DEVELOPMENT OF AN ORAL HEALTH-RELATED QUALITY OF LIFE MEASURE

FOR THE MALAYSIAN ADULTS POPULATION: CROSS-CULTURAL

ADAPTATION OF THE ORAL HEALTH IMPACT PROFILE

By

Roslan Saub

A thesis submitted in conformity with the requirements for

the degree of Doctor of Philosophy

Graduate Department of Dentistry

University of Toronto

© Copyright by Roslan Saub 2004

ivi National Library of Canada du Canada

Acquisitions and Bibliographic Services

395 Wellington Street Ottawa ON K1A ON4 Canada Canada

The author has granted a non- exclusive licence allowing the

National Library of Canada to reproduce, loan, distribute or sell

copies of this thesis in microform,

paper or electronic formats.

The author retains ownership of the

copyright in this thesis. Neither the thesis nor substantial extracts from it

may be printed or otherwise reproduced without the author's

permission.

Bibliotheque nationale

Acquisisitons et services bibliographiques

395, rue Wellington Ottawa ON K1A ON4

Your file Votre référence

ISBN: 0-612-91729-0

Our file Notre référence

ISBN: 0-612-91729-0

L'auteur a accordé une licence non exclusive permettant a la

Bibliotheque nationale du Canada de reproduire, préter, distribuer ou

vendre des copies de cette these sous

la forme de microfiche/film, de

reproduction sur papier ou sur format

électronique.

L'auteur conserve la propriete du droit d'auteur qui protege cette thése. Ni la these ni des extraits substantiels

de celle-ci ne doivent étre imprimés

ou aturement reproduits sans son autorisation.

In compliance with the Canadian Privacy Act some supporting

forms may have been removed from this dissertation.

While these forms may be included in the document page count,

their removal does not represent any loss of content from the

dissertation.

i+i

Canada

Conformément a la loi canadienne sur la protection de la vie privée,

quelques formulaires secondaires ont été enlevés de ce manuscrit.

Bien que ces formulaires aient inclus dans la pagination,

il n'y aura aucun contenu manquant.

ABSTRACT

Development of an oral health-related quality of life measure for the Malaysian adults

population: cross-cultural adaptation of the Oral Health Impact Profile

Roslan Saub Doctor of Philosophy

Graduate Department of Dentistry

University of Toronto 2004

The main aim of this project was to develop an OHRQoL measure for the Malaysian

adult population aged 18 and above by adapting the Oral Health Impact Profile (OHIP). It also

aimed to compare two methods (frequency and regression) of developing a short form of the

measure, to provide preliminary data on the impact of oral conditions on the quality of life of the

adult population in Malaysia, and to compare these impacts between Malaysian and Canadian

population aged 50 and above.

The adaptation of this measure was based on the framework proposed by Herdman et al

(1998). The OHIP was translated into the Malay language using a forward-backward translation

technique. Thirty-six patients were interviewed to assess the conceptual equivalence and

relevancy of each item. The adapted OHIP — L-OHIP(M) — contains 45 items. This

questionnaire was pre-tested on 20 patients. A total of 171 respondents completed the short form

[S-OHIP(M)] to assess the response format. The five-point frequency Likert scale could be used

for Malaysian population.

Field-testing was conducted in order to assess the suitability of two modes of

administration (mail and interview) and to establish the psychometric properties of both the L-

OHIP(M) and the S-OHIP(M).

Both versions [L-OHIP(M) and S-OHIP(M)] were found to be valid and reliable.

However, this study only provides initial evidence for the reliability and validity of these

measures. Further study is recommended to collect more evidence to support the validity and

reliability of these measures.

Both short forms (frequency and regression) performed equally well when tested for

discriminative validity. However, the responsiveness of these measures was not tested in this

i

present study. This could be one of the future research areas to determine which version will be

superior in terms of detecting change.

The preliminary results revealed that a substantial proportion of the sample included in

this study experienced frequent psychological impacts associated with their oral conditions. The

Malaysian population aged 50 and above had slightly higher impacts than Canadian population.

However, the pattern of impacts was similar, where physical consequences were most affected

and social aspects were the least affected.

ill

ACKNOWLEDGEMENT

First and foremost, I would like to thank God the almighty for giving me the strength to pursue this

study.

I would like to extend my gratitude to the following people:

@ Dr. David Locker, my supervisor, for his guidance and advice through out my study.

e Dr. Paul Allison and Dr. Milada Disman, members of my advisory committee, for their criticism

and comments.

e The University Malaya, for sponsoring my study.

T also would like to thanks:

® Dr. Gary Slade for his permission and assistance to adapt his measure.

e The Oral Health Division, Ministry of Health for allowing me to use their database.

® Mr. Rahim Mat Yassim, Ms. Norlie Zolkapli, Ms. Roshidah Hassan, Ms. Sheena Kaur, Ms.

Cecilia Joseph and Ms. Azlina Zainal for their help in translating the questionnaire.

e Mr. Monaj Kumar, Mr. Mahadzar Dayarobi and Mr. Huzaiman Jamil for their hard work locating

and interviewing the respondents.

e The staff of the Department of Community, University Malaya - especially Ms. Pauline Yeo - for

their help conducting the research.

e The respondents who participated in this study.

e Gillete (M) Sdn. Bhd. and Southern Lion Sdn. Bhd. for sponsoring the oral health products.

e@ Ms. Susan Deshmukh for her help in editing this thesis.

Most importantly, my deep gratitude goes to my loving wife, Dr. Norashikin Abdul Fuad, who has

sacrificed herself for my career, providing emotional support, and has lavished me with love and care

thoughout my study. You are an angel to my heart. To my adorable sons, Harith Iskandar and Haziq

Dzulkarnain, I appreciate your understanding and patience while I struggled to complete my thesis. Last,

but not least, to my beloved parents and parents-in-law and my entire family for their support and

encouragement. To all of them I owe a debt that may never be repaid.

This study was supported by the Vot F: Grant no: F0369/2001c, University Malaya, and partial

fellowship award, University of Toronto.

iv

CONTENTS

Page

ABSTRACT ii-iii

ACKNOWLEDGEMENT iv

TABLE OF CONTENT V-ix

LIST OF TABLES X-xil

LIST OF FIGURES xill

CHAPTER 1 - INTRODUCTION l

1.1) Thesis Topic........... ccc ccecece eee ce eee e cee ee eee eee ene ee eee eneetesereeeeenes 1

1.2) Malaysia at Glance...........ccccc cece cee ee eect ee eect eee ee eee e eens enee ena aes 4

1.3) Malaysian Oral Health Care..............cccccc cece eee e eee e eee eaee neces eee 7

CHAPTER 2 - LITERATURE REVIEW 10

2.1) Oral Health Related Quality of Life (OHRQoL) Measures................ 10

2.1.1) The application of HRQoL measures. .......... 0... cece cece eee eee 10

2.1.2) Conceptual bases of health measurement....................0e eee es 12

2.1.3) The properties of the OHRQoL measures.................ce eee ees 17

2.1.3.1) Reliability........... occ cece cece cece eee ee eee eee e ne eee eee 18

2.1.3.2) Validity... cecccece sete eeeeee ee ne essere eee ee eee eas 21

2.1.3.3) RESPONSIVENESS............ cee ee eee ee eee e cence sens ensensenees 24

2.1.3.4) Property for different purpose of the measurements..... 24

2.1.4) Review of existing OHRQoL measures................c cece cece eee 26

2.1.4.1) General (Geriatric) Oral Health Assessment (GOHAI). 30

2.1.4.2) Oral Health Impact Profile (OHIP).................0.0000 39

2.1.4.3) Subjective Oral Health Status Indicators (SOHS))...... 55

2.1.4.4) Oral Impacts on Daily Performances (OIDP)............ 61

2.1.5) Measure chosen to be adapted. ....... 0.0... ccceee ccc e nce ee eee eee ee eens 68

2.2) Cross-Cultural Adaptation of OHRQOL........... eee cece cece cece eee en ees 69

2.2.1) Culture and health... 0... cece cece eee eee renee en eeeereeenanes 69

2.2.2) Approach in HRQoL cross-cultural research...........2......00800

2.2.3) The concept of equivalence. ...............cceccseeenceneeeeeeneeane ees

2.2.4) Technical process in cross-cultural adaptation...................088

2.2.4.1) Translation. 2.0.0.0... cece cece eee e ne ee eee eee ee eeeeenenenan

2.2.4.2) Cultural adaptation. ...............cccececeec een eee snes eee

CHAPTER 3 - AIM AND OBJECTIVE

CHAPTER 4 - METHODOLOGY

4.1) Phase 1: Instrument development and adaptation.......................2e8

4.1.1) Stage 1: Translation and back-translation.....................e eee

Translators Process of translation

4.1.2) Stage 2: Qualitative interviewS..............cceeceee eee eee eeeeeneeens

Respondents

Procedure

Analysis

Committee review

Questionnaire formatting

4.2) Phase 2: Evaluation of the instrument..................ccccccenee eee eeceeeees

4.2.1) Stage 1: Pre-testing. ......... 0... c cece cae e ce ene eee eeeeen eee eeeeneeene Subject

Procedure

Analysis

4.2.2) Stage 2: Testing of response OptionS................cceeeeeeneeneeees Subject Procedure

Analysis

4.2.3) Stage 3: Field Testing. 0.0.0.0... ccc eeceecee eee ee eens eee vecenennes

Study design

Subject Sample size

Procedure

Data entry

Data cleaning Missing data

Scoring

Analysis

4.3 Privacy and confidentiality.......... 0.0... cc cece ee eee eee ene eee enneteeeeeners

4.4 Scientific and ethical approval............ cece cee eee eee reneee eee eeneees

Vi

70

71

78

78

81

36

88

88

88

91

96

96

97

98

CHAPTER 5 - RESULT

5.1)

5.2)

5.3)

5.4)

5.5)

5.6)

5.7)

5.8)

Translation SOO eRe OO REE H RHEE DEH H ERE DE eH OE REED REDE HE DRA HEHE OH ERE E OR EDT ORME HEE TES

Qualitative interview Ce)

Translated and adapted questionnaire ee ee ee ee re ey

Pre-Testing

Test of Response Options i ee ee ee ery

Oe ee ee ee ee

5.6.1) Total Response Rate

Field Testing

ee ee ey

ace mentees enes 5.6.2) Characteristics of respondents and non-respondents

5.6.3) Quality of data

5.6.4) Comparison of mode of Administration-Mail verses Interview...

Cr ee ere

5.6.4.1) Response rate... 0... cee ee ee ee ee ec eee eeeceneneneeenenes

5.6.4.2) Percentage of incomplete data

5.6.4.3) OHIP score

5.6.5) Reliability

eee eee eee ee ee ere ry

OO e oD awe Dae eRe Boda ee HSE reo RHE OEE EHH ree HE HOES EEE OED

5.6.5.1) Internal consistency.............ccecceceeeeceeeeeeeaeeneeees

5.6.5.2) Test-retest

5.6.6) Validity

5.6.6.1) Convergent construct validity............c:ccceeeeeeeeeeeees

i ee ee ee ee i ee ras

i ee ee

5.6.6.2) Discriminative construct validity...................ccee eee

5.6.7) Comparison of the technical properties of the L-OHIP(M) with

the Australian OHIP(A) and the Canadian OHIP(C) data.........

5.6.8) Establishing functional equivalence...............cccceceeens ere ee eee

Comparison of the regression S-OHIP(M) with the frequency S-

OHIP(M). .... 0... ccc cece ec ec nce e ere enenenseeneeeseeeeeeeeeestseaeneeneneeres

5.7.1) Content of the two S-OHIP(M)............0.ccceceeceaeneeecneeeneees

5.7.2) Score for both types of the S-OHIP(M) forms.....................

5.7.3) Technical properties for the two S-OHIP(M) forms...............

The impact of oral conditions on the quality of life of a Malaysian

adult population.......... ccc ccc sce e cen ceceene eee eeee eee eeneeteeseeeeeeeeee nes

vil

11i

11]

116

122

125

127

129

129

130

131

133

134

134

135

136

136

137

140

140

142

145

147

149

149

150

151

133

5.9) Comparison of the oral health-related quality of life between the

Malaysian population and the Canadian population aged 50 and

CHAPTER 6 —- DISCUSSION AND CONCLUSION

6.1) Study limitations............ ccc ccccc ccc cc ccc e cence ee ne eas ee eee eee seen enaenaeenes

6.2) Translation... 0.0... cece cece cece nce e eee ee cece neta en eee ene ee eens eee neenseeee eens

6.3) Cultural adaptation................ cece cece ee ee eee ec ene eeeeeenaeeneeeeeeenas

6.4) Reliability and validity of the S-OHIP(M)................ccceceeeeneeeeeeees

6.5) Comparison frequency S-OHIP(M) with regression S-OHIP(M).........

6.6) The impact of oral health to quality of life of an adult population........

6.7) | Variation in the oral health impact between Malaysian and Canadian

POPULATIONS. ..... 2... cece eee e cence eee eee e teen seen e eee ea eee sesaeeee eee eneeenaenee

6.8) Conclusion....... 0... cece ccece cece e cee ne eee eee cent cette eeeeencenseeenseeeneeeas

6.9) Future research. ......... 0... ccc ence ccc ec eee e eee ene nee reeeeeeeneeneeaeenaeenenes

REFERENCES

APPENDIX

Appendix A Oral Health Impact Profile questionnaire — Original

Appendix B Qualitative interview statement

Appendix C Consent form — qualitative interview

Appendix D Guidelines - qualitative interview

Appendix E Malaysian Oral Health Impact Profile questionnaire —- Long

form - L-OHIP(M)

Appendix F Malaysian Oral Health Impact Profile questionnaire —- Short

form - S-OHIP(M)

Appendix G Introductory letter - Mail respondents

Appendix H Reminder postcard — Mail respondents

Appendix 1 Second reminder letter — Mail respondents

Appendix J Identification letters — The interviewers

Appendix K Appointment card — Interview respondents

Appendix L_ Letter to the police

Vill

158

162

163

164

166

174

175

177

180

181

182

184

195

195

196

197

198

199

200

201

202

203

204

205

206

Appendix M Introductory letter from the dean — Interview respondents 207

Appendix N Impact of oral disorders on daily lives according to the 208

domains

1X

LIST OF TABLES

Table 1.1

Table 2.1

Table 2.2

Table 2.3

Table 2.4

Table 2.5

Table 2.6

Table 2.7

Table 2.8

Table 2.9

Table 2.10

Table 2.11

Table 2.12

Table 2.13

Table 2.14

Table 2.15

Table 2.16

Table 2.17

Table 2.18

Table 2.19

Table 4.1

Table 4.2

Table 4.3

Table 4.4

Table 4.5

Table 5.1

Table 5.2

Table 5.3

Table 5.4

Oral healthcare strategies

Definition of key concepts of the Locker’s oral health model ~

Properties for different type of measures

Summary of an existing OHRQoL instruments

Standard method for reviewing OHRQoL measures

Summary of papers published using GOHAI

Translation and adaptation of GOHAI to other cultures and languages

Summary of papers published using OHIP

Translation and adaptation of OHIP to other cultures and languages

OHIP short form

Scoring method for SOHSI

Summary of papers published using SOHSI

Example of the questions in OIDP

Criteria for frequency score

Summary of papers published using OIDP

Translation and adaptation of OIDP to other cultures and languages

Different concepts of equivalence

Definition of equivalence (Herdman et al, 1998)

Method of translations and its advantages and disadvantages

Method of assessing equivalence

Phase and stage of the study

Example of combined file

Codes for Oral Health Impact Profile domains

Questions asked to assess the questionnaire

Assigned groups and sample size required

Level of translation difficulty

Difficult to translate questions

Sample distribution for qualitative interview.

Impacts of oral disease and oral disorder

Page

13

25

27-28

29

33-36

37-38

43-5]

52-53

54

56

58-60

63

63

65-66

67

73

74

81

82

88

90

94

97

100

112

112

117

118

Table 5.5

Table 5.6

Table 5.7

Table 5.8

Table 5.9

Table 5.10

Table 5.11

Table 5.12

Table 5.13

Table 5.14

Table 5.15

Table 5.16

Table 5.17

Table 5.18

Table 5.19

Table 5.20

Table 5.21

Table 5.22

Table 5.23

Table 5.24

Table 5.25

Table 5.26

Table 5.27

Table 5.28

Relevancy of items

Problem questions

Deleted, reworded and new items

Format of questionnaire

Characteristics of the respondents for pretesting

Summary of pretesting finding

Characteristics of the respondents for test of response options study

Response rate for field testing stage.

Characteristics of respondents and non-respondents

Percentage of missing items for the S-OHIP(M)

Percentage of missing items for the L-OHIP(M)

Characteristics of the eligible sample for mail and interview group

Response rate of mail and interview for the L-OHIP(M) and the S-

OHIP(M)

Percentage of the incomplete data.

Mean ADD-Score and SC-Score

Internal consistency for the L-OHIP(M) and the S-OHIP(M) by mode

of adminstration

Internal consistency for the L-OHIP(M) and the S-OHIP(M) by age

group

Number of respondents who completed the questionnaire for the

second time

Respondents’ characteristics — Test- Retest.

Intraclass correlation coefficient by mode of administration

Intraclass correlation coefficient by age group

Mean ADD-score and Sperman’s rank correlation coefficients for the

L-OHIP(M)

Mean ADD-score and Sperman’s rank correlation coefficients for the

S-OHIP(M)

L-OHIP(M) mean score by perceived dental treatment need and

satisfaction with oral health

Xi

119

121

123

124

125

126

127

129

130

131

132

133

134

135

135

136

137

138

138

139

140

141

141

142

Table 5.29

Table 5.30

Table 5.31

Table 5.32

Tabie 5.33

Table 5.34

Table 5.35

Table 5.36

Table 5.37

Table 5.38

Table 5.39

Table 5.40

Table 5.41

S-OHIP(M) mean score by perceived dental treatment need and

satisfaction with oral health

Mean score of the L-OHIP(M) and the S-OHIP(M) by dental status

Comparison of the reliability of the L-OHIP(M) with the OHIP(A)

and the OHIP(C)

Comparison of the validity of the OHIPCM) with the OHIP(C)

Item content and percent reported “very often’, “often” or

“sometimes”

Mean, median and range of ADD-score and SC score

Comparison of the technical properties of the two forms of the S-

OHIP(M)

Percent responding “very often”, or “often” to one or more items in

each subcales

Response to OHIP(M)

Percent responding “very often” or “often” to one or more items in

each subcales by sociodemograhics and dental status

Background of the Malaysian and Canadian data

Comparison of the percent responding “very often” or “often” to one

or more items in each subscale between the Malaysian sample and the

Canadian sample by gender

Comparison of the percent responding “very often” or “often” to one

or more items in each subscale between the Malaysian sample and the

Canadian sample by dental status

xii

143

144

146

147

150

151

152

153

154

157

158

160

161

LIST OF FIGURES

Figure 1.1

Figure 2.1

Figure 2.2

Figure 2.3

Figure 2.4

Figure 2.5

Figure 2.6

Figure 2.7

Figure 5.1

Figure 5.2

Malaysia map

Potential application of the “patient based outcome measures”

The conceptual model (Locker, 1988)

Simplified model of oral disorders, oral health and quality of life

Measuring oral health and quality of life

Interpretation of a Kappa value by different authors

Conceptual model used for the development of OIDP

Type of cross-cultural research

Distribution of response options

Comparison of the percent responding “very often” or “often” to one or

more items in each subscale between the Malaysian sample and the

Canadian sample

Xiil

Page

1]

13

14

15

19

62

72

128

159

CHAPTER 1

INTRODUCTION

1.1) THESIS TOPIC

Traditionally oral health is measured based on the biomedical model, which focuses on the

presence or absence of disease. However, such measures fail to take into account the burden of

illnesses and disabilities imposed by oral diseases (Gift and Redford, 1992; Locker, 1988;

Cohen and Jago, 1976; Morigama, 1968). In line with the World Health Organization’s (1948)

definition of health, which is defined as “‘a complete state of physical, mental and social well

being and not just the absence of illness”, measuring health should also incorporate

functioning, well being and quality of life. Using this, together with disease measurement, a

comprehensive picture of oral disorders can be captured.

There are a growing number of investigators who are developing ways of measuring the

impact of oral diseases on the well being of communities and individuals. As a result, several

instruments have been developed to evaluate oral health related quality of life. Most of the

instruments have been developed in English speaking countries and in the English language.

Recently, some of the instruments have been translated into other languages.

Health and quality of life are subjective phenomena. The perception of health and the

ways in which health problems are conceptualized and expressed vary from culture to culture

(Guillemin et al, 1993). They are shaped by and embedded in culture so that their measurement

cannot be culturally free (Guarnaccia, 1996). Since most of the instruments have been

developed in the English language and are intended for use in English speaking countries,

particularly in the United States of America, United Kingdom, Australia and Canada, they

cannot be used in other countries, such as Malaysia, with a different language and cultural

background without some form of adaptation.

In general, there are two ways to develop a health-related quality of life instrument for a

different culture (Guillemin et al, 1993): to develop a totally new measure or modify a previous

measure through a cross-cultural adaptation process. The first option is a time consuming

process in which the bulk of the effort is devoted to the conceptualization of the measure and

the selection and reduction of items. In the second option, a transposition of a measure from its

original context to the target population through translation and adaptation is undertaken. This

option is rather cheaper and less time-consuming as compared to the first. Cross-cultural

adaptation not only allows us to describe or evaluate the health status of the target population,

but also allows us to make comparisons between cultures or nations and the differences and

similarities resulting from the various health care systems, the differing attitudes of health

professionals and patients’ health care behaviors (Touw-Otten and Meadows, 1996).

Malaysia is a developing country. As in developed countries, oral health care costs are

increasing each year. In the National Household Health Expenditure Survey 1996 (NHHES’96)

it was reported that the total household expenditure for dental care in Peninsular Malaysia was

four percent of the total household expenditure on health (NHHES-team, 1999). About 96% of

the estimated household expenditures for dental care were on treatment costs. The question is

does this high spending on dental treatment lead to substantial oral health gain?

The focus of the Ministry of Health of Malaysia has shifted from being disease oriented

to emphasizing wellness and the maintenance of the quality of life of the population (Jui-meng,

1998). Recognizing that oral health is part of general health, and that it has been shown to

contribute to the quality of life (Locker et al, 2000), more attention should be given to oral

health. However, data on the impact of oral diseases on the well-being and quality of life of the

Malaysian population is not available. Therefore, it is essential to develop an instrument that

enables us to measure this impact and provide such information. This measure can then be used

to justify resources for an oral health program, to reallocate resources or to evaluate dental care

programs.

Since it is advocated to adapt an existing valid instrument, a measure of the oral health-

related quality of life for the Malaysian population will be developed by adapting a measure

that has been developed elsewhere. There are two important issues that should be taken into

consideration in the process of adapting any health related quality of life instrument (Behling

and Law, 2000). First, it must be valid and reliable and second it must meet requirements for

equivalence relative to the source language measure. To ensure these requirements are met, the

model of equivalence proposed by Herdman et al (1998) was used in this study.

The Oral Health Impact Profile (OHIP) (Slade and Spencer, 1994) was chosen to be

adapted since it meets the aim of this project, to develop a generic oral health related quality of

life measure, which is able to describe oral health and also to discriminate across groups in the

population. This measure was based on a coherent theoretical framework and its validity and

reliability have been tested in different countries (Wong et al, 2002a; John et al, 2002; Allison

et al, 1999; Hunt et al, 1995; Locker and Slade, 1993). Moreover, the OHIP is the most

sophisticated and comprehensive of the measures currently available, covering a wide range of

quality of life dimensions.

1.2) MALAYSIA AT GLANCE

Malaysia is located in South East Asia covering 0.3 million square kilometers with a tropical

climate. It comprises 13 states and 3 Federal Territories (Kuala Lumpur, Labuan and Putrajaya).

Sabah, the Federal Territories of Labuan and Sarawak are located on Borneo Island, known as

East Malaysia, while all the other states are in the Peninsular, separated by the South China Sea

(Figure 1.1). Kuala Lumpur is the capital city of Malaysia with a population of approximately 3

million (Department of Statistics Malaysia, 2001)

Tee

Sergiy & soe

HTS we

KUALA *Kuertet > - gi tNPUR

Figure 1.1. Malaysia map Source: hitp://www.ca.gov/cia/publications/factbook/geos/my. html.

The total population of Malaysia is 23.27 million (Department of Statistics Malaysia,

2001). The average annual population growth rate was 2.6% over the 1991-2000 period.

Malaysia has a rather young population, where the proportion of people age 65 and above was

only 3.9%. The median age of the Malaysian population was 23.6 years. The ratio of males to

females was 104:100. This shows that men outnumber women.

Over the years Malaysia has grown from an agro-based economy to a manufacturing

economy, where the manufacturing sector attained the highest growth rate and is expected to

grow by 5.0 percent in 2003 (Central Bank of Malaysia, 2003). Malaysia’s real Gross Domestic

Product (GDP) growth over the past years has been on average eight percent. However, with the

global economy crisis, the growth rate was 4.2 percent in 2002 and it is projected to grow by 4.5

percent in 2003 after taking into account the current situation (Bumiputra Commerce, 2003).

Meanwhile, the inflation rate, as measured by the consumer price index (CPI), is expected to

remain subdued, rising by 1.5 percent in 2003 (Bumiputra Commerce, 2003). The

unemployment rate is lower than four percent.

Malaysia is a multiethnic, multicultural and multi-religious nation, where Malays,

Chinese, Indian and indigenous peoples live in caring cultural harmony. Malays being the

majority ethnic group, constitute slightly more than 50% of the total Malaysian citizen

population (Department of Statistics Malaysia, 2001). However, the composition of ethnicity

between Peninsular Malaysia and East Malaysia are different. In Peninsular Malaysia, the main

ethnic groups are the Malay, Chinese and Indian, while in East Malaysia, the Indigenous of

Sabah and Sarawak are the main ethnic groups respectively.

Malaysians have strong ties with their family and relatives. In general, family life is

valued more highly than individual accomplishment (Tewolde, 1999). Seniority is greatly

respected within Malaysian households. Politeness and courtesy are very important social values

among Malaysians. Malaysia has been identified as a high power/distance society (MacLachlan,

1997), where less powerful members respect the more powerful members of the institutions or

organizations. For example, students must respect their teachers in every way and the students

are expected to follow their teachers’ instructions.

Each ethnic group practices its own religion and ritual. Islam is the most widely

practiced in Malaysia, with slightly more than sixty percent of the population being Islamic

(Department of Statistics Malaysia, 2001). Although Islam is the most widely practiced, other

religions, such as Buddhism (19.2%), Christianity (9.1), Hinduism (6.3%) and

Confucianism/Taoism/other traditional Chinese religions (2.6%) are also practiced freely in

Malaysia. Religion is closely related with ethnicity, where all Malays are Muslims, most Indian

are Hindus and most Chinese are Buddhist.

Although there are many languages spoken in Malaysia, the Malay language has been

deemed to be the National language under Article 152 of the constitution. However, this

constitution does not prohibit or prevent people from using, from teaching or learning in other

languages. Nevertheless, the Malay language must be used for official purposes, which includes

its use by federal and state governments. Other main languages widely spoken in Malaysia are

English, Chinese and Tamil.

The literacy rate among the Malaysian population has improved over the years. In 2000,

the literacy rate among Malaysians aged 10 to 64 years was 93.5% (Department of Statistics

Malaysia, 2002). Education starts at the age of four or five years old under the preschool

education. Children enter primary school at six years. Malaysia provides eleven years of free

public schooling, with six years in primary school, three years in lower secondary school and

two years in upper secondary school (Ministry of Education, 2003). The education system

places a strong emphasis on literacy skills, mathematics and basic sciences. About one in six of

Malaysians aged 20 and above attained higher education (post secondary, college or university).

1.3) MALAYSIAN ORAL HEALTH CARE

The government of Malaysia realizes that in nation building, it is important to have a healthy

population. Hence, the government has placed great importance on the development of the

health care sector and this had led to the formation of the Ministry of Health (MOH). This

ministry is set up to formulate, administer and manage health care policy and programs in

Malaysia. Recently, the ministry has set a strong vision for health, which stated (Oral Health

Division, Ministry of Health Malaysia, 2001a):

By the year 2020, Malaysia is to be a nation of healthy individuals,

families and communities, through a health system that is equitable,

affordable, efficient, technologically-appropriate, | environmentally- adaptable and consumer-friendly, with emphasis on quality, innovation, health promotion and respect for human dignity and which promotes

individual responsibility and community participation towards an enhanced quality of life.

The ministry has pursued various programs in ensuring that the vision of having a

healthy nation will be achieved. One such program is oral health care. Realizing that oral health

has a significant contribution to the quality of life of the population, the Oral Health Division of

the Ministry of Health was formed, as a main agency for the dental profession and also for the

provision of oral health care to the nation (Oral Health Division, Ministry of Health Malaysia,

2001). Besides the MOH, the Ministry of Defence, the Department of Aboriginal Affair (within

the Ministry of National Unity and Community Development), and the Ministry of Education

(through their dental faculty) also make substantial contributions to the provision of oral health

care to the population. In addition, the private sector has also made a significant contribution to

oral health care, mainly to the urban population.

Currently, oral health care in the public sector is largely subsidized by the government

(Oral Health Division, Ministry of Health Malaysia, 2001a). Pre-school children, school

children up to age 17 years, ante-natal mothers and civil servants, their spouses and school-

going dependents below the age of 21 years, are entitled to free basic oral health care at public

health care facilities. Other groups who are entitled to free care include; those who are

physically, mentally or economically disadvantaged. Nevertheless, all members of the public,

regardless of income may also access public sector facilities at highly subsidized rates, while in

the private sector, oral health care is largely on a fee-for-service basis (out of pocket). To date,

there are few third party payment schemes and there is no national insurance scheme in place.

In Malaysia, oral healthcare is provided mainly by qualified dentists. In the year 2000,

there were 2,144 dentists actively practicing in this country; 65 percent were in private practice

with the majority concentrated in urban areas (Oral Health Division, Ministry of Health

Malaysia, 2001a). Unlike the private dental care facilities, public dental facilities are widely

distributed through out Malaysia in order to ensure that each Malaysian has access to oral health

care. However, because dentists prefer to practice in the private sector, causing a shortage of

dentist in the public sector, this has led to a major equity issue in terms of access to oral health

services in rural areas. Recently, the government has enforced a compulsory three years of

service in the public sector for all newly graduated dentists in order to cope with the shortage of

dentists in the public service.

The Oral Health Division of the MOH, as a leading organization in oral healthcare, has

planned several strategies to ensure that the goal of improving or maintaining the oral health of

the population is achieved. The focus of these strategies is on prevention and oral health

promotion. The strategies are listed in Table 1.1.

Table 1.1. Oral healthcare strategies

1. Increasing oral health awareness of the community through oral health promotion and education.

Fluoridating public water supplies at an optimum level of 0.7ppm. Providing clinical preventive oral healthcare services to all school children in need. improving inter-agency and inter-sectoral collaboration and co-operation.

Providing quality oral healthcare services, which are easily accessed, suitably utilized and technologically appropriate.

Providing maximum coverage to identified priority groups. Rendering the maximum number of school children orally-fit. Providing specialist oral healthcare services to those in need of these services Collecting and analysing data, as well as undertaking research aimed at improving the quality of the oral healthcare services provided.

a Pwh

OOND

Source: Oral Health Division, Ministry of Health Malaysia. Oral healthcare in Malaysia. Malaysia: Oral Health Division,

Ministry of Health Malaysia. 2001. page 20

The constraints on public health resources has led to the setting of priorities by

identifying target groups. The target groups include school children, preschool children,

antenatal mothers and disadvantaged groups (Ramli, 2001). Dental nurses deliver oral

healthcare to schoolchildren below the age of 17 years, mainly under the school oral health

program. The dental nurse is the only operating dental auxiliary available in Malaysia,

introduced in 1948 based on the New Zealand dental nurse model (Oral Health Division,

Ministry of Health Malaysia, 2001a). However, they can only operate in government facilities

under the supervision of a public dentist.

Specialist care is also provided by the public dental service. Currently there is no

specialist register in Malaysia. However, the MOH recognizes five dental specialists; oral

surgery, orthodontics, periodontology, paediatric dentistry and oral medicine/oral pathology

(Oral Health Division, Ministry of Health Malaysia, 2001a).

CHAPTER 2

LITERATURE REVIEW

2.1) ORAL HEALTH RELATED QUALITY OF LIFE (OHRQoL) MEASURES

Oral health status and oral health related quality of life measures, which are collectively known

as “patient based health outcome measures”, were developed to assess the subjective

experiences of people with oral disease or disorders. Because health and quality of life are

subjective phenomena, the development of these measures must be based on a specific

conceptual approach, and in order to be validly used in research, the instruments must possess

adequate psychometric properties. Thus, this section reviews current concepts in health

measurement and the properties that are essential for such measurement. It also reviews some

of the existing OHRQoL measures.

2.1.1) The application of HRQoL measures

Patient based outcome measures have several applications. Locker (1996) identified these

potential applications and classified them into three broad areas (Figure 2.1): 1) political

applications — advocating for resources for dental research and services, 2) theoretical

applications — such as in exploring models of health and illness, and 3) practical applications —

such as in estimating the need for health care, assessing the quality of services, evaluating the

effectiveness of interventions and in cost utility analysis. For example, Srisilapanan and

Sheiham (2001b) integrated a patient based outcome measure into the assessment of treatment

needs in a dentate older population in Thailand. This approach to estimating need is known as

the sociodental approach. They concluded that this approach may better represent those dentate

older subjects’ who will gain more health benefit from a partial denture. OHRQoL measures

10

11

can also be used to assess the effectiveness of oral health interventions, as for example, the

study done by Awad et al (2000), which compared two types of treatment (implant versus

conventional denture) in replacing missing teeth.

Application

¥ ‘ 7 Political Practical |

Public Health |

‘

Need assessment Program evaluation

y

Cost utility analysis

v Theoretical |

Clinical

Figure 2.1. Potential applications of the “patient based outcome measures”

Another potential application of HRQoL measures is in clinical practice. According to

Higginson and Carr (2001), there are eight potential areas in clinical practice where a measure

can be used - to prioritize problems, facilitate communication, screen for potential problems,

identify preferences, monitor changes or response to treatment, train new staff, clinical audit

and clinical governance. However, for a measure to be clinically useful ‘it must not only be

valid, appropriate, reliable, responsive, and able to be interpreted, but it must also be simple,

quick to complete, easy to score and provide useful clinical data.’ (Higginson and Carr, 2001)

12

2.1.2) Conceptual bases of health measurement

It is well accepted now that the measurement of disease alone is not sufficient to describe the

oral health status of the individual or population or when comparing the efficacy and

effectiveness of interventions, since the thinking about health and oral health has changed over

the years. Health is no longer seen as the absence of disease but rather in terms of obtaining or

maintaining optimal functioning and social and psychological well-being (Locker, 1997a).

Such thinking has led to several implications. Firstly, it moves from a concern with disease to a

concern with health. Secondly, it has moved from an emphasis on curing disease to an

emphasis on prevention and health promotion. Thirdly, it has changed the emphasis from health

services to an emphasis on the physical and social environments in which people live and

finally, to see a patient as a person rather than as a body part. This change has also led to a

change in how oral health should be measured.

Because health and quality of life are subjective phenomena, their measurement must be

based on a coherent conceptual model. One such model that has been widely utilized in the

development of oral health measures, is the one proposed by Locker (1988). This model was

derived from the World Health Organization’s disease and its consequences model, which

consists of the following concepts: impairment, functional limitations, pain and discomfort,

disability, and handicap. The definitions of these concepts are shown in Table 2.1. This model

linked the concepts in a linear fashion, which moves from biological to behavioral and

psychosocial concerns (Figure 2.2). However, this model does not link to the quality of life.

Table 2.1. Definition of key concepts of the Locker’s oral health model

impairment:

Functional

limitation:

Pain and

discomfort:

Disability:

Handicap:

Anatomical loss, structural abnormality or disturbance in biochemical or physiological processes, which arises as a result of disease or injury or is present at birth.

Restrictions in the functions customarily expected of the body or its component organs or systems.

Self-reported pain and discomfort, physical and psychological symptoms and other not directly observable feeling states or manifestations which impinge on the individual or others.

Any limitation in or lack of ability to perform the activities of daily living.

The disadvantage and deprivation experienced by people with

impairments, functional limitation, pain and discomfort or disabilities because they cannot or do not conform to the expectations of the groups to which they belong.

a Impairment

Functional limitation Pain

y Disability

Se ‘s Handicap

Figure 2.2. The conceptual model (Locker, 1988) Adapted: Slade GD, Spencer AJ. Development and evaluation of the Oral Health Impact Profile. Community Dental Health 1994; [1:3-1]

13

14

More recently, the term “health related quality of life” is being widely used in the

literature. The measures that were developed to assess the extent to which oral diseases and

disorders affect functioning and psychosocial well-being are now referred to as measures of

“oral health related quality of life” (Slade, 1997a). Whether these measures assess oral health

related quality of life or oral health status remains equivocal.

It is recognized that health status and health-related quality of life represent two distinct

concepts. Although both measures assess subjective experiences of people related to disease or

disorders, health-related quality of life measures, however, incorporate the values placed on

these experiences. Wilson and Cleary (1995) have provided a conceptual framework that

reflects this thinking. This model links, in a linear fashion, biological variables at one end and

quality of life at the other. Linking these two dimensions of human experience are symptoms,

functional status and health perceptions. Locker et al (2002a) then adapted and simplified this

model for oral-health related quality of life, as shown in Figure 2.3. This model suggests that

oral diseases and disorders may compromise the physical and psychosocial functioning of the

individual and subsequently compromise overall quality of life

Oral diseases /disorders

v Functional and psychosocial impacts

Quality of Life

Figure 2.3. Simplified model of oral disorders, oral health and

quality of life. Source: Locker D, Matear D, Stephens M, Jokovic A. Oral health-related quality of life of a population of medically compromised elderly people. Community Dental Health 2002; 19: 90-97.

15

It is clear that thinking about health has evolved from equating health to the absence of

disease to inclusion of subjective experiences and recently incorporates the extent to which

health compromises the individual’s life. This in turn affects the way in which measures should

be developed.

Figure 2.4 depicts the changes that have occurred in measuring oral health. The

contemporary concept of oral health has led to the development of two types of measurement:

oral health status and oral health-related quality of life. The difference between these two

measures lies in the content of the measure. The items comprising a measure of oral health

status capture those aspects of life that are affected by oral disorders or diseases. Oral health-

related quality of life measures, on the other hand, contain items that capture aspects of life that

patient’s value (Gill and Feinstein, 1994).

Concept Measurement type

we Presence or ence of Clinical indices | Traditional J disease. abs —_> ca

3 Aspects of life affected by oral ——p> Oral health status Contemporary disease/disorders.

Aspecis of life that people —-» Oral health-related quality of life value

Figure 2.4. Measuring oral health and quality of life.

16

Therefore, a measure of oral health-related quality of life must not only capture aspects

of life that have been affected by the oral conditions but also the values attached to these

aspects. This can be done in two ways: the indirect and direct method (Locker et al, 2002a).

The indirect method involves establishing an association between the measure and quality of

life measures such as life satisfaction or morale. There are two types of direct method: 1)

weighting — by giving weight to each item; for example Slade and Spencer developed a severity

weight to the 49 items of the OHIP using a panel of dental patients to judge the importance of

each item in relation to other items in the measure, and 2) using the item impact method when

selecting items for the scale. This method begins with pooling items that affect people’s daily

activities. This can be done in many ways — literature reviews, expert opinion, or patient

interviews. The set of relevant items is then shown to the patient to indicate whether they

experience it, if so how important it is to them (Guyatt et al, 1986; Juniper et al, 1996; Locker

et al, 2002a). This method ensures that each item addresses aspects of life that are important to

people.

The question is what does the OHIP-49 assess? Because items were selected from a

pool of items derived from patient interviews, it could be an indirect indication that those items

are, to a certain extent, important to them. In addition, each item was given a weight of

importance. Thus, based on these aspects, OHIP could be considered as a measure of oral

health-related quality of life. Locker et al (2002a) carried out a study to see whether the aspects

of daily life addressed by the OHIP-14 and the GOHAI were in fact important to the study

population. This was done by correlating the score of the OHIP-14 and GOHAI with a set of

broader measures of psychological well-being and life satisfaction, which can be interpreted as

indicators of life quality. Three measures of life quality were used in their study, i.e, morale,

perceived life stress and overall life satisfaction. Their data provide some evidence that the

17

items in both of these measures were measuring aspects of life that were important to the study

population. In other words, these measures can be regarded as oral health-related quality of life

measures.

2.1.3) The properties of OHRQoL measures

This section provides an overview of the properties of OHRQoL measures, that need to be

evaluated before they can be meaningfully used for clinical practice and research in dentistry.

This section also reviews the psychometric properties necessary for different types of measures

(descriptive, discriminative, evaluative or predictive).

Oral health related quality of life is an abstract phenomenon, which cannot be measured

directly. Thus, it is necessary to translate the concept of OHRQoL to empirical indicators,

which adequately represents each element of the overall concept. As with any other

measurements, OHRQoL measures are not free of error. There are two types of error that affect

a measure: random error and systematic error. Random error refers to the error caused by

chance. As such, the direction and magnitude may vary from time to time (Green and Lewis,

1986). On the other hand, systematic error is the one that systematically affects the

characteristics being measured. It consistently tilts the result in one direction or another. As

such, it produces bias in the measurement.

Therefore, before it can be used to provide a precise and accurate measurement, two

basic properties, reliability and validity, need to be evaluated. In addition, for evaluative

measures (which will be discussed later in this section) an additional property that is required is

responsiveness.

18

2.1.3.1) Reliability

The measurement of any phenomenon always contains a certain amount of random error. Thus,

repeated measurements of the same phenomena never precisely duplicate each other. However,

they do tend to be consistent (Carmines and Zeller, 1979). This tendency towards consistency

found in repeated measurements of the same phenomenon is referred to as reliability. In other

words, reliability is concerned with minimizing random measurement error (Green and Lewis,

1986).

There are four different methods of assessing reliability: test-retest reliability, internal

consistency reliability, inter-rater reliability and equivalent-form reliability. However, in this

review only those types which are relevant and frequently used in assessing OHRQoL

measures, are reviewed, These are test-retest and internal consistency reliability.

Test-Retest Reliability.

Test-retest reliability is one of the most widely used methods for assessing the reliability of a

self-report measure, such as OHRQoL measures (Streiner and Norman, 1995). It is a measure

of how reproducible the measurement is. In this method, the same test is given to the same

individuals separated by a sufficient time interval. The correlation coefficient is then calculated

to compare scores from the two sets of responses.

In test-retest reliability, when the scale is continuous, the intraclass correlation

coefficient (ICC) is the more appropriate test to be used (Fayers and Machin, 2000; McDowell

and Newell, 1996; Streiner and Norman, 1995; Deyo et al, 1991). Unlike the pearson

correlation coefficient, which only measures linear association, the ICC measures the strength

of the association and the strength of agreement between repeated measurements. It assesses

the proportion of total variance arising from between-subject variability, using formula 2.1.

ICC = ~~ o..t~<“‘~smtC © subject + O” error

2 O™ subject

19

Whereby O subject 18 between-subject variance and Oo” enor 18 Within-subject variance. Analysis

of variance (ANOVA) is commonly used to estimate these components. If the ICC is close to 1,

then the random error variability is low and a high proportion of the variance in the

observations is attributable to variation between patients. Thus, the measure is regarded as

having high reliability. Conversely, if the ICC is close to zero, then random error variability

dominates and that the measure has low reliability.

In the case where the scale of the measurement is not continuous, Kappa and weighted

Kappa coefficients can be calculated for test-retest reliability. Figure 2.5 shows the

interpretation of Kappa values by several authors.

1.0 —

0.8 -—-

04—-

0.2 —-

0.0 —

Landis and Koch (1977)

Altman (1991)

1.0-—

Almost perfect Very good

Substantial Good

Moderate Moderate

Fair Fair

Slight

Poor

Poor

Fleiss (1981)

Excellent

Fair to good

Poor

0.75

Byrt (1996)

Excellent Very good

Good

Fair

Slight

Poor

No agreement

0.92

Figure 2.5. Interpretation of a Kappa value by different authors Source: Szklo M, Nieto FJ. Epidemiology: beyond the basics. Maryland: An Aspen Publication, 2000.

page 377.

20

Internal Consistency

Internal consistency is an indication of how well the different items in a measure capture the

same underlying concept. This means that the items should be at least moderately correlated to

both the total scale score and to each other (Streiner and Norman, 1995). This method requires

only a single test administration. Internal consistency reliability is estimated by a statistic

known as Cronbach’s coefficient alpha. It is a statistic that reflects the homogineity of the

scale. The formula for estimating Cronbach’s coefficient alpha is shown in 2.2.

a= NO a ceccecececenteeeeceeeeeeseaeaeeeaeesaeaeaeeeneenes (2.2) 1+ p(N—- 1)

N= Number of items

p = Mean inter-item correlation

The basis of Cronbach’s alpha is that if the items are uncorrelated, the mean of the

inter-item correlation will be zero, implying that Cronbach’s coefficient alpha will be zero. On

the other hand, if the items are identical, the mean inter-item correlation will be equal to 1 and

this indicates that Cronbach’s coefficient alpha will be one. This shows that alpha varies

between 0 and 1. The higher the a value, the more internally consistent the measure is. For

psychometric measurements, a Cronbach’s coefficient alpha above 0.7 is generally regarded as

acceptable (Fayer and Machin, 2000).

Based on formula 2.2, the alpha value is influenced by the average inter-item

correlation and the number of items in the scale. Therefore, adding items to a scale that do not

result in a reduction in the average inter-item correlation will increase the reliability of the

measure. However, at some point, adding more items will yield less impact.

For dichotomous items, the KR-20 coefficient is used to establish internal consistency

reliability (Green and Lewis, 1986).

21

2.1.3.2) Validity

Another property of a measure that needs to be evaluated is validity. Validity is an estimation

of the extent to which an instrument measures what it is intended to measure (Kline, 2000;

Fayers and Machin, 2000; Streiner and Norman, 1995; Hays et.al, 1995; Green and Lewis,

1986). A measure is regarded as valid if there is sufficient evidence to support that the

measurement measures what it claims to measure. Basically, validation is a process of

hypothesis testing (Streiner and Norman, 1995).

In general there are four types of validity: content validity, face validity, criterion

validity and construct validity.

Content Validity

Content validity is concerned with whether or not items in the measure adequately represent all

relevant constructs under investigation (Fayers and Machin, 2000; Striener and Norman, 1995).

Thus, content coverage and content relevance are two important aspects of content validity.

Content validity is seldom tested formally. It depends largely upon ensuring that the

instrument has been developed based on wide coverage of items, which could be generated

through consultation with experts in the area of interest, literature or from interviews with the

target population. At the same time, the relevancy of each item is also assessed. Any irrelevant

items should be excluded. This is commonly determined by an expert panel in the area of

interest, or by asking the target population for their opinion as to the relevancy of the items.

22

Face Validity

Face validity involves checking whether items in the measurement appear “on the face of it” to

cover the intended topics clearly and unambiguously (Fayers and Machin, 2000) and it is

assessed after the measure has been constructed.

Criterion Validity

Criterion validity is assessed by correlating the measurement with some other measures, which

have been accepted as a measurement of the phenomenon, which can be treated as a criterion or

a “gold standard” (Fayers and Machin, 2000; Kline, 2000; McDowell and Newell 1996;

Streiner and Norman, 1995; DeVellis 1991; Carmines and Zeller, 1979). In this respect, if the

new measurement does not correlate significantly with the criterion, then it will not be useful.

The higher the correlation, the more valid is this measurement for this particular criterion.

However, then, a question arises, if there already exists a valid and reliable measure,

why would one develop a new one? Streiner and Norman (1995) provide two valid reasons: 1)

when the existing test is expensive, invasive, dangerous, or time consuming; or 2) the outcome

may not be known until it is too late.

Criterion validity is divided into two types: concurrent validity and predictive validity.

If the criterion exists in the present, for example in the case where the existing instrument is

time consuming or too long, thus a shortened version is needed, then the concurrent validity is

to be assessed, by correlating a measure and the criterion, in this case the long version, at the

same point of time. On the other hand, if one wants to develop an instrument that can be used

to predict the future, whereby the criterion will be only available at a different time (future),

then predictive validity is to be assessed.

23

Construct Validity

Construct validity is evaluated by hypothesizing how measures should ‘behave’ and confirming

or refuting these hypotheses (Hays et al, 1995). Therefore, in this validation process,

hypotheses arising from a theoretical framework need to be tested using different procedures

and statistical techniques, depending on the hypotheses that have been generated. Construct

validity is supported when the results are consistent with the hypotheses. However, this does

not mean that the validity is proven rather it is just an indication of the degree of validity.

Therefore, it needs to be continuously evaluated so that there is more evidence of the validity.

In the case where the results are not consistent with the hypothesis, one could suspect that it

could be due to either the theory being wrong, or the instrument being flawed or both (Striener

and Norman, 1995).

Basically, there are three types of construct validity: discriminative, convergent and

divergent validity. Discriminative validity is related to how well the scale is able to distinguish

between groups with known differences (Herdman et al, 1998). For example, among dentate

and edentate patients. Convergent validity describes how closely a measure is related to other

measures of the same construct and to which it should be related. Conversely, divergent

validity recognizes that some dimensions of QoL are anticipated to be relatively unrelated and

that their correlation should be low.

There are different methods available in establishing construct validity, such as known

groups method, correlational method, factorial analysis, and multitrait-multimethod analysis.

However, it is beyond the scope of this review to provide the detail of these methods.

24

2.1.3.3) Responsiveness

Responsiveness concerns the ability of an instrument to detect change within subjects or

treatment effects (Stretner and Norman, 1995; Guyatt et al, 1992; Wilkin et al, 1992; Kirshner

and Guyatt, 1985). There are four ways to assess change in health status (Locker, 1998): 1)

Before and after comparison, 2) Change score, 3) Global transition judgement and 4) Global

transition scale. However, a detailed description of these methods is beyond this review.

2.1.3.4) Properties for different purposes of the measurements.

In general, HRQoL measures can be divided into four broad applications: descriptive (Locker

and Allen, 2002b), discriminative, predictive and evaluative measures (Guyatt et al, 1992;

Kirshner and Guyatt, 1985). Different purpose of measurement requires different properties.

Table 2.2 summarizes the properties for each of the measures.

25

Table 2.2. Properties for different types of measures

PURPOSE OF MEASURES PROPERTY Reliability Validity Responsiveness

DESCRIPTIVE MEASURE: Large and stable inter- Face Not relevant Descriptive measures are used to patient variation Content

measure current status of the Construct

population. (Cross- sectional)

DISCRIMINATIVE MEASURES: Large and stable inter- Face Not relevant Discriminative measures are used to patient variation Content

distinguish between groups or Construct

individuals on some underlying oral (Cross- health related dimension at one point sectional) of time.

PREDICTVE MEASURES: Large and stable inter- Face Not relevant Predictive measures are commonly patient variation Content used in medicine as a mean to identify Criterion

groups or individuals who will develop (Predictive) some target condition or outcome.

EVALUATIVE MEASURES: Small variation Face Applicable Evaluative measures are used to between replicate Content measure the magnitude of longitudinal measure Construct

change in an individual or group on (Longitudinal)

the oral health related quality of life.

Adapted: Kirshner B, Guyatt G. A methodological framework for assessing health indices. Journal of Chronic

Disease 1985; 38: 27-36.

26

2.1.4) Review of existing OHRQoL measures.

Generally, there are two approaches in developing patient-based health outcome measures:

generic and disease-specific. Generic measures are designed to be broadly applicable across

types and severity of disease, different medical treatments or health interventions (Patrick and

Deyo, 1989). On the other hand, disease-specific measures are those designed to assess specific

diseases (e.g. joint disease), or specific conditions (e.g. back pain) or special population (e.g.

older adults) (Patrick and Deyo, 1989). This section reviews some of the OHRQoL measures.

Since 1976 when Cohen and Jago (1976) advocated the development of what they

called at that time “sociodental indicators”, a number of investigators have taken steps to

develop ways of measuring the impact of oral diseases on the well being of communities and

individuals. As a result, twelve instruments have been developed, as summarized in Table 2.3.

They are all self-reported measures, where the development was based on the contemporary

concept of health. However, they are different in terms of the number of items, which range

from 3 to 73, content, the domains, response format and the scoring methods.

Le

“ayy Jo

Ayipenb e800

G66} “Ie

je sigpunes

ysibua/yvsn ‘Uyesy

eso payel-yes

‘uonLnu ‘(swe}!

GL) yyTeey

je10 9S

Aioyusay| ast]

JO AWENH UYeeH

[210

eV66L ‘JONI

pue 194907]

*SUOHElAI [eIOOS

(ISHOS) sso}eo1puy

ysi|Susepeuep

‘uoneoiuntuwoo

‘Bulee ‘swojdwhs

‘Buljeeds ‘Buimayo

ev snjyejs

yyeoH

jeiO eanosiqns

‘deoipuey ‘Ayiqesip

jeioos BY66|

‘leoueds pue

apes

ysyGuzyeyeysny —‘Ayjiqesip

jeojGojoyoAsd ‘Ajyiiqesip

jeoisAyd ‘ured

‘uojoun4 6Y

(dIHO) SYUd

joeduy yyeeH

[10

‘sdiysuonejas ‘asl

C66

‘UNH pue

ssneys ysBusyysn

=‘ jeloos ‘ssourddey

‘souapyuoo

‘yoseds ‘Buljese

‘soueieaddy

GZ (did)

ayole

Joeduy] jejueq

O66} ‘UejOg

puke UOSsIYyo} 7

"SS@USNOIOUOO-}1as ‘AIOM

‘uled (IWHOD)

Xepu] ysibuay/yvsn

‘gouBseedde ‘sjoeju0o

jeio0s ‘yoaeds

‘Burnes ‘BuimeyD

ZL jUSUSSASsSY

YYESH [eIC

D}eUeD

QINSIO} ‘YIOM

686L ‘12

Je SuIsioy

ysibua/yvsn ‘yenjoayjaju!

‘yooeds ‘uoNoeJe}UI

;e1I00s ‘syse]

SwOY sey

el aOld

oediuyj sSauyoIs

686} ‘Te

8 yoo

ysi6us/vSn

UOHESIBAUOD

‘ALIOM ‘ule

¢ xepul

je}ueq GNVY

Q86L ‘12

Je Bulysno

ysi6us/yn

eoueieaedde ‘ured

‘Buiy6ne ‘Buijs

‘Hurye} ‘Buimeyd

PL gjeos

jejuspois0s

abenbue7

suo}! siouny

jpedojaasp Aujuno9

uoIsuBwIGg jO

‘ON yUSWUNI}SU!

BU} JO

oWeN

SJUSLUNIJSU! JOONHO

Suysixe ue

yo Auewiuung

"¢'z ajqey

8C

"yjyeaiq ‘oywod

‘Ajyeuosied ‘eoueul

LoOg ‘peg

8 YEID9/\I

‘WOM ‘pool

‘saujeseo ‘eouapyUOd

‘YBnel/ajiws ‘eouewos

TCM) 1TOOMHO])

ysy6usayyn ‘qe100s

‘deajs/xeja ‘uyeey

‘yooeds ‘aouesesdde ‘Bujey

OL ‘alr

Jo Apenyd

payejoy uyeeH [BO

‘uojowe ‘esougieadde

Q66L ‘WweYlsuS

9g UOUeAiNpY

ysi6uz/-yn ‘Buideays

‘auaiBAy jeio

‘Buryeeds ‘Buea

ui SOURBLUIOLIag

6 SOUBUUOLe,

Ajleq Uo

yoRduy jeIO

9661 “Ie

Je ulssely

ysii6ua/vsn ‘UOHESJOAUOD

‘saHiAyoe jel00s

‘sanianoe Ajleq

€ apr

Jo Aen

payejes-yyeoy] |e1O

G66) ‘weyleys

pue cee)

ys bulyyn

Buijee ‘senanoe

Ajlep ‘ured

‘eoueseadde ‘ojwog

9¢ Buia]

Apeq uo

yoeduuy} jeyueq

abenbue}

swe}! sioujny

jpadojaaep Anunog

UOISUSWIG

JO ‘ON

JUSUINIISU! BU]

jo swe

‘(penuljuod) syusWNAySU!

JOOYHO

Buljsixe ue

jo Asewiuing

*¢°Z ajqeL

29

However, only four measures, which are the most widely tested and used, are reviewed

in this section. They are the Oral Health Impact Profile (OHIP), the General (Geriatric) Oral

Health Assessment Index (GOHAN), the Subjective Oral Health Status Indicators (SOHSI) and

the Oral Impacts on Daily Performance (OIDP). These measures have been used in countries

other than the one in which they were developed. Moreover, these measures have also been

translated into other languages and tested for their psychometric properties. A review of each of

the measures follows the headings as listed in Table 2.4.

Table 2.4. Standard method for reviewing OHRGoL. measures.

Purpose: Brief statement of the purpose of the instrument.

Developmental process Description of the development processes, which includes the conceptual framework used.

Description Description of the content, Domains measured

Response option Time reference

Method of scoring Administration

Time taken to complete

Property tested Brief summary of property tested.

Population/setting Original group/setting Other subsequent group/setting

Comments General comments.

30

2.1.4.1) General (Geriatric) Oral Health Assessment Index (GOHAD

Purpose:

The General, (formerly known as Geriatric), Oral Health Assessment Index (GOHAI) was

developed in the United States of America by Atchison and Dolan (1990). Its aim was to

measure oral functional problems and to estimate the degree of psychosocial impacts associated

with oral diseases in the elderly population.

Development Process:

Conceptual framework: The GOHAI was based on the definition of oral health as a freedom

from pain and infection and consisting of a comfortable and functional dentition (natural or

prosthesis) that allows the individual to continue in his or her desired role. Based on this

definition, three hypothetical dimensions were derived: 1) physical function, including eating,

speech and swallowing, 2) psychosocial functions, including worry or concern about oral health,

self image, self-consciousness about oral health, and avoidance of social contacts because of

oral problems, and 3) pain or discomfort.

Item generation: The items were generated using three methods: 1) literature reviews around the

area of oral health and disease impacts, oral functional status, patient satisfaction, oral

symptoms and self-esteem and socialization, 2) consultation with health care providers, and 3)

interviewing people attending a senior center in Los Angeles and a Veterans Administration

hospital dental clinic. A total of 36 items were generated to reflect the three dimensions.

Item reduction: Item reduction was based on the internal consistency and frequency distribution

of the response options. A final instrument containing 12 items, which exhibited the best

31

distribution of responses and maximized the Cronbach’s alpha for the instrument, were chosen

to represent the three dimensions. The items chosen were worded both positively and negatively

to discourage respondent acquiescence.

Description:

The GOHAI contains 12 items with a six-point frequency Likert scale response option (always

[5], very often [4], often [3], sometimes [2], seldom [1], never [0]). The time reference used is

three months. The GOHAI score is obtained by adding the response code for each of the 12

items after reversing the coding of the three positively worded items. The GOHAI score ranges

from 0 to 60.

Properties tested:

Reliability, in terms of internal consistency, was assessed in the original development with a

Cronsbach’s alpha of 0.79. Component principal factor analysis showed that the instrument was

measuring a single construct. Two types of validity were assessed: criterion validity and

construct validity. It was found to be valid. An Initial study on the responsiveness of the

GOHAI, found that the instrument was sensitive to the provision of dental care among the

elderly population (Dolan, 1997)

Population/setting:

Table 2.5 shows a summary of papers published using the GOHAI and Table 2.6 gives a

summary of papers describing its translation and adaptation to other languages and cultures. The

GOHAI was originally developed for use with older adult populations. However, recently it has

32

been used with population of younger adults (Atchison et al, 1998). It has also been translated

into Chinese (Wong et al, 2002b) and French (Tubert-Jeannin et al., 2003)

Comments.

Because it is a short instrument (12 items) the administration of this instrument is rather easy

and cheap. Since this instrument is an index, it gives a net impact of oral health, thus it may not

be possible to disaggregate the contribution of different domains of health to the overall score.

ee

"9409S XSPU!

||2IBAO xa]dwoo

OU} SB

[JOM SB

'|VWHOO yuswpede

BU} JO

SUS} [ENPIAIPUl

JOIUaS 9]

UO SBOlAJas

je}USp

poayene

SNOLIPA Jo

joed UI

Sy} ~younyo

pue}sJepun

O}] pepssu

‘pezipisqns

S| YOJBOSOl

|BUO!IPpe Ajjesepel

e

yBnouye ‘aueo

JO sjuapisey

|e}Uap Jo

UOIS|AOJd au]

(Buydwes ‘aBueyo

Joe}ep

0} SAISUAS

SI [YHOO

anss! |@9

MO|AJO}U} (sueah

Z) DHEWUSISAS) |

0} AyIge IWHOO

2661 ou}

yeu} siseH6ng |

ibojopoujeyy 90BJ-0}-9084

ysn j2uIpNyGHuo0 7

‘+69 OOZ=N

ssesse O}

uelog

"s9jdwes JudlayIp

OM} BU}

Ul IWHOOD

‘saidues jo

uoHeysiujwpe

Jo Ajauea

e ul

Bulag WO, seledod

-|}aM pue

Bujuo youn ol4jowoyoAsd

pue

uO SUOI}IPUdS

JEUOTNGLISIP SU}

jeJO JO

Joedu! G4}

Jo asedwos

0] (Z

Joyeolpul ue

se |VWHOD

‘suemisAud

jo asn

penuijuco Ajtunuiwos

Buisn

jsoebins sijnsey

- sjusijed

aueoipey\

‘USUI JEpjo

jo sejdwes

(2 jo

ajdwes

om] UsEMjeq

pue Alyejueseidas

e pue

seo

Buowe

saouasejjip Aywesu)

yyeey WA

Buisn

o1ydesHowuapojoos

ZPG=N

sjuaned ojew

O}] SAIPSUAS

S$] }|

- :ejdwes

WA -

Jepjo jo

ajdwes

‘sejdues (ajdwes

3 B

Ul WHOO

yjlog ul

seliedoid eu

OUa!IUaAUCD) 4}

UO S8l00S

oyeuuOYyoAsd

:ajdues VA

- 66Z=2N

UUM peyeloosse

Z66L

Asopoejsiyes A8ains

auoyds|a} jeuoyjoas

+9

‘ojdlues SJO}JOR}

‘yey

SUGIUXS |[VWHOD

- UyeoH

| ‘aseo

aueolpey -

ysn -SS0l9

Udi aJBOIPS

- euiuexs

Of (LP |

ulssely

vol elslu|WwipYy uoIsnjoueD

uonesjddy

IVWHOS

uolje007 | ublsep

Apnis

aby

ajdwues wu

souny

IWHO®O Suisn

peysijqnd sueded

yo Arewuung

‘¢°Z ejqe

ve

Qu} JO

uoHeys|uWupe s,uelisAyd

pue S,jsHUep

UsEeMjeq juaeweeibe

pooh

Buyesipul 19°70

40 JUSIOIJ9OD

UOEfOI09

SSPJOBU! OY!

- ‘peyedionue

aie peou

Jo sjena]

yBiy WOUM

ul uoneyndod

e ul

joo} Bulugeios

jnjesn

@ JOU

SI WHOS

SUL -

‘pooB Jo

poo6 AJaA

‘Uua||eoxe

se yyeoy

[eso JI@Y}

payes sjuedioyed

Bu} JO

%OZ ‘speeu

pue eseesip

ybiy aydseq

‘speou

-uou Aq

pasn

8q PINOD

IWHOO

Ou} JOUJOUM

suiuexa of

(¢ ‘uoyejndod

Aepya [ej

Qu] Ui

JUSUeSI}

je}Uep jo

pseu ul

suosied

BuiAyquep! Ul

[YHOS

94} jo

ssaujnjesn ay}

SUIUeXe OL

(Z “2018S

JBdIPaiAy SOY

s,uo}sog Aq

peases sjusned

punoqetwoy au}

Jo snye}s

Yyjeoy

jUSLUJeel)

BAl}Oe[qo (Buydwes |

jeio pepodel-jes

pue saiieo

‘sso| wopued

pue peulwajep

6661

JOO} JO

|eAg} UBIY

Asains j2ualoas

onewe}sAs)

Aypeoruyjo yejeo

pey ajdwes

siyt -

UyeoH MAlAJO]U}

yen -ssain

| Apeply

OS=N |

au} ssesse

Oo] (1 |

saiqereg

‘UBILOWY

“sjeuolssajoid -UBDLIY

pue

yyeoy jejuep

sojuedsi

JO} ajGenjea

ose jeu}

suojeyndod

uyeey [eso

Jo sjoedse

p

juepodui

ssiw Aew

oousoquBbiau

UyJeey JO

sesnseow SWOSU!

peyiodal-}es -MO}

OM ‘UBOLIOLUY

gueUueb Jo

asn su}

pue (6uyjdwes

-UBOLIY

yyeay jessueb

wo

@nisodind) pue

soiuedsip

JOUNSIP SI

UBESy [es

- 9c=N

jo ajdwes

ynpe

"sojdues ‘ueoWeuy | e6e-\je

ue ul

pesn

asenip Ajjeojuyje

pue UBOHY

USYM IWHOOD

8661

4e6unod ul

pesn usum

anssi jeo

AASIAIOIUL j2UOI}OeS

O8SZ=N 8uj

JO ATIpHeA

ye je

DHEA SI WHOS

eUL - |

Gojopoyjew 908}-0}-998

4 vsn

“SSOlD wnpy

‘soq1uedsiy |

oy} ayeHyseAu!

OL | UOSIYOIY

St

uyjeay jeso

ajeiidoidde

0} sseo0y

- “AyuNUUWOo

“SJENPIAIPUul ou}

Auyapje pesiwoidwoo

wow} sjueed

Ajjesipaw Wve

- 9sSsu]

jo UOHOBysies

"Sonoel

aj} pue

Buleq-|jom ouy

e1e9 UO

JOOS JuBOWIUBIS

eB W9}-Buo0]

| “aujUueo

OJe9-UW9}

BABY SJSPJOSIP

jEIO -

ysouoAeg Huo]

& Uy}

"SJOPJOSIP [BJO

JO MLE

- Peal]

WOUM

Jo

ynsel e

se swaqold

‘aqUueg |

jsow sjenplAipul

yeioosoyoAsd pue

ysqunheg pesiwosdiuod

jeuoHjouns paouaiedxe

au} ul

A Ayyeorpewu

uole;ndod

jjuepusdepul Jo

uoeyndod

Ayepje pesiwoid

woo PaAl|

%6rv -

e jO

oy

Aljeoipau GZZ=N |

jo Ayjenb pajeja.

27002 B

jo uoWodoid

Agains MAIAJOIUI

jeuoyoes ajdues

-yyeey jes

eyo

jeyjueysqns -

UYeoH e@0eJ-0}-80e~4

epeuey -SSOID

|

OOL-ZG | saUaIUeAUOD

eu} ssesse

O] JEyO0"]

98=N (aue9

[e}UNp

"aleo Bulyoes

Huryjees jou

asou} pue

Ayaayor)

eseo Buryeas

Ajaaqgoe sjuened

JENpIAIpUl UBEMJOq

jeyuaq -

ales Bulyees

S}CNUSIALIP O}

ZG=N

JOU BSOU}

YM

age st |VWHO

Su -

:(aue9 jejUep

peiedwioo aeo

‘WHOS

pue uyeoy

Buryees jeyuep

Burjysas

Jeo peyel-jjas

jeqoj6 S8-SG9

AJaayoe AJ@AOe

BSOU}

au} Buisn

painseow ueHed |

jou) ejdwes

Ul Sa9UeJEYIP

se aseo

jejuep Bulyaes

jeyueq eawoy

jo uoyenjeas

asou} UeU}

YyeeY jeso0

G6-EL Bulsuny

- 24}

Ul [YHOO

6661

JayJeg pey

ose

jeyuep Aaains

feuoljoes “OLUOU

‘ajdwes ay}

Jo Ayiqe

seyu

Bupjees jou

esoyu -

uyeeH POUOHUSLU

JON ysn

-ssoij |

Bulsuny | eouaUsAUOD

au} ssesse

Oj |

aueosey|

‘WHOS

“uo}eyNsUoS

jejuep Buysenbe.

JO} poeou

QU} SUILUJE}Op

0} sjeudissajoud

uyjeey jejuep

9¢

"JUBLOC

UI

Buleq se

psebol

SIENPIAIPU! BSeu}

yeu} apy

jo sjoodse

Buynseew oe

bi-dIHO PUB

|VWHOD -

“OHl] JO

Ayyenb |JesaA0

sAosduut

0} Ajeyl|

St eed

Le

— OPO

:uaITejasJ0O gjBOS-tWs]]

- 98'0

:eudje s,yoequolg

-

JUSIO1jj809 UONEIaU09

$,UOSIed¢ :UOI]e|esJ09

8109S "Yous

sem abenbue|

jsiiJ @SOUM

‘IWHOS 284}

‘Aiqeljes pue

Appiyea -LUd}|

PUB WA}-O]U]

=e | syenpialpul

fenBuljiq eouy

| JO

UODISJOA BOURL-

€00Z

ajqej;deo0e payeujsuowep

"eydje S.yoequolg

oe ‘WoVe|SuelT P

leMIO4

au} Jo

AypyeA |

‘jeje ujuUeer

UOISIOA [WHOS eouel4

out AWIqeney

youeiy 0}

ys Bug

pSuoHUSU

JON OU}

}SO} OL

-Hoqn,

‘ajdues BOUSIUBAUOD

- ALAPJOe OZ

‘BUSS Old

“s}sq}Uuep juspusdapul om

"ysi{6uz 0}

eseulyd ‘Uohejsuey piemyoeg

sioy}ne ‘peeu

jueWUjeol} peAlsoled

auy Aq

auop —

uorejsued}

(syoefqns pue

einseaw pJEMJO}

OU} UI

pasn sem

YYEOY Jes

Psjes-jjos We}!

uoissnosip dnos6

snoo} -9[Guls

@ pue

81098 |YHOD |

ay)

wow UOeWWOJUI

aU, useMmjeq

UOEOssy

oe ‘asaulu9

0} Ysi6uy

28'0 {JUuaIdIye0o

sisjeue J0}Oe|

/UOe|suely Premios

UOHe{909 SSBjejul

oe jUSUOdWOD

jedioUd =

"21098 [WHOOD

€93'0 JUSINIJe0o

Aypijea yonsysuog

“senss! uyeoy

au} soueN|jul

uonejauo0o suewuesdS

8 jeso

jnoqe suJeoUoS

Aewi Ajqissod

}SO}O/-]S9 | ‘(9am

Wey} sseidxa

ALLapje Jey}

Su0}OR}

190 auo)

eddey payyhiem

au} MOY

pue}suepuUNn a}eBI|SOAul

O| -

-97'0 1UOI}e}a1U09

a}eos-tuay} - |

PUB JUS!D1JJoO9

UOHEje1109 0}

— Apepja

JO} azyue0

‘Buoy

L3°O ‘Bude

s,yoequoly

- yues

suewueeds

jeioos suo

pue swoy

Buoy

ui Ayapyle

jSo}eljsol

Ayseple om)

ul peyonpuos

30} IWHOD

JO

“Ayyiqeryjes pue

Aypyea UOHEaOD

BJeos WY

=e | | auam

sdnoi6 snoo}

eauy tL

UOISI@A BSeUlyy

giqejdeooe payesjsuowep

"eydje s.yoequoig

BU} BJEPIeA

GZ00¢

UOISISA [YHOD 8seulyD

Syl AWUPIGEHOY

asouly

o} ysy/6ug

pauojuew

JON | pue

ajejsues Of -

‘je Jo

Buona

yYiOMOWes |

uoisnjoUueD pessessy ealy

uoljesuel] uoneydepy

wy

soupy

afenbue] pue

sein}jnd 48430

0} [WHO

24} Jo

UoHeIdepe puke

uONelsuel] 9°Z

B1q2L

Be

28'0

0} LG'Q

:eddey peyybley,e

28'0 :JUa101Jje09

UOI]TEJG1IOO S,UOSIES

=e

‘JSO}Ol-]SO]

82'0

“snjejs feyuap

jeoiusjo

BY} UM

84098 |VWHOD

ay} pue

‘sesuodsal wal}

syuedioed

oy) Buiedwoo

‘AJIPIJEA JUBUILULOSIQ)

“UHeESY eso

UM

UOHORJsHes pue

peau JUSLU;eeL}

PeAleoied spolqns ‘aunseow

uyeay Jeo

peyel-jjes we}!

-ajGuls e

pue 91098

WHO

usemjeq ucielossy

‘AUPIJEA JUALINDUDD-—®

Aypljea yonysuog

‘(syaam ¢)

eddey

PUB JUSIDIJJO09

UOHE|e109 $,UOSIedd

JSO}Or-]SO] ¢

‘9jdlues s}Npe

O6

‘BURSS} JO}ld

"ysi6uz sem

obenbue] si aSOUM

sjenpiaipul yenBuiliq

suoHe|Suel) puemyoeg

39

2.1.4.2) Oral Health Impact Profile (OHIP)

Purpose:

This instrument was developed in Australia by Slade and Spencer (1994a). The aim of this

instrument was to provide a comprehensive measure of self-reported dysfunction, discomfort

and disability attributed to oral conditions, which can provide information about the burden of

illness within the population and the effectiveness of health services in reducing that burden of

illness (Slade, 1997b).

Development Process:

Conceptual framework: The conceptual model proposed by Locker (1988) was used to identify

conceptual domains (see page 12 for the description of this model). Seven hypothetical

dimensions, namely functional limitation (e.g. difficulty chewing), physical pain (e.g.

toothache), psychological discomfort (e.g. self consciousness), physical disability (e.g. changes

in diet), psychological disability (e.g. reduced ability to concentrate), social disability (e.g.

avoiding social interaction), and handicap (e.g. being unable to work productively) were

created.

Item generation: The items were generated from open-ended interviews with a convenience

sample of 64 adult dental patients in Adelaide, Australia. A total of 535 statements regarding

the adverse impacts of oral conditions were identified.

Item_reduction: A content analysis was conducted and resulted in identifying 46 unique

statements. These 46 statements were categorized into those seven conceptual dimensions.

40

Three additional statements from an existing inventory were adapted for use in the handicap

dimension and gave rise to a total of 49 items.

Description:

OHIP contains 49 items grouped into seven subscales. The number of items in each

domain varies. A five-point frequency Likert scale is used as the response option and treated

both as ordinal and interval data. The time reference used is one year. Each item is given a

weight indicative of the severity of the problem it describes. Unweighted and weighted scores

can be computed for each domain and the OHIP overall. The design of the OHIP permits it to

be administered both by interview and self-completed questionnaire.

Properties tested:

Test-retest and internal consistency were assessed in its original development. The Intraclass

correlation coefficient for test-retest reliability ranged form 0.42 to 0.77 for six subscales, but

only 0.08 for social disability. Cronbach’s alpha values for internal consistency ranged from

0.70 to 0.83 for six subscales, but only 0.37 for handicap. Construct validity was evaluated.

They concluded that this measure was valid and reliable for use in the target population, that is,

Australian elderly population. A study done by Allen et al (2001a) demonstrated that OHIP has

good ability to detect change (responsiveness).

Population/Setting:

Table 2.7 summarizes papers published using the OHIP and Table 2.8 shows a summary of

papers describing its translation and adaptation to other languages and cultures. Originally, the

OHIP was developed for the Australian elderly population. Subsequently, other countries, such

41

as Canada (Locker and Slade, 1993) and America (Hunt et al, 1995) have used it for their

elderly populations. Recently, it was used in an adolescent population in the U.S (Broder et al,

2000). It has also been translated into German (John et al, 2002), Chinese (Wong et al, 2002a)

and Canadian French (Allison et al, 1999).

Although the OHIP was developed for descriptive studies, it has been used as an

outcome measure for evaluating clinical interventions such as osseointegated implants (Awad et

al, 2000) and in resource allocation by identifying patients who are most likely to benefit from

dental treatment (Locker and Jokovic, 1996).

Alternate form:

The OHIP has a short form - OHIP14. This short form was developed using a controlled

regression method in which the OHIP score was the dependent variable (Slade, 1997c). Two

items from each domain were selected based on their contribution to the R-square. Locker and

Allen (2002b) produced an OHIP short form using the item impact method. Table 2.9

summarizes papers describing the OHIP short form.

Comments:

OHIP is considered to be the most sophisticated measure of oral health, since it covers a broad

range of quality of life dimensions. It was based on a coherent conceptual framework. Although

the way in which items were selected is not clearly explained, they were selected from a wide

pool of items, generated from interviews with the target population. This ensures, to some extent

that the content coverage is satisfactory.

The OHIP can be administered both by interview and self-completed questionnaire.

However, a randomized cross-over study of Australian adults by Slade et al (1992) found that

42

the OHIP-49 underestimated oral health impacts when administered as an interview. One of the

limitations of the OHIP is the length, increasing the cost of data collection.

ev

"sosoujsold

peziiqeys -jue|cduut

SU} WO}

HJeUaq 0} Ajayi]

s}uaed

BulAyuep} uj

(nyasn sieedde

diHO -

“@W0d}NO yuowjeo}

Jo s0}01PaAId

poob e

aq 0}

jou Jeedde

suojeypedxe

SPUBHE

- "jueWea.}

paJisep St=N

MOY) PSAl@dal

‘(jueuaoRdas

sjuayed I

aunjuap "9WU09}NO

BIOLU SEM

Aji] JO

J2UOIJUBSAUOD uo

uoHe}edxe

Ayyjenb poyeje.

pajsonbei aaneiadoaid

~Uyeay au}

JO GUM

9@sou}) jo

aouenyut

JUSLU9A0JC LU]

- Z ainjuap

au} ssesse

Of (Z

“sjuaed jeuo|jusAUOD

‘sounseau snjejs

snojnjuepe OZ=N

:(uejduyt |

yyesy oyjoads-jes0

jo af)

Jo Ayyenb

aAlo0eL ‘paseq-jueyed

psyeyes-yyeoy you

pip jueydul

Buisn siaieam

ayy uo

yoR du} paysenbel

ainjuaep 9}8;dWOs

e@anisod @ eAeu

OUuM 2souy})

jo smjejs

uyeoy

ueo sesoyjsoid

|. aunjuap

Sy} UO

siseujsoid

peziiqe}s JEUOHUBAUOD

PSZiiqeys

~juejduu! Jo

wopBbuly peuipny6uo}

O¢=N ~juejduu

joedu! qL00é

JBUOIJUSAUOD - |

jel} [BOIUD

pauoljUSw JON

perun aanoadsold

ynpy | :dnos6

jueyduy au}

ssesse OF

(1 ‘je

yo UdTIY

diHO 4°

uoisnjaue4

uoeoyddy | uoyesjsiujwpy |

uojeso7 | uBlsep

Apnig

aby

ajdwes

wy

soyuny

diHo Sursn

peysyjqnd sueded

jo Arewuwings

“2°Z e1qeL

vv

Buowe

peyJeuu

Ayeinoyied sem

yoeduul slut

-

"SUOITIPUOD Je10

HOU} UM peyeosse

SOAl Ajtep

HEY} UO

~eCdUI

eneBeu

saey 0}

peyoda

ajdoad Jepjo

jo uonodoid

JEHUBYSqNS -

:eyep AJBUIUUIOl

“AUDIEA (SHNPY

JEPIC | “S}INpe

Japjo jo

ayy

yonsjsuoo poob

jo uyjeaH

Jo Ayyenb

peyejes

eyeajsuOWwag -

{240 8u}

jo ujeey

jes0 uO

“06'0 Apmg oejuo)

eyep Aseutwuyoud

0970 ‘eydje

(8ve=N) apiaoid

of (Z

S,YORqUOold -

Ayunwwoeo

soiedod

:sansedoid Asains

(jew) epeuey

j2uol}oes ueyjodojew

juguainseaw |

SE6L epeIS

DLUJ@WUOYOAS uiesH

pajadwico-jjas Ol12JUO

-SSOlD +05

-UON eu)

ssesse of

(, |

pue 194907]

“diHO au}

UJIM peunseow

yl Jo

Ayyenb

“Bjl| JO

payejel-yyJesy

Ayyjenb payejes

uo ‘sjuejdui

-Uyeay fes0

omy Aq

payioddns

ul JUQWJEOL]

sesouysoid

jEUO|JUBAUOD (pauBbisse

‘SA SOINJUSP

JOA0 Ajwopuey)

JEUOIJUBAUOD

JUSLUSAOICUI Jejngipuew

WWt8}-OUS Sb=N

‘wuslnjuepe

jueowubls (S9-Se)

dB

aunjuap JO)

SJUSWWES]

saplaold (dIHO

40 JEL

PEOUND ynpe

JEUDIJUBSAUOD jo

sodA}

JUSUW}BO} UOISJOA

YOUS!4) epeued

payosju0g sno

pr=N OM}

JO sjaye

0002

queiduy |

jeu} jeouyo

pajajdwoo-}}3S ‘Teeqjuoyy

pezimopuey |

jnjuepy :d6

yuejduy] ay]

eredwoo Of] |

je je

pemy

Sv

“sueaA snolaed

ey Pepusye pey

OuM pue

saspuaye onewoydwAs

9J8M OUM

@SOU} Aq

payiode,

yoeduul sayBiy

YM

‘susJeyed

uoiyeZziyn J2}Uap

JOU} 0}

5ulpsoooe

suosied

ayejuap Buowe

SUOIELIBA 4e6se7-

‘yoedull Jayeoi6

Ajueowubis

UM

pejeloosse Sem

abe JApPIO

- ‘snojnjuepa ayes}

Ue} yoediui

alow

Ajueoyiubis

peyode

ajew

snojnjuepe asaum ‘Buree

pue Burmaya

0} pejejei

seaue ui Aeynoned

‘joeduul jeioos

p SZHEUOHMNYSU!

*sdnos6-qns Aq

paousiiedxe

yoeduul

[BI00S Jo

}U9}XE Qu}

Ul SUOI}BUA

ssesse O01

(Z

ALepye

pezyeuolynyysut -UOU

BU} JO

UOHIPUOS jeJO

YJIAA peyeloosse

jyoedu

juenba aol

-UOU (J

LZ-=N) ye1o0s

peyode!

Gr661

peyodai uosied

Aaains eljeuysny

jeuoioes ajdwes

jo Aduenbea.y

ay) Jaoueds

snojnjueps -

uU]eoH P9}9/dW09-}jeS

yjnos -SSOlD

+09 wopuey

yuauunsop Of

(1 pUe

SpE|S

"“gued jejuap

JO} S}ISIA

sejnBed

yew

JOU pip

oym sjenpiaiput

snojnjuep pue

snojnjuepe UjoqG

ov

Jebuo; Buynp

“uosiad Bulljeap

SUOI}puco -AyUNUIWUOO

jeio jo

syoeduut Buowe

UU9}-}4OYsS SUOHPUCD

JBJO Jo

eousuedxa

yoedul peyodey

ul

synpe AdAins

(re) eyjeasny

e6ueyo jo

wepjed eO66/

Jepjo Aueyy

YESH

pajejdwuos-jag YInNOS

jeuipny6u07 +19

O6=N aquoseap

Of | |e

je 8peis

“sdnol6 j2joes

OA} BSA}

uaemjaq Aioj}siy

HSIA jeyuEp

DUP SNye}s

jes0 Uf

SSOUBIALIP 0}

pexul ae

Ajayi} syoeduui

peyodal

Ul SSOUSJO}IP

aul -

"SOUYM (Apms

a}e]Uep Japjo

uoyeindog |

siinpe epym

Jepjo

uey} suuayqoud

Ayeapyy pue

sjinpe yorq

jEJO WO’

ul oseasiq

JEplo Usemjeq

yoedui asow

j2IO Jo

yoRduy| SJOPJOSIP

jeJO JO

peyiode. syoesq

Aaains euyjoles

Jeuc}oeS feloos

ay1) |

yoedui ul UoeWeA

S66L

BIEJUSP JBP/O

~ Yee

pe}a}dwiod-Jas ULON

“Ssol) +99

Ory =N |

24} aeHyseAu!

OL j@

}@ UNH

“s]inpe Jepjo

ay} Buowe

wisiinjuepe pue

ssoj JOO}

JO ayeJ

yBiy Buipnjou

‘g0uUa|Jedxo

eSeasip jo

SAAR] SAISUBIXO

$}OeJel

Apnjs sitj

ut peyoda. joedu

jo Aouenbey

yBiy aul

-

LV

B pey

pl-dlHO PUB

6PdiHO

- "seunjuap

feuouUeAUcd Bulyjees

asou}

UB] SSO}

Y}OO} Aq

peddeolipuey “dIHO

$0 SWA}!

7}

jo

spoued

pue pajqesip

pue 6

Jo Ayipifea

‘pasreduuit ei0w

eu} sledwos

O| (Z

Ayueoubis

“‘SJOIBOM aJNjUSp

aJOM jEUdIISS

Qr=N

‘Jueldwy Q}9|]GWO9

Ul S$so}

R661

sdnoui6 yuejdui

Asains wopbuly

-SSOJO GE=N

(}04}U0D YjJoo}

Jo oe dul

UBINAOIN

ul syoalqnes

- UNeeH

POUO!}USL

JON perun

jo.jUOo-asey) ynpy

‘sdnoi6 om

ay} ssasse

of (1

pue udjiv

eplejepy ul

ued

pepunj-oijqnd

GHulaleoe, sjuaied

jeJuap jeraueb

yy UOSsedu0s

‘sjuened Ul

UOHOSsu!

jesouoh ue}

Al UM

sjueyed

yoeduii jeloos

GrL=N

jejuep Buowe

Jo sjengy

soyeoub :sjuaned

SUOI}IPUOS |eJ0

Ayueoyiubis jeluap

jesoues Jo

yoeduu jeioos

payoda,s ASAINS

eljeasny jeuoyoes

PS =N

pue snjejys

uyeou 9661

sjuaned AIH

UJeoH

pejajdwoo-}jaS ujnos

-SSO0ID Wwpy

juaned AIH

jeso ssesse

O| | je

je sayeod

BONO

JO

S}E}S payun

pue eeisny ‘epeueo

‘joedul jetoos

uj uolejndod

Ss} pue

yyeey Z¢@y=N

‘VSN Ayapje

Buowe

jeso Bulouenyul

9¢6 SUOHIPUGD

JeJO

SJO}OR peinyjno

vsn

=N ‘eleasny

jo yoeduyi

jej90s

pue |elo0s

ASAJNS eieusny

jeuoyoes 612

QU} Ul

SUOITEEA

qo661

ase aay t

yee

peje|dwi09-jjaS epeuey

-SSOID +9

= N

:epeuey

ssasse oO]

je 38

epR|S

“UYEoY j240

jo yoedtuui

peajeoiod

ut Ayyiqeys jesodtue}

8V

Jo dnoib-qns

e peeu

OYM siinpe

Ayquap! 0}

9/qe (SHNpY

J8PIO Japjo

Ayjuap! eze

Agu} ing

jo yyeoH

0} suoyeo1pul

*s}00} Buluselos

{JO 94}

JO yyeey

|elo 9661

se ||JOaM

Asains epeued

jeuojoes Apnyg

ouejyuc) | eagoolqns

jo Ayyige

oIAOyOr

pewuopied JON

yjeoH paje|dwioo-jjes

OWeIUO ~SSOlD

+0S €67=N

ey} ssesse

o| | pue

s94907]

‘aides Apnys

siyj Buowe

9E-AS Ou;

UBY} SUOIPUOd

yyeay [eso

jo sjoeduj

ou}

UIIM pajeloosse

AjuBiy BJO

8G 0}

sueedde qIHO

-

“snyejs UYeoy

jeso peoruyo

100d Jey}

“sjusosajope Ayo

UIIM payeloosse

~Jauul ay}

Buowe

sjoedu UjJeay

felo jo

yealBojouoAsd slojeaipul

jeoluijo

pue 0}

‘diIHO eu}

pue

jel90s ‘jeajsAyd

9E4S GNVY

84} a|qesopisuod

‘gwioojno payiode,

psoueledxy -

-}J9S jO

Sounseew

‘aidwes Apnjs

om) jo

diysuoiyejei

ay} Buowe

8} pue

SUOITpUCd

spseu jawun

sjuo ajdwes

yyesy jeso

jo

pue saves

Asains jeuoloes |

osajope

BOUdIUSAUOD yoeduul

panisaied 000z

Jo j9A9]

YBIH -

uylesH PSUOIJUSW

JON yvsn

-SSOln | Ayoul)

S6=N au}

ssasse Oo} |

‘Je }e

sapolg

“Bunes

jeaiusjo @

Ul ple

jnjesn e

aq

Aew ¢L-diHO

- ‘sdnou6

ay] USeAjeq

@}EUIWWLOSIp

0} Ayjiqe

Jey wus

6P

JSOLU SU}

WJyeUaq Ajqeqoid

pjnom

OUM aj}

Ajtep uo

Agueoyiubis

poyoedu!

SUOI}INUOD Jeoruyo

asouM

S/ENPIAIpU!

jugu}ee) f2]UEp

SOINSBOU O41]

JO ajdwes

Ayyenb payejas

anssi je

eyesjsny jeurpny6uo|

WOPUeY

“aBueyo

-yyeey |e1E |

Hojopoujey| Pe}o/dWiod-jeS

yInoS |

jeuoyeAlesqO +09

86v=N fOJUBUUSSESSY |

8661

SPEIS “sdnob

usemjeq SSQUAIALD jueoiubis

BulAyjuep! Jo

Sw9} Ul

Spoyyow Buyoos

49yjo Oo}

pesedwoo

dIHO St}

“dlHO Jo

souewuoped

(sIINPY 4EP|O |

aU}

jo eoUeWOLed

ey} sAoiduut

jo yyesH

@U] O}

a}Nqu}UOS

Ajyeueysqns EIQ

OY} JO

S}yH}am We}!

L661 you

enssi jeo

epeueg jeuonoes

Apnjyg ove)

OU JO

JaYJOYM 194907]

pip BunyGien, |

Gojopoujeyy pa}a|duo0d-jaS

OWe}UQ “Ss0l1D

+0S ooG=N

SUIUNO}OP OL

pue ually

(sHINPY JepIO

JO UIESH

JEIQ St}

jo Apnyg

oveyug) (8pe=N)

"g0}eoipul Aytunwiwos

jeoruyo uepjodosjew

SJoJeoipul

pue ainseew

-UON jeoluyo

pue

payioda 9 |

eunseew peyiode

“9S UBEMjOqG

(6G9=N) “Hes

Usemjeq v66L

uoyejes09 enssi

eo epeueg

jeudOeS Ayiunuiwos

diysuonjeje: |

apes pue

yea |

Bojopoujay| paj}eiduw0o-3aS

OuRIUO -SSOlQ

+05 ueyjodojey\

QUIWEXE OL

489907]

Juewyes} jeJuep

Woy

Os

JEUOIUSALO)

OZ=N ‘(juejdu

“S}UBIOM @AI900)

jugwigjejs Buisn

jou pip

juejduui

Aq paaoiduul

peysenbal

jou sem

ouM esou})

Aysedoid siul

- | ainjuep

*poo6 sem

jBUOHUSAUOD “dIHO

jo eBueyo

dIHO 40

eBueys anssi

jeo wop6uly

OZ=N 0}

AVAIISUaS 21002

0} AyAnISUas

- | IBojopouUey\

psa}edwoo-jjes payun

ynpy |

:dnos6 juejduy

Su] ssesse

OL ‘Je

JO UST

“‘sounseaw o148ue6

Uuey} Saseasip

(e10 Jo

S@wi0o}nNo Buunseaw

ul esn

sayees6

JO @q

[IM auieuuonsenb

oloads je1O ‘you

seo0p 9E4S

au) ayum

‘sdnoib ‘9¢4S

eyeiedsip LZ=N

[J04}U09 ay}

‘eINSeSW }I]

Aypearuryo e}ejueq

jo Ayyenb

payejes

usamjaq GE

=N :J]O4}U00

-yyeoy oeueb

@}EUILUOSIP JUaWU}eaL}

JBUOI}OSS snoinjuep3

2 uM diHO

ued dIHO

eu} anssi

jeo BuAIeoel

JOU wopBuly

-SSOID Zo=N

ay} Jo

AyipHea 6661

yeu) saBGng |

ibojopoujeyy pa}ajdweo-}jaS

poyun

jOsjU0d-aseg ynpy |

:dnos6 jueyduy}

ay} aedwos

Oo] "ye

jo USTIV

“soipnys

peulpny6uo| ul

eGueyo

HuizAjeue pue

Buizyenjdeouca

JO} Sajyixedui0s

Mau Buyeeio

‘snjeys Uyeay

ul UOReJOWa}Ep

pue jusWeAd/duI

yjog aunjdeo

Ig

Se=N (JuaWecRdel

eainqusp

jeudl}JUBSAUOD

pejsenbel

OUM 8Sou})

Z aunjuep

cs

‘VL O<

reudyje s.yoequalg

-

c6'0 -€9'0

:}Se}el-]SE| -

‘saipnys jeurpnyBuo|

pue jeuoiyoes

-SS0J0 JO)

9jqeyns $1

}I ‘snuL

‘saljiedoid oujeuOYyoAsd

eayenjeas puke

SAITEUILWHOSID JUSIOLYNS

SEY D-diIHO

UL

-

Jeak-| ‘aousuedxe

SlUues!| spored

jeoay -

“Ss}UBIOM

Wa}! Jo

UOHJONIsUCD

‘UBLUINY 8

PUelg ‘eyde

s,yoequorg

‘(SY9OM ZOO!

}S9}01-]S9} ‘AWIQeNOY

- SSOUdAIsUOdSsAY

- (sdnoi6

pue jug6sJeauos)

Apyea yonsysuog

- *(,OAOU

@p,) SUOIPUOD

je10

JUdJEJID YM

sjualed OZ

UUM MaIAs9}U|

“AYPIEA JUBJUOD

-

JO JUeuedap Ou}

Ul! Payse} jOji¢f

jeuoissajoid pue

s}sijUap UBLUAS JuSpUSdepUl

OM ysi|Guy

0} ueWes) ‘UORe|suey-yoeg

yvsn au]

WO. Jeyeeds

aalyeu Aq

pesiAay

UOISIOA PasiAas

psonpoid s]sijUuap

UBWIIAg Ag

payenjeag

‘ys!|6uz jo

aBpeaymouy Jua|jeoxe

YIM Jayeads

ueuas

aaneu jenBulliq

auO UeWJED

oO} Ysbuy :‘WoRe[suey

PleMIO-

‘ueULas 0}

ys!|/6uq Ppajejs

}ON

eaoge pue

9g) :aby

“diIHO JO

UOISIOA UBLUIAS)

9} jo

Hul}s9} oLewoUoAsd

pue uonejsued

| 2006

‘je Je

UUOr

‘saipnys Bunjes

jesnyjno seu}

ou} ssouoe

Adua}sisu0o

jo sai6ap

sjqeuosesi

2 SEM

Jou} ‘AJOUM

9uU} UO

‘seyeosqns

UIUTAA SHUI

494} pue

}YBIaM Swe}!

jo apnyuBew

ay} ul

yO

‘peAJesqo 338M

SeodualayIp seljue}sqns

awos yBbnouyy

‘suosyedwoo paued

JO poyjew

auojsuny ‘uBipeueg

Burjeeds-youds4

pue

ueyeajsny Bupyeeds

ysi6uzy ‘ueipeueg

Buryeeds ysbuq

usemjeg uosedwog

:s]UBIem Wed}

ajdues Bupjeads-youss4

:Pd}Se] JO

‘yeziueBio yoofoid

ay} pue

siojejsues} XIS

eu] ‘jUStISsEessy

[eu

“abenbue

ysuly

sem ysiGuz

-sjen6Buljig jJuUspuedepul

¢

ys6uy

0} youel4

‘uoTeisuely yoeg

‘eBenbue

jsuy

sem youes

-sjenBuljiq juspuedapul

¢

youel., o}

ys6uq

:doyejsuen premio [

‘juauUSsesse joued

pue enbiuyoe}

uonelsuey yoeq-pjemo4

"youel4 uejpeueg

o} ysi|bug

jenjdaouos -

jenuayedxe

-

OE WOIP!

- OUBWAS

-

(e661) ‘|e

38 UWaTINg)

14Hiam Swe}

aseu}

auedwios O|

(Z ‘uoyeindod

veqent) Buryeads

youel4

e pue

uoljejndod

oWe}UO Bupjeeds

-ysi/6usy ue

LUO) S}YBIOM

We}!

diHO ule}qo

O1 (1

6661 ‘Je

9 UOSIIIV

uoIsn}sUuey

pessessy Boly

uonejsuesy ylOMOUIEI

4

uoye\depy

wy

AOUuINY

sobenBue] pue

seinjnd 18430

0} gIHO

eu} Jo

UOIje}depe pue

uoNelsues) ‘eZ

a1qeL

es

260-220

JUDIOIYSOO UOL}E[al09

$,UOSJES¢)) JSOJOI-]SO |

78°0-69'0 ‘eyde

s.yoequoig

“AUIGEN 4

pue Aypyea

poob Pa}esSUOWAP

UOISJOA

dIHO eseulyd

eyt

Aypijea yonuysuoey

"eydje s,yoequolg

‘(OAM

BUC) JUSIDIYJEOO

UOHe|a109 $,UOSIdq

ysojol-}s9 1 ‘AW/IGENOY

“awoy Ayepye

Su] Ul

S}USPISEl JO

syuaHed saya Aepla

0Z jO

ajdwes aoua|UaAUOD

— Pd}s9}

IOI

dIHO ys!j6uz

ey) ym esedu0g

‘yen6Buljiq aie

OyM s}sijUuap

yuepuedapu! om,

‘UoResuey-yoeg

‘“gsaulyy jeinbojjoo

0} 9}2]SUeJ}

0} dnos6

snoo} woy

uonetusoju! Buisn

:UoHe|SUe] PIeEMIO{

payonpuos sdnoi6

sno) sesy

‘asoulyy 03 ysi6uy

peuolusl JON

‘dIHO WJOJ-JOUS

SSOUlYyD e aAuep

o| (¢

+09 :oby

Buoy Buopy

ul Ayepjea

ay} Buowe

asn JO}

JU@uNysu! payejsuely

ay} ayeplje

oO} (Z "UOISIOA

aSBUlYy oyu!

cIHO JO

UOISIaA ysj6uq

jeurby0

ay) ayeisuey

OL (,

22002 "ye

yo BUOAA

‘yJeoos UJUOWW-|

‘|eoeJ “s]s1}Uep

UeLUJES OM)

Aq POMIIANY

AUBLLIAS

‘SOHUOPOU]SOs¢e

Table 2.9. OHIP short-form

54

Author Method used Country Note

Slade,

1997c

Controlled regression method.

Stepwise regression

Dependent variable: Total OHIP score

Independent variable: All 49 individual

items.

1) Find item that makes the least

contribution to R-squared and which

is in a dimension that already has

more than two items in the model 2) Delete it and refit the model 3) Goto step 1

Repeat until it ends up with only 14 individual items.

Australia sample Contained low

prevalence items.

Locker and

Allen.2002b

item impact method.

Step 1: in calculating item impact scores

(IMS): iMS=frequency rate X importance rate.

Frequency raie= was obtained by summing

the response codes for subjects responding

‘hardly ever’ to ‘very often’ and dividing this sum by number of subject with those responses. Importance rate = item weight

Step 2: Selection of items: 1) Rank the IMS within the subscales.

2) Select two top score from each subscale.

Canada sample - Only two items were identical

with Slade’s short

form.

Wong et al. 2002a

Controlled regression method. Same as Slade (1997)

Hong Kong

Chinese

- Only 5 items

were identical

with Slade’s

short-form.

~ accounted for

94% of the

variation in the

long form OHIP score.

~- Cronbach's

alpha: 0.86

Test-retest

(Pearson correlation

coefficient): 0.94 - demonstrated

good construct

validity.

55

2.1.4.3) Subjective Oral Health Status Indicators (SOHSD

Purpose:

SOHSI was developed in Canada by Locker and Miller (1994a). The aim was to describe the

functional, social and psychological outcomes of oral disorders and conditions. It was intended

to be used as a descriptive measure in an oral health surveys (Locker, 1997b).

Development Process:

Conceptual framework: The development of this measure was based on the conceptual model

proposed by Locker (1988) amended from the WHO’s International Classification of

Impairments, Disabilities and Handicap (see page 12 for the description of this model).

Item_generation/ Item reduction: This measure was developed in an ac hoc fashion, whereby the

items were derived from a series of studies of older adult populations. In its initial development,

the instrument consisted of an index of chewing ability, derived from the work of Leake (1990),

an oral and facial pain index, an index of other oral symptoms used by Locker and Grushka

(1988) and a psychosocial impact scale partially based on the questions used in the RAND

Health Insurance Experiment (Locker, 1997b). The items were expanded in its final version.

Description:

This instrument consists of eight subscales that measure the impact of oral disease or condition

in three areas identified by the model; functional limitation, pain and discomfort and

psychosocial impacts. Functional limitations are measured by a six-item index of chewing

capacity, and a three-item index of ability to speak clearly. Chewing capacity was measured

based on the ability of the respondents to chew or bite six-indicator foods varying in texture and

56

consistency. Respondents are considered as having limitation in chewing capacity if they are

unable to chew or bite at least one of these foods.

Pain and discomfort were measured by means of a nine-item index of oral and facial

pain symptoms and a ten-item oral symptom index. Respondents were asked to indicate whether

or not they had experienced pain or other oral symptoms in the last four weeks.

Psychosocial impacts were measured by means of a three-item scale of problems with

eating, a four-item scale concerned with communication/social relations, a six-item scale

measuring other limitations of activities of daily living and a two-item scale concerned with the

degree of worry and concern about the health or appearance of the oral cavity. All items referred

to last year. They were scored on a five-point frequency Likert scale with the following

categories: never, sometimes, fairly often, very often and all the time.

Since this instrument constitutes a battery of indices, an overall health status score

cannot be obtained. Thus different scoring methods were employed for each of the subscales as

shown in Table 2.10. It was designed in a way that it can be administered by mail or other self-

completed survey questionnaire.

Table 2.10. Scoring method for SOHSI

e Chewing capacity index } Count number of ‘no’ responses.

e Ability to speak e Oral and facial pain Count number of ‘yes’ responses

e Other oral symptoms

e Eating impact scale 2 methods:

e Communication/social relation scale 1) Count number of ‘all the time’, e Activities of daily living scale ‘very often’, ‘fairly often’, e Worry/concern impact scale ‘sometime’ responses.

2) Sum the response coded.

57

Properties tested:

Reliability in terms of internal consistency and test-retest was evaluated and it was found to be

satisfactory. Concurrent and construct validity assessments were performed. All hypotheses

postulated were confirmed (Locker and Miller, 1994a).

Population/Setting:

Table 2.11 shows a summary of papers published using the SOHSL Initially, it was developed to

be used in older adults in Canada. Subsequent a study by Locker and Miller (1994b) suggested

that this instrument is useful for descriptive oral health surveys of general populations of

Canadians. The performance of this instrument was evaluated in the United Kingdom in an

elderly population aged 60 to 65 years-old and was found to be reliable in terms of internal

consistency and test-retest reliability and demonstrated satisfactory construct and concurrent

validity (Tickle et al, 1997a).

Comments:

One of the advantages of this instrument is that it allows for the exploration of the links between

distinct dimensions of health. Because it is a battery, an overall score cannot be computed;

hence, it cannot relate different outcomes to a common measurement scale.

8S

SWIed pue

jeioUab +0 p

abe

ul UyJBey

j20 sood

sueipeuey

pey payesolajep

OUM Buyjemp

asouy

‘ued Bursneo

~AWUNWWWOS

auljeseq ye

SUON!PUOD Jo

jes0 O1L00ds

uolejndod

e

e@aey 0}

peseodde

ul poliad

1e0A

PeAoJduui OUM

SSOU |] -

-38JU} OU}

‘pouad seeA-901u}

@y} JOAO

PSAOIC UI

J@AO Pes1iINIDO

yeui snes

pey %G’°O]

pue awes

uyeey jelo

8} paurewies

pey pealeoiod

%G'89 ‘payeioejep

(Buyjeip y61p

~}98 UI

2661 pey

Uyeey eso

W98uy AeAins

wopuey) |

sabueyo ey}

“O1AO yor

yeu} peyodai

%G'0Z -

U}TeeH MOIAJO}U]

epeued jeulpnyibu07y

+0S LL9=N

aulwexes O| |

pue Ja@x4907]

"JSOj YOO}

OY} Jo

uoIpuoD ay}

uodn “yyeay

je1o

Bulpusdep eanebou

Jo uoydeaoied

JO aAiisod

oq Aew

ayy

snjejs yyeey

UO Sso|

jo aBueyo

ou}

YjOo} jo

yoRdiui ay

~ 0}

Sajqeiien

"snjeys yjjeay

jeso (Guyjelp

yOip jeoiuyjo

ul eBueYo

paajeosed wopuey)

pajyogjes jo

-}]8S YM

payeloosse sjoelqns

uoHNgijUuoCS

Se|qeueA jearuljo

AjuO Aegains

aye]u9q eu}

9/661

GU} SI

SSO] YIOO|

- uyeeH

MBIAJO}U] epeuen

jeulpny6uo07 +0S

669=N ayewijse

OL "19990"]

“JOAO pue

JA g|

pebe

uoyeindod

ype

ue jo

"SJOPJOSIP snyejs

yey

jeso Aq

pasiwojdiuoo (Buyjdwes

| jeso

peyode

a190M (+91)

sebe {je

Wwopues Ajeanoelqns

qr6sl

jo syjnpe

jo Buleg

jem A@Ains

(yew) jeuoyoes

abejs om)

au} “JOININ

pue yyjeay

[eio SUL

uHesH payejduo0o-}}aS

epeuey -SSOID

+8L €SS=N

aquosep of |

pue

1984907

ISHOS

jo uoIsnjaUueD

uoneojddy | uoljeisiuiupy

uOHeI07} uBisep

Apnis aby

ejdwues uly

souny

ISHOS

Buisn paysiqnd sisded

jo Arewiung

"LL'Z aiqey

6S

"SUOITIPUOD [BJO

JO spaye

feloosoyoAsd ay}

BuluiWeyep

ul SWU9|qOJd

jeuoiouny pue

uyed 0}

eoueLodtul Alepuooes

Jo ae

saouejsuinodio OIWOUODe

-O100S ,S|@NPIAIpU!

- ‘uoyeindod

paaudep

ay} Wo jOedu

sayeos6 Ajueopubis

Jo 9q

0} payode.

SeM ojeosqns

jeloosyoAsd Quo

puke

ejeosqns jeuojyoun)

QUO AjUO

‘JOA@MO} -

YN

*SUOINPUGD Je10

Aq ul

uojejndod

Aem awos

ul peweduil

juenye

Sem 9jij

Jo Ayjenb

pue peaudap

JU} Jey]

peyode. jo

snjeys

juenyye pue

peaudep uyeey

jeso

yoee woud}

SJeNplAIpul (ej\dwes

| aajoelqns

eu}

Jo suojjodoud

ASAins (EIN)

wop6uly jeuoljoes

wopuey) |

eledwoo pue

a/66L

feyueysqns -

UESH

pa}9)dwod-JaS peyiun

~ssolD 9-09

CVE=N | Sinseew

oO] - |

Je je

[SHoIL

“AyIpiea qU@.LINDUOD

pue yonujsuoo

Asojoejsies payesjsuowadq

-

06°0-08'0 wopBbuly

reudje s,yoequoly

- peyup

ou}

8'0- 70

:eddey Ul

ISHOS §0

66'0-8G'0 ‘}Ua!OLjeoo

(ajdwes |

sauewojed

UONE{AU09 S,UOSIea|

anssl Jeo

(ew) wopBuly

WeUOIOaS wopuey)

ou} 2/66)

‘Jsajer-]S9 | - |

[Hojopouje| peya|dwioo-}aS

powun “SsOlD

9-09 CVSEN

ayenyjeas OL |

“j@

Je jexolL

‘punoibyoeq

pebejueapesip JOU WO.

09

“swojdwAs

yons

jo yoedult

ay} pue

swojdwis

|e10 pessesse-}jas

jo Bunodes

say} Ui

pue|Guz \seo

-UINOS ul

JUSPISes

(a4 M

pue UBSQqHed

xoeI1q ‘ asouiyy

‘uelpul) sdnoi6

ojuyye ino

WO} SJENPIAIpUI

Buowe

swioj}dwAs

yons jo

joedull

ay] pue

swoydwAs

YN ey}

ul sdnoi6

(Buydwes je1o

passesse

oluyje sno}

Guowe

Agains wop6ury

jeuoljoes YJOMION)

“J[9S JO

[OAR] £002

JSIXO SSOUSIOYIC)

- u}eeH

paje|dwi0a-jjas peyun

-SSO1D +GP

99e=N QU}

SUILUIA]EP O] |

‘Je je

UOMEN

61

2.1.4.4) Oral Impacts on Daily Performances (OIDP)

Purpose:

The OIDP was developed in the United Kingdom by Adulyonan and Shieham (1997). The

purpose of this measure was to assess serious oral impacts on a person’s ability to perform daily

activities.

Development Process.

Conceptual framework: This measure was developed based on the WHO model of disease and

its consequences amended by Locker (1988) for use in dentistry. However, a modification was

made by dividing the consequences into three levels (Figure 2.6): Level 1 refers to oral status,

including oral impairments, which are measured by clinical indices. The second level is called

“the intermediate impacts”; these include pain, discomfort, functional limitation and

dissatisfaction with appearance. The third level, “the ultimate impacts”, represent impacts on the

ability to perform daily activities, which consists of physical, psychological and social

performances. According to this model, any of the dimensions on the second level may impact

on the performance ability. The OIDP measures the third level of the consequences of disease.

Item generation: The items were generated from a literature review. Nine items were derived,

which covers the physical, psychological and social performances.

62

Level 1

Level 2 - »| Functional p»| Dissatisfaction with Intermediate impacts limitation P appearance

Level 3 Impacts on daily performance

Ultimate impacts Physical Psychological Social

Figure 2.6. Conceptual model used for the development of OIDP Source: Adulyanon S, Shieham A. Oral impacts on daily performances. In: Slade GD (Ed). Measuring oral

health and quality of life. Chapel Hill: University of North Carolina, Dental Ecology 1997. page 153

Item reduction: Analyses of internal consistency and item-total score correlations were

conducted. Based on these analyses one item was found to be redundant and was then excluded

from the final instrument. The final version of the OIDP contains eight items.

Description:

The OIDP contains eight items with a multilevel question (example shown in Table 2.12).

Respondents are asked to report on experiences over a six-month reference period by recording

the frequency or duration and severity of impact. The score for the frequency is shown in Table

2.13. For severity, the respondents are asked to indicate the level of severity on a Likert scale

from 0 to 5, where 0 indicates no effect and 5 indicates a very severe effect.

The final score for the OIDP is calculated using the formula in 2.3. The OIDP was designed for

administration by interview.

63

Table 2.12. Example of the questions in OIDP.

1) In the past six months, have problems with your mouth, teeth or denture caused you any of difficulty in ....(performance’)....?

0 No (Go to next activity) Yes (Go to 2a)

Performance* a) Eating and enjoying food

b) Speaking and pronouncing clearly c) Cleaning teeth d) Sleeping and relaxing e) Smiling, laughing and showing teeth without embarrassment f) Maintain usual emotional state without being irritable g) Carrying out major work or social role

h) Enjoying contact with people

2a) If yes, how often? ~ Less than once a month (Go to 2b) 2 Once or twice a month

3 Once or twice a week 4 3—4 times a week

5 Every or nearly everyday (5+ times/week) 2b) If less than once a month, around how many days in total?

1 Up to 5 days in total 2 Up to 15 days in total

3 Up to 30 days in total 4 Up to 3 months in total

5 More than 3 months in total

3) Using a scale from 0 to 5, where 0 is no effect and 5 is a very severe effect, which number would you say reflects what the difficulty in ...(9erformance).....nad on your daily life?

Table 2.13. Criteria for frequency score.

Category Score Never affected in past 6 months

Less than once a month, or a spell of up to 5 days in total

Once or twice a month, or a spell of up to 5 days in total

Once or twice a week, or spell of up to 30 days in total 3-4 times a week, or a spell of up to 3 months in total

Every or nearly every day, or a spell of over 3 months in total ObwWN-A

OIDP score = > (Frequency score [x Severity score |) X 100.0... eee eee (2.3) 200

| = performance

# = Maximum possible score.

64

Properties tested:

This instrument was initially tested among a 35-44 year-old adult population in Thailand. Test-

retest reliability, using Kappa, of the items ranged form 0.91 to 1.0. Cronbach‘s alpha was 0.69.

The construct validity and criterion test validity studies showed satisfactory results.

Population/Setting:

Table 2.14 shows a summary of papers published using the OIDP. It was developed in the

United Kingdom and was initially tested in Thailand among middle age adults for the Thai

version. It has also been used to measure the oral health status and to estimate treatment needs

among the elderly population in Thailand (Srisilapanan and Sheiham, 2001a; Srisilapanan and

Sheiham, 2001b). Subsequently, the OIDP was tested in Great Britain (English version) and

Greece (Greek version) among elderly populations (Tsakos et al, 2001a).

Comments:

The OIDP was developed to measure the most serious oral health impacts. This results in a

severe “floor effect”. As such, the use of this measure as an evaluative measure is impossible.

The design of the questionnaire only allows the OIDP to be administered through interviews,

and this will increase the cost of administration. Robinson et al (2001) recommended that one

way of reducing the cost and time is to conduct interviews over the telephone. However,

telephone interviews will not be appropriate in countries where a telephone is not owned by all

the households.

$9

( Buowe

Bujee uo

Buljdwes Jejnoyed

ul pue

*spooj jo

sodA} wopuel |

ot Ajlep

Jo Ayyenb

UOLULWWOD jBJOARS

Jes 26

eu} UO

peYy Adu}

0} Ayyiqe

ouy ‘sejnoned

ys (NW)

S}OBYS OU}

PUE ul

pue ‘ejdoed

Japjo jo

Aue

spedull poyeja.

yl Jo

Ayyenb ay}

sjoaye puadepul

yyeey jeio

Ayuenbe.y Ajsiey

a}doad wop6uly

jeuoloes BulAry

Jo aougjeaaid |

100Z 12

18

Japlo jo

snyejs jeso

ayy | Aesuns

yyeoH MAIAJ8}U|

pe1un -SSOID

+69 |

€G/=N-

au} ssasse

O] WeUleUS

‘aidoad

JapjO sJe]USpP

Jo

‘ajdoed jeu]

Jepjo Jo

Spaeu JUusU}e9l)

sp9au JUSWIeS]

je}Uuep oeujsoid

ssesse

oneujsoid ssesse

0} o]

sayoeoidde

yoeosdde sajpewsou

pue jeyUapoloo0s

pue

jejJuspolsos eB UseMjeq

Ajue (ajdwes

eAanewou

Bursn qLo00z

punoj iam

peau puadepu!

jue pseu

pessesse |

‘weuleyus

jUSW}eeL] peyewise

jeuUOH}oeS Burary

| jweauog)

Ul SSOUSIOLIP

pue ue

Jo saousiayip

oBe7] |

Aeains yyeoH

MOIAJOU] pueyeul

-SSOlD 2-09

67S=N ayessnyioj

| uedeysus

*sleyl

"SJ8AS JOU

Japjo ul

ayy} Apiep

jg UoWWOD

AaAI}e/eJ Aju

(ajdwes 9}

uO sjoRedul

aJsam ajdoed

Japjo pusedapul

jue peyejei

jes0 BLOOZ

Jo op

go Ayyenb

pejooye jeuaijoes

BulAl] |

jUeAUdD) jo

aouajeaaid ‘je

jo ue

yeu} sjoeduu!

je10 | AeAins

yyeeH MA@IAIB}U]

pueyeu |

-Sssoln Z-09

ZOZ=N au}

ssasse oj

| uedelisus

‘uoeindod

jo spsou

jeyuep ajewlse

0} sjuewuebpn!

‘uoyeindod

SAIEWOU

Jesiaaiun AjUO

aseesip jejuep

Bursn jo

Bulwoo}oys MO}

& Ul

Seinjeay

Ou} SESSONS

SI] -

(| peyejes

trey} pue

‘uorye(ndod Buydwes

seoUeWJOLed

SaLI2@O MO|

UL UBAS

wopuels |

Ajfep uo

syoedul

uBiy sem

joedui yyeey

jeuoyoes Jaysnjg)

| eso

yo BouepIOU!

| 9G6)

‘1212

jelo Jo

soUEpIoUl SY L- |

AeAINS UYyeSs}

MBIAJO}U] pueyeuL

-SsOl9 pr-Se

LOS=N einseaw

of} | uouesnpy

uoHensiulupy

uBisop

uoisnjauce5 uojeonddy

dao

uoHeI07

Apnig aby

ejdwes

wy

souny

daio Suisn

peysijqnd sueded

jo Arewwng

‘pL'z eqeL

99

“ule

pue @98015

Ul GUAT

snye}s jeso

yeotuyo Je}Wis

Jo ajdoed

Japjo usemjeq

peou JusWjee}

‘ajdoed ‘Ayapye

panleoied Japjo

ul sjoedtuy

20

jo Buialy

que |

= pue ayy

jo Ayyjenb

uoydaosed ul

souenyu! puedepuj

poayejal-yyeoy

jeunyjno juspuedepuy-

j2JO Ul

SBOUBIOJIP

‘sjuedojyed ysnug

LS9=N jesnyno

QU] UeY}

YeeIH ou}

1908015) |

-SSOJO JUROIUBIS

Jo} sayBiy

sem syoeduui

9080!9 jeudljoes

€G/=N |

sue eeu,

JeuloyM |

= gLOOZw ‘Ie

je10 Jo

gougjeaeid oys-

| Aeauns

yyeeH pue

ulejug -SSOID

+69 iuleyg

suiex9

oj | ja

soyes]

“‘sjuayed ‘saunjuep

ainjuap JBUOHUSAUOS

ajejdwos

ueu} ayy}

Ajlep uo

yoedui jJEUOHUSAUOD

$so| peouaiedxe

pue UJI

uosedwioo

AYNOLWJEP SSB}

U}IAA SLE}!

Ul SAun}USPsOAO

pooy Jo

eBues apm

OvV=N Jejngqipuew

2 jee

pjnoo ‘seinjuap

‘eunjuap | peziigqe}s-jue|duui

HOU} JO

YO}WOO peuo

jeyuep wo}

ay} YIM

peysyes alow

nusAUOD |

uleB yyeey

e sem

OJOM SBINJUSPJBAO

wop6uly jeuoyjoes

€p=N JOY)

JOUJOUM £00z

‘je

juesdul uM

syuened |

Aesains yyesH

MOIALO}U] peyun

-SSOID +9

juejduy ssesse

O| je

sejoyy

(ajdwes wopue.)

pezije UONNPSU]

“uone;ndod

cOc=N

- Ajs@p19

YN OU}

L9

“AUpyeA JONYSUOD

pooh pej}eujsuoweg

0'L 0}

09'0=eddey-

28°0=0U s u

ewleeds-

wuBees

so

JSOJOI-]S9 | seq

jo Ayissanun

L9 Q=U0/}2I}SIUIWUPe ay}

Buipuaye

puc” Ayipyjea

yonujsuos :AyIpyeA

synpe uejuezue |

£002

€9 O=uOHeySIUlUpe

‘yede sAep

Gj, -jse}e.-]se | -

0} dco

"WUOJSY

1S|- AQUS}SISUOD

feUJS}UI- jo

Ayqeaydde

pue

seydye $,yoequolg

AUPGeNOY uonejsues}

ON |

peucyUeW

JON OU}

QUILUEXO OL

nyeseyy

“Aupiien ul

Bulay soyenpeibisod

808019) PUE

poo6 payesjsuowep

oa “pue

solwepeoe Uleyg

Jealgd jo

SUOISIOA UO

YeosD Jo

joued e

Aq eo

spon

69'o=usnua- -Aypen

pessesse pue

uoljeysuey @

J0 salyedoud

LL Q=824-

| JONISUOD

PU UOEPID

‘JUa}UOO ‘soe

:AUIPIIEA, PUeM}eQ-PAeMIO3

ajeawoysAsd 21002

seydje s,yoequolg

AQuUalsISUOd jeUIE} Uj

:AWGetey yesig

oj ys!|Bug

| pauonusw

jON au}

eulMexe Of |

[2 }e

OVeS]

yIomoues{

pessessy eoly

uo}esuel |

uojejdepy

wiy iouny

uoisnjouc9

sebenbue] pue

sainjyjno Jeyjo

0} gqiO

94} JO

UOHe}depe pue

uoHejsuell “GL'z

a1qe 1

68

2.1.5) Measure chosen to be adapted

After extensive review of these four measures, we decided to adapt the Oral Health Impact

Profile (OHIP) since it meets the aim of this project, that is, to develop a generic oral health

related-quality of life measure, which is able to describe OHRQoL and also to discriminate

across groups in the population. OHIP has been used in both health surveys and clinical trials.

The psychometric properties were the most widely tested including responsiveness. Moreover,

the OHIP was developed based on a coherent theoretical framework and the items were selected

from qualitative interviews with the target population. As such, it increases the likelihood that

the OHIP is a measure of oral health-related quality of life rather than oral health status. The

OHIP is the most comprehensive measure currently available that covers a wide range of quality

of life outcomes. In addition, the format of OHIP enables it to be administered both by interview

and self-completed questionnaire.

69

2.2) CROSS-CULTURAL ADAPTATION OF OHRQoL

The demand for developing cross-cultural health-related quality of life measurement is

increasing. This rising demand is due in part to the development of global perspectives on

health and health care (Herdman et al, 1997; Phillips et al, 1994) and the necessity to include

minority groups in clinical research (Guarnaccia, 1996; Cella et al 1996). Since culture has a

significant impact on an individual’s life, the measurement of health cannot be culturally free

(Guarnaccia, 1996). Therefore, instruments that have been developed in one culture cannot be

used in another culture without adaptation. Thus, an appropriate methodology is needed so that

the results will be valid. This section reviews the issues surrounding the process of cross-

cultural adaptation of health status questionnaires.

2.2.1) Culture and health

It has long been recognized that how people lead their life is very much influenced by the

culture to which they belong and the social environment in which they live. Culture is “a set of

values, assumptions, perceptions and conventions, based on a shared history and language,

which enable members of group or community to function together.” (Henley and Schott, 1999).

Thus, culture affects every aspect of life, how we think and behave, and the judgements and

decisions we make. In other words, culture will influence how we look at the world, define and

interpret it (Henley and Schott, 1999; Helman, 1990). However, cultures are not static. They

change in response to new situations and pressures. For example, with the effect of urbanization

in Malaysia, Malaysian society has become more individualistic.

Culture influences how people perceive health and illness and also the ways health

problems are expressed. Within every culture there is a range of norms, values and assumptions

relative to health. Thus, what people in one culture consider as healthy may not be considered

70

healthy in another culture. For example, obesity is viewed as unhealthy in some, predominantly

western cultures, but healthy in others. A cross-cultural comparison study by Buck et al (1999)

on the impact of epilepsy on the quality of life among the European countries found that those

living in Spain and the Netherlands were less likely to experience an impact on their lives and

less likely to report feeling stigmatized due to epilepsy. In fact, they had higher levels of

physical and mental well-being. One of the explanations for this observation is how people view

epilepsy in different countries. Perhaps, epilepsy is viewed less negatively by the public in Spain

and the Netherlands than in some other countries, which in turn means that those who have

epilepsy are less likely to feel “socially different” and hence perceive low levels of impact from

their condition.

The way in which health problems are verbally expressed also differs from culture to

culture, pain expression for instance. In some cultures or families people are brought up not to

make a fuss, to be stoical and to show strength and fortitude and to be “good patient”. In other

families or cultures, people are more vocal and demonstrative. As such, it may be normal for

them to moan and cry out when in pain (Henley and Schott, 1999). Thus, it is wrong to assume

that all cultures conceptualize and express health and quality of life in the same ways.

2.2.2) Approach in HRQoL cross-cultural research.

Cross-cultural adaptation is oriented towards measuring a similar phenomenon in different

cultures. Basically, it is a process of producing an equivalent measurement developed in one

setting/culture to be used in another setting/culture (Beaton, 2000; Guillemin et al, 1993). Thus,

the notion of equivalence becomes an important aspect of the cross-cultural adaptation process.

An extensive review on the matter of equivalence in health-related quality of life cross-cultural

adaptation by Herdman et al (1997) found that there is a considerable amount of confusion

71

surrounding the ways in which equivalency is defined and established. In order to resolve these

confusions, Herdman et al, (1997) adopted a cross-cultural psychology orientation to health-

related quality of life research.

Basically, there are three approaches to the issue of culture and health: absolutism,

universalism and relativism. Absolutism assumes that “culture” has only a minimal impact on

the construct being measured and how it is operationalized. Thus, there will be only small

variations across cultures. Therefore, standard instruments can be used to measure the same

construct in different cultures. However, this can result in what is known as an “imposed etic”

whereby a construct is designed and operationalized in one culture is imposed directly onto

another culture. On the other extreme is relativism, where it is assumed that the role of culture in

behavioural variation is substantial and therefore standard measures across cultures cannot be

used. It requires the development of measures specific to each culture. In between these two

extremes lies the universalist approach. This approach is more open to the suggestion that

culture will have a significant impact on the way concepts are expressed across cultures. It does

not make any prior assumption that the construct will be the same across cultures. As such, it

allows a measure to be adapted. Thus, Herdman et al recommended that researchers adopt such

approach.

2.2.3) The concept of equivalence

There are two related concepts in cross-cultural research: equivalence and bias (van de Vijver

and Leung, 1997). Although they are related, they refer to different aspects of cross-cultural

research: cross-cultural adaptation and cross-cultural comparison (Figure 2.7). Cross-cultural

adaptation is referred to as the process of producing an equivalent instrument for measuring a

similar phenomenon in different cultures. On the other hand, cross-cultural comparison refers

72

to the comparative study of a phenomenon across cultures. Thus, the concept of equivalence is

associated at the cross-cultural adaptation in which scores are obtained in different cultural

groups. Bias indicates the presence of factors that challenge the validity of cross-cultural

comparisons. Hence, cross-cultural adaptation is a prerequisite for cross-cultural comparison.

Since the present study is concerned with the cross-cultural adaptation process, the concept of

equivalence will be the focus of this review.

Cross-cultural research |

WV

Cross-cultural adaptation Cross-cultural comparison

Prerequisite

equivalence Bias

Figure 2.7. Type of cross-cultural research

There are a number of guidelines for the cross-cultural adaptation of health status

questionnaires (Beaton et al, 2000; Herdman et al, 1998; Touw-Otten et al, 1996; Phillips et al,

1994; Flaherty et al, 1988). However, there has been little standardization in terms of the

definitions of different types of equivalence and how they should be tested (Herdman et al,

1998). For example, Table 2.16 indicates types of equivalence and their definitions as specified

by Hui and Triandis (1985), Flaherty et al (1988) and Guillemin et al (1993). Recognizing that

there is a need to develop a standardized framework, Herdman et al (1998) developed a model

known as the “model of equivalence” on which the discussion of the concept of equivalence

will be based. We have chosen Herman’s model as a framework for the present study because

73

this model addresses every aspect of the questionnaire and provides a clear framework on how

to achieve equivalence.

Table 2.16. Different concepts of equivalence

Hui and Triandis, 1985*

1) Functional equivalence: Pertains to whether or not the items ina

translated version of a

scale have a meaning similar to the source version.

2) Operational equivalence: Pertains to the comparability of procedures (e.g self versus interview) used in obtaining the information needed, including comparison of

psychometric

performance. 3) Scale equivalence:

Pertains to the extent to which individuals in different cultural groups respond to similar items

in similar ways. 4) Scalar/Metric equivalence:

Concerns whether or not

a measure orders

individuals along a continuum in a

comparable way across language and cultural groups.

Flaherty et al. 1988

1) Content equivalence: Pertains to whether or not the items of the

instrument are relevant

to the phenomena being studied in each culture.

2) Semantic equivalence: Concerned with whether

the meaning of each item is the same in each culture after translation

into the language and

idiom of each culture.

3) Technical equivalence: Pertains to the method

of obtaining information (e.g self-versus interview).

4) Criterion equivalence: Concerned with the

normative interpretation of the variable between

the two cultures.

5) Conceptual equivalence: Concerned with whether

or not the instrument is

measuring the same

construct in each culture.

Guiliemin et al. 1993

1) Semantic equivalence: Concerned with achieving the same meaning of the words comprising eacn

item. 2) Idiomatic equivalence:

Concerned with finding an equivalent expression of

idioms or colloquial expressions.

3) Experiential equivalence: Concerned with the relevancy of the items to the target culture.

4) Conceptual equivalence: refers to the validity of the concept expiored and the events experienced by

people in the target culture, since items might be equivalent in semantic meaning but not conceptually equivalent.

* cited in Bullinger et al. 1993

This model was developed based on the universalist approach. It includes six types of

equivalence; that is conceptual, item, semantic, operational, measurement and functional

equivalence. Herdman et al (1998) believe that this model provides a sufficiently complete

74

framework for the examination and achievement of equivalence. The definitions of each type of

equivalence are shown in Table 2.17 and each of them will be discussed in turn.

Tabie 2.17. Definition of equivalence (Herdman et ai, 1998)

Type of Definition equivalence

Conceptual Ways in which different populations conceptualize health and quality of life (QoL) and the values they place on different domains of health and QoL.

item Concerns the way in which domains are sampled. Item equivalence exists

when items estimate the same parameters on the latent trait being measured and when they are equally relevant and acceptable in both cultures.

Semantic Concerned with the transfer of meaning across languages.

Operational Refers to the possibility of using a similar questionnaire format, instructions, mode of administration, and measurement method (response

format).

Measurement Ensuring that different language versions of the same instrument achieve acceptable levels in terms of their psychometric properties — reliability, responsiveness, and validity.

Functional The extent to which an instrument does what it is supposed to do equally well in two or more cultures.

Conceptual equivalence:

According to the Herdman et al (1998) model, the most important aspect of equivalence is

conceptual equivalence. Conceptual equivalence refers to ways in which different populations

conceptualize health and quality of life (QoL) and the values that they place on different

domains of health and QoL. Because no prior assumption is made on how people from different

cultures conceptualize health and quality of life, this must be established before one could think

of adapting the instrument. If conceptual equivalence is not established, the result would be

biased. For example, Aracena et al, (1994) conducted focus group to determine if the meaning

75

of child abuse was similar between the American and the Spanish population. She found that

behaviors, which would be considered abusive in North America, were seen as part of the

continuum of normal child rearing practices in Chile. Although the child abuse scale can be

translated from a linguistic point of view, the results would be biased.

Item equivalence:

Item equivalence concerns the way in which domains are sampled. It exists when items

estimate the same parameters on the latent trait being measured and when they are equally

relevant and acceptable in both cultures. Frequently, these items were selected based on the

source culture and may not be relevant to the target culture. Therefore, the relevancy of each

item in the instrument needs to be examined to ensure that it is “culturally appropriate” and

“culturally sensitive” to the target population. “Culturally appropriate” means that the items

have to be relevant in the target culture. If the behavior is not practiced in the population or it

has become a norm to the target culture, the items could be considered as not relevant to the

target culture. For example, Hunt et al (1986) in their work in adapting the Nottingham Health

Profile (NHP) to be used in Egypt found that the item “I find it hard to stand for long” was

inappropriate since most tasks are carried out sitting on the ground and very little actual

standing is done in the rural Arabic world. ‘Cultural sensitivity” is concerned with items that

may be offensive or taboo to the members of the target culture. For example, Hunt et al (1986)

observed that words relating to what are perceived as “intimate issues” caused offence to the

people of middle-eastern cultures.

76

Semantic equivalence:

Semantic equivalence is concerned with the transfer of meaning across languages. This means

that the words from the target language have to be denotatively and connotatively the same as

with the source language. Therefore, items should not be translated literally; rather they should

be translated in a dynamic way so that the conceptual meaning of items is retained. In addition,

one has to ensure that the level of language used is appropriate to the language skills of the

target population. Wong et al (2002b), in the process of adapting the GOHAI to the Chinese

elderly population in Hong Kong, found that several items could not be translated directly due

to a lack of understanding among the elderly population. For example, the question “How often

did you limit contacts with people because of the condition of your teeth or dentures?” The

meaning of “limit contacts with people” was not understood by the elderly. In such cases there

was a need to modify the item without losing its general meaning.

Operational equivalence:

Operational equivalence refers to the possibility of using a similar questionnaire format,

instructions, mode of administration and response format in the target culture. How a

questionnaire is operationalized can affect the results obtained. For example, a study by Gilmer

et al (1995), which compared the functional characteristics of older adults, using the Iowa Self-

Assessment Inventory (SAD, of White rural elderly and elderly rural African Americans from

South Carolina and Louisiana, found that there was a significant difference between South

Carolina and Louisiana, which was not expected. This prompted the authors to investigate

further and found that there was a high illiteracy rate among the Louisiana sample, so that

many of the inventories were being read to the elderly rather than self-administered or sent by

mail, as was the case with the rural South Carolina sample (Gilmer et al, 1995). Slade et al

77

(1992) found that the OHIP score was lower when it was administered by interview compared

to a mail questionnaire. These examples illustrate that without uniformity in data collection

procedures, the validity of the data comparisons can be suspect.

Another important aspect, which many researchers ignore, is the suitability of the

response format used. For example, French et al (1998) found that younger children in

Australia had difficulty in using the response options of the Childhood Asthma Questionnaire

that had been developed in the United Kingdom. They concluded that the UK children were

more cognitively advanced than Australia children of the same age and may even be more

advanced in their reading skills since they had earlier formal education than the Australian

children. This demonstrates that the suitability of the response options cannot be assumed but

must be established.

Measurement equivalence:

Measurement equivalence aims to ensure that different language versions of the same

instrument achieve acceptable levels in terms of their psychometric properties — reliability,

responsiveness, and validity. Many researchers agreed that once the instrument has been

adapted, there is little assurance that the psychometric properties of the instrument remain

constant (Badia et al, 2000; Atchison et al, 1998; Streiner and Norman, 1995, Flaherty, 1988).

Therefore, the psychometric properties of the adapted instrument need to be re-evaluated using

data collected from members of the target culture.

Functional equivalence:

Functional equivalence is concerned with the extent to which an instrument does what it is

supposed to do equally well in two or more cultures. Since the Herdman et al (1998) model was

78

based on the universalist approach, this depends on two fundamental issues. Firstly, how health

and quality of life are defined and conceptualized in the target culture. Secondly, how

successful the adapted instrument is in measuring the trait which it is supposed to measure.

Therefore, according to this model, an instrument has achieved functional equivalence when all

other types of equivalence, which deal with these two fundamental questions, have been

achieved.

2.2.4) Technical process in cross-cultural adaptation.

The cross-cultural adaptation of an existing instrument involves two processes: translation and

cultural adaptation. Translation is a process of transferring individual words or sentences from

one language to another language, while cultural adaptation is a process of ensuring that the

measure is appropriate to the cultural context and lifestyle of the target population. However, in

some instances, cultural adaptation is needed even when the language remains the same,

because the culture or life experiences of populations speaking a given language can differ

(Geisinger, 1994; Guillemin et al, 1993). Both processes will be discussed in the following

section.

2.2.4.1) Translation:

To date, all of OHRQoL measures were developed in English speaking countries and in the

English language. Hence, in most instances, these measures need to be translated to other

languages. However, the translation of a HRQoL questionnaire is not a straightforward task.

Thus, several techniques have been developed over the years to ensure the quality of the

translation. Three methods are commonly used in translating existing HRQoL instruments: One-

way translation, translation by committee and forward and backward translations (a summary of

79

each method is shown in Table 2.18). Since the forward-backward translation technique is the

preferred method and strongly recommended (Kim and Lim, 1999), a detailed description of this

method will be described here.

Forward-Backward translation is an iterative process and it requires a minimum of two

bilingual independent translators. It is recommended that the translators should be linguistically

competent and they must be fully conversant with both languages as well as familiar with both

cultures (Carlson, 2000; Leplége and Verdier, 1995; Geisinger, 1994). It is also suggested that

the translators translate into their mother tongue (Beaton et al, 2000; Guillemin et al, 1993). The

forward-backward translation method involves four steps (Behling and Law, 2000):

1) A bilingual individual translates the source language instrument into the target

language.

2) A second bilingual individual translates this target language version into the original

language.

3) The original and back-translated versions are compared.

4) If there are substantial discrepancies between the two versions, then another target

language draft is prepared and modified to resolve the discrepancies.

This process continues until the two versions are identical or contain only minor differences.

The ability to compare the back-translated version to the original source language

version of the questionnaire is a major contributor in providing the researchers with an

objective indication of the semantic equivalence of the target version and pinpoints the nature

of specific problems with it (Behling and Law, 2000; Sechrest et al, 1972). For example when a

statement in English such as “I get tense before an examination.” was translated into Urdu and

back translated into English, it came out as “I get excited before an examination” (Sechrest et

al, 1972). This raised a question whether the words “tense” and “excited” are in fact the same.

80

Thus, this gave the researcher an indication that this translation needed some further

translation.

One of the drawbacks of this method, as identified by Geisinger (1994), is that when

translators knew that their work was going to be subjected to back translation, they would use

wording that ensured that a second translation would reproduce the original version rather than

a translation using optimal wording in the target language. However, Geisinger acknowledged

that:

“If used as an iterative procedure in the sequential process of translation, test

review and test retranslation, the method may indeed help guarantee similarity of meaning across languages.”

Leplége and Verdier, (1995) recommended three steps in order to obtain a high quality

forward translation. Firstly, the face validity of the translated questionnaire should always be

checked by a layperson because the bilingual translators tend to have higher educational status

than the average respondent and could potentially produce wording unsuited to the general

public. Secondly, the translators should be fully briefed on what they need to do. Finally, they

recommended that the translators should be fully involved in the task by asking them to

comment on their own translations; for example, to identify any difficulties encountered or how

satisfied they were with their translation.

According to Herdman (1998), there are three possibilities of the translation: 1) easy, 2)

difficult and 3) impossible to translate. The problematic items need to be given special attention

when the instrument is assessed by a sample of the target population. It may be also possible to

replace items, which have proven to be impossible to translate.

[8

Bulwuinsuos atu

"$10}}SuUed} OM]

JO asn

— ANsoD

‘UOISJAA [EUIBLIO

au} soNnpoida

pinom yeu}

Guipsom esn

pjnopy

“Bulyjos eayIWUW09

Ul JOUJOUL

QUO BZIOUS

O} yuRysey

aq Aew

10 spunos6yoeq

JOU] Ul

SAIPIE|!LUIS OF

ONP MAIAP|JOM

UOLWLUOD B aJeUs

Aeu Siaquaw seyww0y

"JUDLUNSUL Meu

By} Jo

ALIpIeA

pue Ajiqeljas

Mo} Ul

s}iNnsey ‘peulBo

au} jo

BujueaW aAlye}OUUOD

UB] JOYE

SAIJE}OUSP Su}

0} Spuodsa09

yeu} UO}e|sued}

e ui

synseu Ayensn

enbiuyos} siq yp

sebeyueapesiq/suoyeywry

"JUSLUNISU! OU}

JO JUdLUdO|eAep

OU} Ul

ajo Jofew

e Aejd

oO} sieyoleesa

jenBuljouow swe,

‘ssed AjIpee

JOU |[IM

SWI2} yuajeainba-uou

Aqajeym Jaq}

2 Se

soyesodo ‘aquaYy

PUB SUOISIOA

OM} 9}

UBAaMjaq UOSIEdLOD

SsMOTY

“uole|suel -yoeq

uey} Buiwinsuod

ou}

sse7q

ealsuedxaul pue

ajdwis

sobejueapy

"SOOUSIOIIP Joulw

ule}Uuod AjUO

JO jeoQUEp!

ae SUOISIOA

OM} AU}

[JUN SeNUUOD

ssedold SIU]

"sajouedasosip OU}

SJEUILUI[a O] SUOHeapou

Bululeyu0S pasedaid

sj yeip

aBenbue] ye6e}

JOYOUS U8]

‘SUOISJ9A OM]

BY} UBemjag

sslouedaJosip jeyueysqns

ae

alieau} |

(p “‘paredwioo

ase SUOISJOA

pejyejsuel-yoeq pue

jeulsuo oy,

(¢ ‘abenbue]

jeul6i0 ayy

oyu! uoIsJoA

aBenBug| jabie}

siuj seyejsuey

lenplaipu fenBurjiq

puosas y

(Z ‘abenBue|

je6ue} 9U}

OJU! JUSLUNI}sUI

OBenBue] soinos

dy} Saye[suey

jenpiaipul jenBuijig

V7 (1

:Sda]s INOL

S@AJOAU! YdIUM

SSB00id SAesa}]

UB Si]

“Suo}ejsuey juapusdepul

om} Jo

WNuUiW

e Seuinbay

‘UOHJE;SUBI] OY}

UO YOOUS

oO} SJuedxe

jo jaued

e Aq

majadl juspusdepuy

=e

‘UOdI}eIOqe|]O9 Ul

Jo Ajayesedes

Joupe Bupom

‘eyesue.) 0}

sjenplAlpul yenBurjiq

aso

JO Om)

SouInbayy =e

‘aBenbury| jaGse}

OU} OJU!

UOISISA [eUIBLIO

OU} 9}e|SUBJ]

0} yenplaipul

yen6uyig e

Buryse Ajuo

speyusy

SS390/d

uoHe|suew)-piemyoeg pue

puemio4

O}}!LULUOD Aq

uonejsues |

uoHejsuel AeM-euCO

poujoy sebejueapesip

pue sebejyueape

s}i pue

uoie|sues) Jo

poujopy *91'Z

eIGeL

82

2.2.4.2) Cultural adaptation:

It is recognized that if measures are to be used across cultures, their content must not only be

translated well linguistically, but also they must be adapted to the cultural context and lifestyle

of each culture. Therefore the ultimate aim is to achieve equivalency as discussed in the

previous section. Since the cultural equivalence of HRQoL measure cannot be assumed but must

be assessed, Herdman et al (1998) recommended a method of assessing each type of

equivalence as summarized in Table 2.19.

Table 2.19. Method of assessing equivalence

Type of equivalence Method of assessment

Conceptual e Locai Literature

e Consultation with expert interview the target population

item e Review literature

e Expert judgments

e Interview with target population

Semantic e Forward-backward translation

Check by lay panel

Operational e Pretest the questionnaire

Measurement e Reliability test e Validity test

e Test of responsiveness

® Factor analysis

Functional e Examine other equivalence

Methods for establishing conceptual equivalence. There are three ways to establish conceptual

equivalence as suggested by Herdman et al (1998): 1) Review of local literature, 2)

Consultations with experts, and 3) Interviews with members of the target population. The first

resort is to review related or similar literatures that have been conducted on the target

83

population, such as topics related to perceptions of health, well-being, illness and disease or

perhaps topics related to emotion. This may provide an initial overview of convergent or

divergent with the conceptual structure of existing instruments. Secondly, it is possible to

consult experts in the target culture. Herdman recommended that the experts consulted should

be broad, including for example anthropologists, medical sociologists, linguists, as well as

health professionals.

The third approach recommended is to involve the target population themselves, through

various techniques such as focus group discussion, participant observation or open-ended

unstructured interviews. The advantage of this approach is that the concepts can be gleaned

from the target population perspective, thus increasing the likelihood that their views are

captured. In most instances, the assessment of conceptual equivalence needs to apply all three

approaches.

The investigations of conceptual equivalence have four possible outcomes (Herdman et

al.,1998): 1) the domains in the original instrument are equally relevant and important to the

construct being measured in the target culture and the emphasis placed on different domains is

also appropriate in both cultures, 2) the domains in the original instrument are equally relevant

to the construct being measured in the target culture but the emphasis placed on the domains

varies between the two cultures, 3) some of the domains in the original instrument are not

relevant to the concept being measured in the target culture or domains which are relevant to the

target culture are not included in the original instrument and 4) the domains are totally different

to the construct being measured between the two cultures. In the case of outcome (2) it may be

possible to weight the domains in order to reflect their differing importance in the target culture.

For outcome (3), it might be possible to use the same domain but it only allows a partial

comparison of the construct across cultures. Adaptation cannot be done for the outcome (4). In

84

this case, perhaps adaptation of another instrument should be considered or a culture-specific

questionnaire should be developed.

Methods of assessing item equivalence. The assessment of item equivalence involves examining

the relevancy of the items to the target population and this can be done in different ways. Firstly,

one can review available documents on lifestyle patterns and habits, which can suggest that

certain items would be irrelevant. For example, the item “Do household chores (e.g., taking out

trash ect)” in the Childhood Health Assessment Questionnaire, the example given “take out

trash”, was not relevant in Argentina because trash is removed late each night by adults

(Moroldo et al, 1998). In the case where literature on the target culture is scant, consultations

with experts can be conducted to elicit their “expert judgments’ regarding the potential

relevance of items. However, in many instances, the most appropriate source of information is

members of the target population, who can be assessed through different methods.

There are four possible outcomes of investigating item equivalence (Herdman et al,

1998): 1) Items can be used in target population without modification (other than translation), 2)

items require minor modification, but may be used more or less in their original form, 3)

Replacement items must be used, 4) Neither existing nor replacement items can be used because

they deal with subjects which are considered offensive or taboo.

In the case of outcome (3), the substitution of the item will be largely based on the

researcher’s judgment. Replacement items can be derived from literature reviews, expert

opinions and most importantly from the target population themselves. In the case of the outcome

(4) it may be possible to omit the items.

85

Methods of achieving semantic equivalence. The possible methods of achieving semantic

equivalence have been discussed earlier in this section (see Translation, page 77).

Methods of investigating operational equivalence. The most common method used to

investigate operational equivalence is to test the instrument on a sample of the target population.

Other sources of information, such as statistics on literacy rates, can provide a good indication

of the possibility of successfully using written questionnaires.

There are three possible outcomes of assessing operational equivalence: 1) The same

methods of data collection can be used, 2) Some aspects of operationalization need to change

and 3) it is impossible to achieve operational equivalence.

Methods of establishing measurement equivalence. The psychometric properties of the adapted

instrument need to be established. These include reliability, validity, and responsiveness using

the methods described in a previous section (see the properties of HRQoL, page 25).

It is desirable to achieve very similar or equivalent results in some of these areas, notably

in terms of reliability, however in other cases it might be unrealistic to expect similar results.

For example, effect size and item weights.

Methods of establishing functional equivalence. According Herdman et al (1998), the functional

equivalence can be established by assessing the degree to which each type of equivalence

included in the model has been achieved.

CHAPTER 3

AIM AND OBJECTIVE

GOAL:

The main goal of this study was to develop an Oral Health-Related Quality of Life measure for

the Malaysian adult population (those who aged 18 and above) through the cross-cultural

adaptation of the Oral Health Impact Profile (Appendix A). In order to achieve these goals, a

two-phase study was designed. The aim and objectives for each phase are stated below.

PHASE 1:INSTRUMENT DEVELOPMENT AND ADAPTATION

AIM:

The aim of this phase was to translate the OHIP questionnaire into the Malay language and to

assess conceptual and item equivalence.

Objectives:

1. To translate the OHIP questionnaire into the Malay language.

2. To examine how two populations, Australia and Malaysia, conceptualize health and

quality of life in terms of how oral diseases/disorders affect individual daily life.

3. To examine whether items in the original questionnaire were relevant to the people in

Malaysia.

86

87

PHASE 2: EVALUATION OF THE INSTRUMENT

AIM

The aim of this phase was to establish the operational (format, instruction, response format, and

mode of administration), measurement and functional equivalence of the adapted instrument

[OHIP(M)].

Objectives:

1. To assess the comprehensibility and clarity of the OHIP(M), including format and

instructions.

2. To test the response format.

3. To assess two modes of administration (mail and interviews)

4. To examine the reliability of the OHIP(M).

5. To examine the validity of the OHIP(M).

6. To compare the psychometric properties of the OHIP(M) with the original questionnaire.

In addition to the above mentioned, this study also aimed:

1) To compare two methods of developing short forms of the OHIP(M).

2) To provide preliminary data on the impact of oral conditions on the quality of life of the

adult population in Malaysia.

3) To compare the impact of oral conditions on the quality of life between Malaysian and

Canadian population aged 50 and above.

CHAPTER 4

METHODOLOGY

This chapter outlines the methodology used in this study. This study was divided into two

phases and each phase was divided into stages in order to achieve the goal of the study (Table

4.1). The aim was stated and the method was described for each stage.

Table 4.1. Phase and stage of the study.

Sample Sample

size

Phase 1: /nstrument development and adaptation.

Stage 1- Translation and back-iranslation process

Stage 2- Qualitative interviews Purposive 36

Phase 2: Evaluation of the technical properties of the

instrument.

Stage 1: Pre-testing. Convenience 20

Stage 2: Testing of Response options. Convenience 171

Stage 3: Field testing. Random 426

4.1) PHASE 1: INSTRUMENT DEVELOPMENT AND ADAPTATION

This phase was divided into two stages (Table 4.1).

4.1.1) Stage 1: Translation and back-translation.

The aim of this stage was to produce a Malay version of the questionnaire, using the forward-

backward translation technique to establish semantic equivalence (Behling and Law, 2000;

Del Greco, 1987).

88

89

Translators:

The process of translation, for this study, involved six translators. Three translators were

involved in forward translation (Forward translators) and another three were involved in the

back translation (back translators). All translators were bilingual.

All three forward translators were Malay and Malay language was their mother tongue.

Two of them were lecturers in the Faculty of Languages and Linguistics, University Malaya

and both of them have a background in Malay studies. They were considered as the Malay

language expert. The third translator was a primary school teacher, where English was her

teaching subject.

Ideally, it is recommended that the translators translate into their mother tongue (Beaton

et al, 2000; Guillemin et al, 1993). However, it was not possible in this study to find someone

bilingual, with English as his/her mother tongue. Therefore, it was decided to appoint “English

expert” as back-translators. Of the three backward translators, two of them were considered as

English experts, since they were English lecturer at the Department of English Language,

University Malaya, and one was a primary English schoolteacher.

Process of translation:

Forward translation: The forward translators (FT) translated the original instrument, which is

in English, into the Malay language, the national language of Malaysia. Prior to the translation

process, the translators were briefed by the investigator on the aims of the instrument and the

population on which it would be used. This was to ensure that the words used were appropriate

to the target population (Herdman et al, 1998). The translators were also informed that the

focus of the translation was to retain the meaning of the items (semantic equivalence) rather

than a word-by-word translation. Therefore they were expected not to translate items literally.

90

They were also asked to note any items, which were not easy or impossible to translate. The

investigator combined all three translations, together with the original questionnaire (OQ) in

one table to facilitate comparison (Table 4.2). To avoid revealing the identity of the translators,

symbols (T1, T2, T3) were used instead of their names. The combined translation was given to

each of the FT before a reconciliation session was held. At the reconciliation session, all three

FT together with the investigator decided on the single Malay version (MQ).

Table 4.2. Example of combined file

Original T1 T2 T3 Solution

Q1) Have you Adakah anda Pernahkah anda | Pernahkah anda had difficulty mengalami mengalami mengalami chewing any masalah ketika kesukaran kesukaran untuk foods because of | mengunyah mengunyah mengunyah problems with makanan sebarang sebarang your teeth, disebabkan makanan makanan

mouth or masalah gigi, disebabkan disebabkan oleh dentures? mulut atau gigi masalah gigi, masalah

palsu? mulut atau gigi berkaitan paisu anda? dengan gigi,

mulut atau gigi paisu anda

Backward translation: Once a single forward translation was achieved, the backward

translators (BT) then back translated this Malay version into English. The original version was

not given to the back-translators to avoid bias in their back translation. The same process as in

the forward translation was carried out but this time was to come out with a single back

translated English version (BTQ).

Committee review: The MQ, BTQ and the OQ were compiled in a single table and brought to

committee meeting to ensure that the translations were satisfactory. A committee comprised of

the investigator and the six translators, reviewed the translation version in terms of:

91

1) The comprehensiveness of the translation. This includes the translation of the

introduction and instructions.

ii) Semantic equivalence. This was to ensure that items, which have been modified during

the translation process, retained their original meaning.

Once it was satisfactory, it was then sent to the evaluators to assess for any discrepancies.

Evaluator assessment: Evaluators were comprised of the developer of the original version (Dr

Gary Slade), the supervisor of this project (Dr David Locker), and one of the PhD committee

members (Dr Paul Allison). The backward English version and the original version were

emailed to the evaluators for them to make comments on the translation whether it had

achieved semantic equivalence. Items that were considered by the evaluators as problematic

were brought to the committee for discussion and amendments were made. The amendment

version was emailed again to the evaluators. This process continued until the evaluators were

satisfied with the result.

Both the Malay version and back-translated English version were used in the qualitative

interview phase of the study.

4.1.2) Stage 2: Qualitative interviews.

The aim of this step was to investigate the conceptual and item equivalence between the

source and target culture with respect to oral health, to ensure that the dimensions comprising

the OHIP are appropriate and comprehensive and to identify impacts/problems not covered by

the OHIP. In-depth interviews were conducted. How oral diseases/disorders affect one’s well

being and daily living in the target population was the focus of these interviews.

92

Respondents:

A convenience sample of the target population was drawn from patients attending clinics at the

Faculty of Dentistry, University of Malaya. The reason for selecting a convenience sample

rather than a random sample is because of the likely range of oral diseases with consequent

social impact (Slade and Spencer, 1994a). Patients were recruited from three departments,

namely Prosthetic, Orthodontics, and Primary Care. The purpose of recruiting patients from

different departments was to ensure that patients with a variety of oral conditions/disorders

were obtained. Patients were equally distributed in terms of ethnicity (Malay, Chinese, Indian)

and age group (Young adult, Middle adult, Old adult). These groups were known as the

“natient group”. Patients were added until the investigator was satisfied that the data were

complete and interviews were no longer eliciting additional information regarding perceptions

of oral health in the Malaysian culture. A total of 36 patients were interviewed.

Procedure:

Face to face interviews were conducted. The interview was conducted in Malay language or

English depending upon the language preference of the individual. The trained assistant (Dental

Surgery Assistance, Ms Pauline) approached the patient and briefly explained about the study.

Once the patient agreed to participate, he or she was brought to the interview room.

Each patient then was given a written statement of the aims and objectives of the study

and the nature of their participation. This information sheet was printed on the Faculty of

Dentistry, University of Malaya letterhead and was available in two languages, Malay and

English (Appendix B). Once patients understood the nature of their participation, a written

consent was obtained from the patient using the consent form (Appendix C). The consent form

was also printed on Faculty of Dentistry, University of Malaya letterhead and also available in

93

Malay and English. Each patient was informed that refusal to participate would not affect his or

her eligibility for services provided by the Faculty of Dentistry. They were told that they could

refuse to answer any questions they were asked and that they could withdraw at any time during

the course of the interview if they so wished. Patients were also informed that they could request

that the tape recorder not be used to record their responses at any time during the course of the

interview.

The first part of the interview was conducted to assess whether or not Malaysians

conceptualized oral health in a similar way as Australians. The interview was semi-structured.

That is, it consisted of open-ended questions and probes designed to elicit as much detail as

possible from the respondent concerning their perceptions of oral health (guideline of the

interview as shown in Appendix D). Patients were asked to describe why they had come to the

Faculty of Dentistry and the nature of the problem or problems that they wished to have treated.

The focus of the interview was on the ways in which their oral problems impact on daily life and

psychosocial well-being. Following the responses to these questions, they were asked about

areas of daily living that they had not mentioned spontaneously. Probes were again used to elicit

as much information as possible. The interview was ended by asking for demographic

information. The interview was tape recorded for the purpose of retrieving information.

After this component of the interview was completed the appropriate language version of

the OHIP was shown to each patient. Each statement was read to the patient, and the patient was

asked to comment on the relevance of the statement (i.e. does the problem it describes apply to

them) and whether or not the statement is clear or unclear in terms of its meaning. Patients’

comments concerning relevance and clarity were recorded. After the statements from each

OHIP domain were discussed, the individual was asked to describe any problems they may have

experienced not covered by the items in that domain. This information was also recorded.

94

Analysis:

Each of the interviews was transcribed exactly directly into a computer using Microsoft word

processor. The transcription was then checked against the tape for accuracy. The spelling was

also checked. Once the accuracy of the transcription was ensured, the coding and sorting process

began, using The Ethnograph version 5 software. Words or phrases that describe the functional

and psychosocial impacts of oral disorders were identified and a code was inserted into the text.

The codes were assigned according to the domains from the original questionnaire, as shown in

Table 4.3. The data were then sorted according to the domains. If additional domains were

identified, these would be given codes.

Table 4.3. Codes for Oral Health Impact Profile domains

Domain Code

Functional limitation FL1

Physical pain P1

Psychological discomfort P2

Physical disability D1

Psychological disability D2

Social disability D3

Handicap H1

Committee review:

Revision of the content of the questionnaire based on a list of impacts/problems obtained from

the qualitative interview and patients’ comments concerning the relevance and clarity of the

items was carried out. The committee that developed the translated versions of the questionnaire

reviewed these revisions. Discussion on the appropriate response format was given emphasis,

since the response format used in the original OHIP, which is the Likert scale with 5 points, may

encounter some difficulty when used with a Malaysian population, since Malaysians are modest

95

in giving their opinion. The time reference for the response scale was discussed. At this stage,

the committee worked only with the Malay version. This version was used in the second phase

of the research.

Questionnaire formatting:

Two sets of questionnaires were formed: Long (Appendix E) and short (Appendix F) versions.

The main reason for developing a short form was to provide an alternative, cheaper and less

time consuming way of collecting information related to oral health related quality of life. The

short version was created based on the items most frequently reported by the patients who were

interviewed at the qualitative interview stage. The two most commonly reported items from

each of the subscales were chosen to form the short version. The short version has 14 items.

96

4.2) PHASE 2: EVALUATION OF THE INSTRUMENT

This phase was divided into three stages (Table 4.1).

4.2.1) Stage 1: Pre-testing.

The aim of this stage was to check the wording, clarity, as well as comprehensibility of the

adapted questionnaire since it had undergone some modification (new items were added and

some items were modified). This phase was also used to check whether respondents were

interpreting the meaning of each item in an appropriate manner.

Subject:

A convenience sample of 20 respondents was selected from the patients attending clinics at the

Faculty of Dentistry, University of Malaya.

Procedure:

The investigator approached the patients while they were in the waiting area of the clinics. After

obtaining verbal agreement to participate, each patient was given a copy of the translated Malay

OHIP version to complete. After the patient had completed the questionnaire, he/she was asked

to comment on the questionnaire as a whole (Table 4.4). Then random questions were selected

and the interviewer probed by asking the patient a question: “What do you think is meant by this

question?’’, Patients were encouraged to elucidate their understanding of the items in an open-

ended manner. Patients’ comments were recorded verbatim.

97

Table 4.4, Questions asked to assess the questionnaire

Area tested Question asked

Format and instruction of the questionnaire:- | Did you have any difficulty to complete the questionnaire?

Were the instructions easy to follow?

Wording of questions:- Were there any questions that you found difficult to understand?

Time to complete:- How long did the questionnaire take to complete?

Analysis:

Data from the questionnaires were analyzed and any necessary changes to the wording of

items, instructions or response formats were carried out. The questionnaire would be re-tested

on a group of ten patients recruited from the same clinics if substantial changes were made to

the content or wording of the items comprising the questionnaire.

4.2.2) Stage 2: Testing of Response options.

The aim of this stage was to test whether the response format of the questionnaire is appropriate

for a Malaysian population. The response format is a five-point Likert frequency scale, with the

options ‘very often’, ‘fairly often’, ‘occasionally’, ‘hardly ever’ and ‘never’.

Subject:

A convenience sample of 200 patients attending dental clinics at the Faculty of Dentistry,

University Malaya and Bangsa dental clinic was selected. Patients were chosen to be the

participants at this stage because it was expected that they would be more likely to respond to

the more extreme options.

98

Procedure:

To test the response options, the short version of the questionnaire was used. Patients were

asked to complete the questionnaire (self-completed). The reason for using a short version rather

than long version is because the short version consists of the most common items reported,

therefore it was expected that the respondents would be more likely to use the full range of

options.

Analysis:

Questionnaire data were used to determine if the respondents used the full range of response

options by plotting the frequency distribution for each item.

4.2.3) Stage 3: Field Testing.

At this stage, the mode of administration, mail verses interview, was assessed and the

measurement properties, in terms of reliability and validity, were evaluated. The latter

determined whether the Malay language version of the OHIP had acceptable psychometric

properties. The results were then compared with similar assessments undertaken in Australia

and Canada in order to assess the measurement equivalence of the OHIP(M).

Study design:

A cross-sectional study design was used in this study, since the intention was to develop a

descriptive and discriminative measure, that is, measure that can be used to describe the impact

of oral conditions on the quality of life and are able to identify groups or individuals that are

most affected by their oral health at one point in time.

99

Subject:

Since a population-based sample was required, participants for this phase were a sub-sample of

the Malaysian National Oral Health Survey of Adults (NOHSA 2000). Permission to use the

NOHSA’s sub-sample and access to the database was obtained from the Dental Director,

Ministry of Health Malaysia. In the NOHSA, an adult was defined as those aged 15 and above.

However, for this study, only respondents aged 18 and above were included. The sample in the

NOHSA study was based on a subsample of Enumeration Blocks (EB) and Living Quarters

(LQ) from the fourth Quarter Labour Force Survey 1998. The country is divided into states and

each state was divided into urban and rural strata. A two-stage stratified random sampling was

used. The first stage involved a random selection of EBs from urban and rural strata within

each Malaysian state. The second stage was the selection of LQs randomly within these EBs.

All subjects age 15 years and above in all households in the selected LQs were interviewed and

clinically examined. The response rate was 66%.

For this study, the Selangor state was chosen as the sampling area. Selangor is one of

the 14 states in Malaysia. Administratively, it is divided into nine districts; Gombak, Klang,

Kuala Langat, Kuala Selangor, Petaling, Sabak Bernam, Sepang, Ulu Langat and Ulu Selangor.

Selangor was the most populous state in Malaysia with a total population of 4.2 million

(Department of Statistics Malaysia, 2001).

A sample of Selangor state was identified from the NOHSA database. Those who were

below 18 years by the year 2002, were removed from the sample. For the purpose of

convenience, Petaling district, was used as the sampling frame for the interview group and all

the other districts were used for the mail group. Respondents in each group were randomly

assigned either to complete the L-OHIP(M) or S-OHIP(M).

100

Sample Size:

The sample size was calculated based on differences in expected response rates between

interview and mail administered versions of the OHIP(M). If the former produces a response

rate of 60% and the latter 40%, then 77 subjects per group was required for a one-tailed test of

proportions with alpha at 0.05 and beta at 0.20 (Lwanga and Lemeshow, 1991). In order to

examine variations by age for the mail questionnaire where literacy rates are an issue, the

sample for the mail component was doubled. Test-retest reliability assessment with the null

ICC set at 0.6 requires 40 subjects in order to detect an ICC of 0.8 (Donner and Eliasziw,

1987). Sample sizes were increased to ensure that sufficient data was obtained for respondents

for other analytic purposes (See Table 4.5).

Table 4.5. Assigned groups and sample sizes required

Type of administration Age L-OHIP(M) S-OHIP(M)

Mail 18+ 150 150

interview 18+ 100 100

Procedure:

Two methods of administration, mail questionnaire and interviews, were employed and two

types of questionnaire, L-OHIP(M) and S-OHIP(M) were used.

Mail questionnaire: For those in the Mail group, a set of questionnaires together with an

introductory letter (Appendix G) and a prepaid stamped return envelope was sent to the

participants. In order to identify the respondents, a number, which was assigned to the

respondent, was stamped on the return envelope. A pen was enclosed to show an appreciation

for their participation.

101

Participants were asked to complete the questionnaire and return it to the sender using

the envelope enclosed. To maximize returns, the steps outlined by Dillman (1978) were

followed. Seven to ten days after the first mailing, a postcard (Appendix H) was sent to thank

those who had returned the questionnaire, and remind the others of the study’s importance. The

card also indicated to those who had mislaid the original where they can get another copy.

Three weeks later, a second letter was sent to those participants who did not return the

questionnaire (Appendix I). A second copy of the questionnaire and a return envelope was also

included. If there was no response from them after one month from the date the second

questionnaire was sent, a participant was regarded as a non-respondent.

Interviews: Household interviews were carried out for the interview group by the trained

interviewer: Investigator (Dr Saub) and three dental students (Mr Monaj, Mr Mahadzar and Mr

Huzaiman). The interviewers were trained by the investigator on how to conduct the interview.

They were given the following items to be carried with them when performing the interviews:

1) a set of location maps, 2) the name and address of the respondents, 3) laminated

identification letters (Appendix J), 4) the questionnaires, 5) pen (to be given to the respondents

as a token), and 6) the appointment card (Appendix K).

For security purposes, the police where the interviews were conducted were informed

(Appendix L). In order to obtain better response from the respondent as well as to make them

aware of the study, one month before the interviews started, an introductory letter (Appendix

M) regarding the study from the Dean of the dental faculty of University Malaya was sent to all

respondents. Subsequently, the interviewer went to each respondent’s house to interview him or

her. In the case where the respondent was not available, an appointment card (Appendix K) was

left at the house, asking him/her to contact the interviewer so that an appointment could be

102

made. However, if he/she did not get back to the interviewer within one week, a second visit to

the respondent’s house was made. The respondent was regarded as a non-respondent after two

visits were made. For convenience, the interviews were carried out area by area.

Re-administration: In order to carry out test-retest analysis, a second administration was carried

out. Fifteen days after the first administration of the questionnaire, the second administration

was carried out on a selected sub-sample. For the mail questionnaire, the sample was selected

based on those who had returned the questionnaire within two weeks from the time that the

questionnaire was first posted to them. For the interview questionnaire, after fourteen days

from the first interviews, the second interviews were made. All areas were revisited once. No

second visit was made. The same method of administration and type of questionnaire as at the

first administration were employed. At this stage one additional question regarding whether

their oral health had changed since the first administration was added.

Data entry:

Before data were entered into the computer, the source data were coded and then entered direct

into the computer using SPSS+ program. L-OHIP(M) and S-OHIP(M) data were entered

separately.

Data cleaning:

Before analysis was performed, data were first cleaned by two procedures: range checking and

contingency checking (Aday, 1996).

103

Range checking: this was done by running the frequency distribution for each item and verified

that only valid ranges of numbers were used. If any number in the coding was not valid, the

original questionnaire was used to determine the correct answer.

Contingency checking: Cross-tabulation was performed on the related questions to check for

accuracy in data entery. For example, if the question was meant for denture wearers, then those

who were not wearing dentures should only respond to the option “not applicable” for that

particular question. If another option was entered, then it was corrected to the “not applicable”

option.

Missing data:

Two procedures were applied in the case of missing data: total exclusion and mean item

imputation (Slade, 1997b):

1) If more than twenty percent of the data (nine and more items for L-OHIP(M)

and two or more items for S-OHIP(M)) were missing (blank entries or “don’t

know” responses), then it was excluded from the final analysis.

ii) In the case where less than 20 percent of the data were missing (blank entries or

“don’t know” responses), then the item was imputed by the mean of that

particular item.

Scoring:

Two methods of scoring were computed: Additive (ADD score) and Simple count (SC score)

(Allen and Locker, 1997). ADD scores were calculated by adding up the response codes and

SC score was calculated by summing the number of items reported as “very often” and “often”.

104

For L-OHIP(M), the ADD score could range from 0 to 180 and the SC score from 0 to 45. For

S-OHIP(M), the ADD score could range from 0 to 56 and the SC score from 0 to 14. A high

score indicated poor OHRQoL.

Analysis

Analysis was performed using the Stata version 7 program to deal with the complex sampling

design used in the study. The variances (standard deviations squared) of estimates obtained

from the complex designs generally differ from those based on a simple random sample (Aday,

1996). The stata statistical program can deal with this design effect.

Appropriate statistical analyses were performed. A p value was set at 0.05. The analyses

performed were:

1) To compare the two modes of administration (Interview vs mail) in terms of

response rate, completeness of data and the OHIP(M) score.

2) To assess the reliability of the measures.

3) To assess the validity of the measures.

4) To compare OHIP(M) with OHIP(Australia) and OHIP(Canada)

5) To compare the two types of S-OHIP(M) derived using two methods (frequency and

regression),

6) To describe the impact of oral conditions on the quality of life of the adult

population.

7) To compare the impact of oral conditions on the quality of life between Malaysian

and Canadian populations aged 50 and above.

105

1) To compare the two modes of administration (Interview vs mail)

The comparison of the two modes of administration was made by assessing three parameters:

response rates, completeness of data and OHIP(M) scores. A questionnaire was considered

incomplete if more than 20% of responses were left blank or marked don’t know: nine or more

items for the L-OHIP(M) and two or more items for S-OHIP(M) (Slade and Spencer, 1994a).

The following hypotheses were tested using the ra test or Mann-Whitney test.

Hypothesis:

Response Rates e The response rate for the L-OHIP(M) would be lower than response rate of S-

OHIP(M), both interview and mail.

e The response rate for mail questionnaire of the L-OHIP(M) would be lower than mail questionnaire S-OHIP(M).

Percentage of Incomplete data (%ID): e The %ID for the L-OHIP(M) would be higher than %ID for the S-OHIP(M), both

interview and mail.

e The %ID for mail questionnaire of the L-OHIP(M) would be higher than mail questionnaire S-OHIP(M).

OHIP score

e Ho: The score obtained through mail questionnaire would be no different with the score obtained through interviews, both L-OHIP(M) and S-OHIP(M).

2) Reliability

Two types of reliability were assessed: internal consistency and test-retest reliability.

Internal consistency.

Internal consistency was assessed using Cronbach’s reliability coefficient a based on the

responses from the first administration. Scores of 0.6 or more indicate good to excellent

reliability (Locker and Slade, 1993).

106

Test-retest reliability

To ensure that the questionnaire was reproducible in stable subjects, intraclass correlation co-

efficients were calculated using scores from the repeated administrations of the OHIP(M).

sample. The value of 0.7 was considered as an acceptable level (Aday, 1996).

3) Validity

To ensure that the instrament measures what it is supposed to measure, validity tests were

carried out. Data from the first administration of this part of the study was used to assess the

validity of the instrument. Because there was no “gold standard”, construct validity was

assessed. Two types of construct validity test were performed: Convergent validity and

discriminative validity.

Convergent Construct validity

Convergent validity describes how closely a measure is related to other measures of the same

construct to which it should be related. Correlation between the global rating of oral health and

OHIP(M) score using Spearman’s rank correlation test was performed to assess convergent

validity.

Discriminative Construct validity

Discriminative validity is related to how well the scale is able to distinguish between groups

with known differences. This was assessed by examining the association between:

i. The OHIP(M) score and satisfaction with oral health.

ii. The OHIP(M) score and self-perceived need for dental treatment.

iti. The OHIP(M) score and dental status (Dentate/Edentulous).

107

The following hypotheses were tested using the Mann-Whitney test for comparing two groups

and the Kruskal-Wallis test for comparing more than two groups.

Hypotheses:

e Those who were not satisfied with their oral health would be more likely to have

high OHIP scores

e Those who perceived that they needed dental treatment would be more likely to

have higher OHIP scores than those who did not.

e The edentulous subjects and dentate subjects wearing dentures would be more

likely to report more functional problems and have higher OHIP scores than

dentate subjects not wearing dentures

4) Comparison of the technical properties of the OHIP(M) with OHIP(Australia), and

OHIP(Canada)

The results of reliability and validity analyses were compared with those from the OHIP

(Australia) and OHIP (Canada) to establish measurement equivalence. The information on the

psychometric properties for OHIP (Australia) was obtained from a published paper by Slade

and Spencer (1994a), while OHIP (Canada) was obtained from the database of the Ontario

Study of the Oral Health of Older Adults (Locker and Slade, 1993).

The Cronbach’s alpha values and Intraclass correlation coefficients were compared.

Two analyses were performed in order to compare the ability of the measures to distinguish

between groups: the differences in the mean ranks (DMR) between categories of the

independent variables, obtained from the Mann-Whitney test and odds ratios based on the

median splits (Allen and Locker, 1997). To perform these analyses, the OHIP scores were

standardized to a range score of 0 to 100 using the formula below (Fayers and Machin, 2000):

108

Z = X [100/(m x (k-D}

Z= Standardize score

X= Sum-score

m= Number of items

k= Response categories

5) To compare two types of short form developed using two different methods.

Although a long measure provides more comprehensive data, some research settings do not

permit its use, thus, most health related quality of life measures develop a short form to provide

an alternative form of collecting data. There are different methods to shorten a measure. Slade

(1997c) derived a short form using regression analysis, which contains 14 items, two from each

domain. In this study, a short form was developed based on the most common items reported

by the patients who were interviewed at the qualitative interview stage. The two most

commonly reported items from each of the subscales were chosen to form the short version.

This method is known as item-frequency method. The question is between these two methods —

regression and item-frequency - which is a better method to produce a short form?

Thus, in order to assess the performance of the two different methods of deriving short

form, L-OHIP(M) data was used in creating the two forms - Frequency-S-OHIP(M) and

Regression-S-OHIP(M) - so that the comparison could be made on the same sample. The

Regression-S-OHIP(M) was reproduced using the controlled regression analysis as described

by Slade (1997c). The total score of the 45 items was used as the dependent variable and all the

45 individual items as an independent variable. Stepwise regression was performed. The item

that makes the least contribution to the R-squared and which is in a dimension that already has

two items in the model was deleted. The model was then refitted. The procedure was continued

until two items from each dimension were selected. To compare these two forms, the following

analyses were performed:

a. To compare the content,

109

b. To compare the score obtained using these two version of S-OHIP(M).

c. To compare the reliability (Cronbach’s alpha and ICC) of these two version of

S-OHIP(M).

d. To compare the concurrent and construct validity of these two version.

e. To compare the ability of the measures to distinguish between groups. Two

analyses were performed: the differences in the mean ranks (DMR) between

categories of the independent variables, obtained from the Mann-Whitney test

and odds ratios based on the median splits (Allen and Locker, 1997).

6) To describe the impact of oral conditions on the quality of life of the adult population.

Data obtained using the L-OHIP(M) was used to describe the impact of oral conditions on the

quality of life of the adult population of Malaysia. In this analysis, the response categories were

reduced to three: i) “very often” or “often”, ii) “sometimes” and iii) “once in a while” or

“never”. The cut off point used in estimating the proportion of impacts was “very often” and

“often”. This represents the most stringent cut off point and identifies those who experienced

oral health related problems on a relatively frequent basis (Locker and Slade, 1993).

7) To compare the impact of oral conditions on the quality of life between Malaysian and

Canadian population aged 50 and above.

The Malaysian data obtained from the present study was compared with Canadian data

obtained from the database of the Ontario Study of Older Adults. For the purpose of this

comparison, the age was standardized at 50 years old and above. The cut off point used for this

analysis was “very often” and “often”.

110

4.3) PRIVACY AND CONFIDENTIALITY

All participants involved in this study were assured that all information they supplied would be

kept confidential. No identifiers such as names or patient numbers would appear on any of the

instruments used during the study and no names would be used in any papers or reports. Once

the tape recordings of the in-depth interviews had been transcribed, the tapes would be

destroyed. Completed questionnaires would be destroyed once the data had been entered into

the computer. The signed consent forms would be retained but kept in a locked filing cabinet

separate from other documentation and data files created for the study.

4.4) SCIENTIFIC AND ETHICAL REVIEW

The protocol of this study was reviewed and approved by the Faculty of Dentistry Research

Committee and the Ethical Committee University of Toronto. There was no Ethical review at

the Faculty of Dentistry, University of Malaya. Thus, Ethical review was only obtained from the

University of Toronto.

CHAPTER 5

RESULT

5.1) TRANSLATION

This section describes the results of the translation process to produce an equivalent Malay

questionnaire (MQ) of the OHIP using a forward-backward translation technique. It involved

translating the original English questionnaire (OQ) (Appendix A) into the Malay language.

One of the drawbacks of the forward-backward translation technique is that when the

translators are aware that their work will be evaluated by back-translations, they tend to translate

literally (van de Vijver and Leung, 1997; Geisinger, 1994). However, in this study the

investigator was convinced that the translators (FT) had translated the questionnaire in a

dynamic way rather than in a literal way, and it is shown in the example below:

Example

OQ: Have you felt that your breath has been stale because of problems with your teeth, mouth or dentures?

Literally translated: Pernahkah anda merasakan yang nafas anda menjadi_basi

disebabkan oleh masalah gigi, mulut atau gigi palsu?

In this example, if the question had been translated literally, the translation would have no

sensible meaning in Malay language because there is no such phase as “nafas menjadi basi” in

Malay. The word “basi” that was literally translated for “stale” is commonly used to describe

food but not breath.

Table 5.1 shows the distribution of the level of difficulty in translating the content of the

questionnaire into the Malay language. The level of difficulty, i.e, easy, difficult, or impossible

(Herdman et al, 1998), was based on the consensus given by the forward-translators: at least two

of them agreed upon the degree of difficulty.

ill

112

Table 5.1. Level of translation difficulty.

Result No.ltem Easy to translate 37

Difficult to translate 12

Impossible to translate 0

More than 75% of the questions were rated as easy to translate and none of the items

were considered impossible to translate. Twelve questions were considered difficult to translate

(Table 5.2). These difficulties are discussed below.

Table 5.2. Difficult to translate questions

Q3) Have you noticed a tooth which doesn't look right?

Q5) Have you felt that your breath has been stale because of problems with your teeth, mouth or dentures?

Q8) Have you felt that your digestion has worsened because of problems with your teeth, mouth or dentures?

Q10) Have you had a sore jaw?

Q16) Have you had sore spots in your mouth?

Q20) Have you been self conscious because of your teeth, mouth or dentures?

Q21) Have dental problems made you miserable?

Q26) Have you felt that there has been less flavor in your food because of problems with your teeth, mouth or dentures?

Q32) Have you had to interrupt meals because of problems with your teeth, mouth or dentures?

Q34) Have you been upset because of problems with your teeth, mouth or dentures?

Q42) Have you been a bit irritable with other people because of problems with your teeth, mouth or dentures?

Q45) Have you suffered any financial loss because of problems with your teeth, mouth or dentures?

113

The first difficulty encountered in this study was related to finding equivalent Malay

words to explain the English concept underlying them. For example, the words “self-conscious”,

“miserable”, and “irritable” describe internal feelings. Members of Malay society are known to

be very shy and non-expressive, especially in matters related to personal feelings or emotions,

and this could place limits on the number of Malay words that describe feeling or emotions.

This is supported by a statement made by Asmah Haji Omor (1985), a Malay linguistic expert.

She wrote:

“It has long known and accepted fact that language is an index to culture. This

means that language reflects the culture of its speakers — the way of life they lead

as well their physical and social environment. Such function is borne by the

vocabulary of the language concerned. As such, the language of people engaged in agriculture is expected to be rich in the vocabulary of agricultural.”

To overcome this problem, the committee used the “Dewan English-Malay” dictionary

to look up Malay equivalents of these words (self-conscious, miserable, and irritable). The

Dewan English-Malay dictionary was used because the aim of this dictionary is to provide an

equivalent for Malay to English words. For example, it was found that the equivalent word for

““rritable” is “lekas/cepat marah’”’. Therefore, this phrase was used to mean “irritable”. However,

when this was translated back to English, the word “irritable” did not show up. Rather, it was

translated as “got angry easily”. Hence, the word “irritable” is assumed, to a certain extent, to be

similar to “got angry easily”.

The dictionary gives “sedih” as the equivalent for “miserable”, but “sedih” is also used

as the equivalent word for “sad” and “upset” (Q34). Thus, when these two questions (Q34 and

Q42) were translated into Malay, they ended up being the same. Therefore, one of the questions

had to be deleted.

Another word that caused difficulty was “self-conscious”. We found that one specific

dictionary (Dato Asmah Haji Omar, 2000) translated “self-conscious” as “sogok”. However,

114

“sogok” is not a widely used word and, hence, it has not been internalized by the Malaysian

people. Thus, this word could not be used. “Segan” was the closest word that was found to

represent “self-concious”. However, the word “segan” is commonly used interchangeably with

the word “malu” (“embarrass” or “shy’”), although academically these two words have slightly

different meanings. Furthermore, the word “segan” has a different meaning in different parts of

Malaysia. For example, in the northern part of Peninsular Malaysia, the word “segan” is used to

describe laziness rather than embarrassment or shyness. Nevertheless, the word “segan’” was

kept to be examined in the qualitative interviews.

The second problem in translation is related to the clarity of the words or the sentences.

This was seen clearly when translating “sore jaw” and “sore spot”. These two words were not

familiar or understood by the forward-translators. Thus, clarification of the meaning was

obtained from the developer of the OHIP. Based on his explanation, the investigator changed the

word “sore jaw” to “pain in the jaw” and “sore spot” to “ulcer”. These changes enabled the FT

to translate these words to the Malay language.

The third problem was related to the terms used, many of which could be misinterpreted

by the respondents. One example was “dental problem”. This phrase was easily translated into

Malay language as “masalah pergigian”. However, it was anticipated that by using this word,

the focus would be only on the teeth. Thus, to avoid such a problem, the phrase “dental

problem” was translated to “teeth, mouth and denture”. Question number 45, “Have you

suffered any financial loss because of problem with your teeth, mouth or dentures?” was totally

restructured at this stage to “Have you had to spend a lot of money due to problems with your

teeth, mouth or dentures?” This was done because the word “financial loss” was not suitable in

this context.

115

At this stage, no items were removed. However, there were questions which seemed to

have a similar meaning after they had been translated into the Malay language. This translated

questionnaire will be used in the qualitative interviews, and is discussed in the next section.

116

5.2) QUALITATIVE INTERVIEWS

This section describes the results of the in-depth interviews that were carried out to establish the

conceptual and items equivalence (content validity). Herdman et al (1998) suggested that the

most important aspect of cross-cultural adaptation is to determine that the two populations

(source and target) conceptualize health and quality of life in similar ways.

The results will be presented in two parts. Part 1 presents the results of the interview

with the patients in eliciting information on how oral diseases and oral disorders affect their

daily life. Part 2 presents the results related to assessing the content and face validity of the

translated OHIP questionnaire.

The interviews took five months to complete, starting September 2001 and ending in

January 2002. The interviews were conducted in an isolated room. This had at least two

advantages: 1) it improved the quality of recording, which was extremely important to facilitate

transcription, and 2) it helped to reduce distractions, which made the patient more focussed on

the interview. An external microphone was added to the tape recorder after the interviewer

realized that the first few interviews were not very clear.

To encourage participation of the patients, an incentive was given. This included a dental

examination and appropriate treatment by the investigator, tokens in terms of oral care products

and money. Such incentives, especially immediate dental examination, have been shown to be

good motivational factors in encouraging participations.

On average, each interview was about 20 minutes long. In general, the interviewer

probed heavily for answers, especially in matters related to social well-being. The interviews

were conducted in Malay or English, depending upon the patient’s preference. However, most

of the time the interviews ended up using both languages. It was also observed that Chinese and

Indians preferred to speak English during the interviews.

117

A total of thirty-six patients were interviewed (Table 5.3). Of those, 17 respondents were

Chinese. About half of the total sample was a middle-aged group (40-59 years old). The

patients interviewed had a range of oral diseases and disorders. It is important to note that the

purpose of sampling was not to get a representative sample but, rather, to get a broad view on

how oral diseases or disorders affect people’s lives across age and ethnicity in Malaysia.

The number of Chinese interviewed was higher than Malays and Indians. This could be

due in part to the time frame in which the interviews were conducted. The months of November

and December are fasting months as well as the religious celebration for the Malay. During

those months, the number of Malay people attending dental clinics drops drastically. The low

number of Indian participants could be due to the fact that most Indians have good teeth and,

therefore, do not to go to dental clinics for treatment. In addition, the interview months were at

the end of the year, which happens to be a school holiday.

Table 5.3. Sample distribution for qualitative interview.

Agegroup Malay Chinese indian Total

18-39 6 5 2 13

40-59 6 8 4 18

60+ 0 4 1 5

Total 12 17 7 36

Part 1: From the transcriptions, significant statements on the impact of oral diseases and oral

disorders were extracted and treated as raw data. Then, duplicate statements were eliminated,

which resulted in forty-nine significant impacts as listed in Table 5.4. Of these, three were

considered to be new impacts, which had frequently emerged in the interviews. They were

“unable to enjoy their favorite foods”, “losing appetite” and “‘no self-confidence”.

Table 5.4. Impacts of oral disease and oral disorder.

118

1. OONAAEWN

20. 21. 22. 23.

Difficult to chew foods (especially hard foods: nuts, chicken bone, sugar cane)

Can't bite hard foods.

Can't pronounce certain words (eg: “s”) Face look ugly “like old person” Look awkward

Loose denture

Unsatisfied with the look of the denture

Cause headache Cause pain on the other part of the body (eg. Ear, hand, neck)

. Toothache

. Pain on the gums

. Ulcer

. Sensitive to cold water

. Denture pain

. Feel shy

. Worry to laugh or talk. . Feel uncomfortable due to food stuck in

between the teeth or dentures.

. Scared that it might worsen

. Scared that the tooth next to it will be

affected

Bad breath

Feel worry about it Feel very unhappy

Uncomfortable denture

24.

25. 26. af. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39.

40. 41. 42. 43. 44. 45. 46. 47. 48.

Avoid certain foods (eg. Sweets, cold drink, hard foods) Avoid chew on the pain side.

Avoid talking Avoid laughing Difficult to brush teeth. Unable to eat favorite foods Have to cover the mouth when smiling Have to blend the foods. Food feel less tasty Difficult to clean Have to brush more frequently

Avoid using that side. Tend to close the mouth Disturb sleep (have to take pain killer)

Sleep ail day long Unable to concentrate (eg: to other people conversation, study, work). Lazy to do other works besides sleeping. Easy to get angry Loss appetite. Can’t enjoy the foods Preferred to stay at home. Less confident to talk to people Tend to be quiet Tend to stand far from people Have to take medical certificate (mc) Financial burden.

These impacts were then analyzed and grouped into a hypothetical domain as in the

original development of the OHIP. They are “functional limitation”, “physical pain”,

“psychological discomfort”, “physical disability”, “psychological disability”, “social disability”

and “handicap”. Although there were impacts that were not listed in the original questionnaire,

they belonged to one of the domains (Appendix N). No new domain emerged, nor was a domain

excluded. Thus, it could be concluded that Malaysians and Australians conceptualize oral health

similarly in terms of how oral diseases or oral disorders affect daily life.

Part 2: In the second part of the interview, the patients were shown the translated questionnaire

and asked to comment on the relevancy and their understanding of each statement. Table 5.5

119

shows the distribution of items based on relevancy and the respondents’ understanding of the

questions. Most of the items were understood and relevant. Two pairs of questions, in the

translated version, were viewed by most of the patients interviewed as being the same. They

were question Q20 “self-conscious” with question Q38 “embarrassed” and question Q21

“miserable” with question Q34 “upset”.

Table 5.5. Relevancy of items

Situation Number of items Understood and relevant 40

Understood and Similar meaning 2 pairs

Difficult to understand/Ambiguous 4

Understood but irrelevant 4

Although there is no precise Malay word for “self-conscious”, the concept does exist in

Malaysian society. This was shown in the following narrative.

“Kita gigi rongakkan, jadi kita rasa janggallah, tapi kawan-kawan tak kisah

pun.... Dia orang buat macam biasa je.., tapi saya malu” (I don’t have teeth, so I feel odd, but my friends don’t bother... they act as

normal.. but me myself feel shy.) 52-year- old Malay woman

The word “malu’” here was used to describe her own feeling of being self conscious rather than

being embarrassed. However, the word “malu” was also used to describe the feeling of being

embarrassed, as shown in the following narrative.

“sometime saya rasa malu bila when people making joke of my braces....gigi besi” (sometime, I felt embarrassed when people making joke of my braces.....iron teeth)

32-year-old Malay man, orthodontic patient

120

Therefore, the word “malu” was used interchangeably, depending on the context where

it had been used. However, in a dental context, it very often referred to being self-

conscious rather than being embarrassed.

Four questions were considered as difficult to understand and ambiguous as listed in

Table 5.6. For example, in Q3, the majority of respondents did not understand the questions.

Whereas question Q9 “the pain in the mouth” was confusing and too broad, and most patients

interviewed asked whether it included toothache or gum pain. For question Q29 and Q48, the

words “diet pemakanan” and “berfungsi sepenuhnya” were not understood by most of the

respondents, especially those who were older and uneducated. In one question, Q6, “sense of

taste” seemed to be irrelevant, since most patients interviewed did not feel that his/her sense of

taste had anything to do with their teeth, mouth or dentures. One of the patients felt that the

cause of her reduced sense of taste was due to her age: she believed that it was related to her

blood circulation and not because of her teeth, mouth or dentures.

121

Table 5.6. Problem questions

Difficult to understand/ambiquous:

Q3) Pernahkah anda perasan terdapat gigi yang kelihatan tidak sempurna? (Have you noticed a tooth which doesnt look right?)

Q9) Pernahkah anda mengalami kesakitan di dalam mulut anda? (Have you experienced pain in your mouth?)

Q29) Pernahkah anda merasakan diet pemakanan anda menjadi tidak memuaskan disebabkan masalah gigi, mulut atau gigi palsu anda? (Have you felt that your diet became less satisfactory because of problems with

your teeth, mouth or dentures?)

Q48) Pernahkah anda tidak dapat berfungsi sepenuhnya disebabkan masalah gigi, mulut atau gigi palsu anda? (Have you been unable to function fully because of problems with your teeth, mouth or dentures?)

Understood but irrelevant:

Q6) Pernahkah anda merasakan yang deria rasa anda menjadi semakin kurang disebabkan masalah gigi, mulut atau gigi palsu anda? (Have you felt that your sense of taste has become worse due to problems with your teeth, mouth or dentures?)

122

5.3) TRANSLATED AND ADAPTED QUESTIONNAIRE

Based on the translation process and qualitative interviews’ results, a Malaysian version of the

OHIP questionnaire was produced. This adapted questionnaire will be referred to hereafter as

the Malaysian Oral Health Impact Profile [OHIP(M)], which will be used in the subsequent

stages of this study. The word “M” in brackets indicates “Malaysia”.

Seven items were excluded from the original version either because the items were

ambiguous, yielded similar meanings to other items, or were not relevant (Table 5.7). Three

items were restructured and/or reworded and three items were added, as shown in Table 5.7. All

patients interviewed reported that they had experienced food getting stuck between their teeth or

under their dentures, and that this had become a normal occurrence for them. Thus, the

committee felt that it was more meaningful to ask patients whether or not they had experienced

discomfort due to food getting stuck between the teeth or under the dentures.

In Q8, when “digestion has worsen”, was translated to Malay, the understanding was that

the person already had a degree of digestive problems and whether or not dental problems had

made this worse. Therefore, the committee felt that the question was not appropriate. Hence, the

question was reworded to “Have you felt that foods you eat have not digested properly due to

problems with your teeth, mouth or dentures?”

In Q41, “trouble getting along with other people” was regarded as giving a negative

impression to people. Therefore it was anticipated that it would not elicit honest answers from

the respondents. Hence, the question was reworded as shown in Table 5.7. Three new questions

were included in the questionnaire (Table 5.7). This was done because most patients interviewed

reported these impacts.

123

Table 5.7. Deleted, reworded and new items

DELETED ITEMS:

Q3) Q6)

Q9) Q20) Q21) Q29)

Q48)

Have you noticed a tooth which doesn't look right? Have you felt that your sense of taste has worsened because of problems with your teeth, mouth or dentures?

Have you had painful aching in your mouth?

Have you been self conscious because of your teeth, mouth or dentures?

Have dental problems made you miserable? Has your diet been unsatisfactory because of problems with your teeth, mouth or

dentures?

Have you been totally unable to function because of problems with your teeth, mouth or dentures?

REWORDED/RESTUCTURED ITEMS:

Q7)

Q8)

Q41)

Have you had food catching in your teeth or dentures? (Have you experienced food getting stuck in between your teeth or dentures?)

Have you felt that your digestion has worsened because of problems with

your teeth, mouth or dentures? (Have you felt that your digestion is

becoming poorer due to problems with your teeth, mouth or dentures?)

Have you had trouble getting along with other people because of problems with your teeth, mouth or dentures? (Have you experienced problems having friendships with others because of problems with your teeth, mouth or dentures?)

NEW ITEMS:

A25) A34) A44)

Have you felt discomfort due to food getting stuck in between your teeth or dentures?

Have you felt that foods you eat have not digested properly due to problems with your teeth, mouth or dentures?

Have you been unable to mix around with

others because of problems with your teeth, mouth or dentures?

Have you been unable to eat your favorite foods? Have you felt a loss of appetite due to problems with your teeth, mouth or dentures?

Have you felt less confident of yourself due to problems with your teeth, mouth or dentures?

124

The OHIP(M) at this stage was divided into four parts as shown in Table 5.8. Part A

consists of 45 questions of impacts as compared to 49 questions in the original questionnaire.

Part B consists of three global self-rating questions. Part C has four questions regarding

dentition and dentures. Part D consists of questions regarding respondents’ socio-demographic

background.

Tabie 5.8. Format of questionnaire

Part Part A: OHIP questions

45 questions

Part B: Global self ratings: i. Self-rating of oral health

ii. Self-perceived treatment need iii. | Satisfaction with oral health.

Part C: Dental status 4 questions

Part D: Sociodemographic: ii Age

ii, Sex iii. Ethnicity iv. Level of education

Type of data Ordinal/Interval

Ordinal

Dichotomous

Dichotomous

Dichotomous

Interval

Nominal

Nominal Nominal

Categories Likert scale (5 point scale). (Very often/often/sometimes/ once a while/never)

Very good/Good/ Fair/ Poor Yes/No Yes/No

Yes/No

Birth date

Male/Female Malay/Chinese/India/Other No education/ Primary/

Secondary/ University/ Other)

Two versions of the OHIP(M) were developed: a long form consisting of 45 items and

designated as L-OHIP(M) (Appendix E), and a short form consisting of 14 items and designated

as S-OHIP(M) (Appendix F).

125

5.4) PRE-TESTING

A total of twenty respondents completed the L-OHIP (M) questionnaire. Table 5.9 shows the

characteristics of the respondents. The mean age of the respondents was 44.1 years old, ranging

from 18 to 88 years old. Sixty percent of the respondents were female, and the majority were of

Malay ethnicity. All respondents had at least a primary education.

Table 5.9. Characteristics of the respondents for Pretesting

Characteristics N (%) Mean (SD) Age (n=20)

18-39 7 (35.0) 44.1 (19.4) 40-59 10 (50.0) 60+ 3 (15.0)

Gender (n=20) Male 8 (40.0) Female 12 (60.0)

Ethnic (n=20) Malay 9 (45.0) Chinese 7 (35.0) indian 4 (20.0)

Education level (n=20) Primary 1 (5.0) Secondary 9 (45.0) University 10 (50.0)

All respondents agreed that the format and instruction on how to complete the

questionnaire were easy to follow. Almost all respondents completed the entire questionnaire.

This indicates that the instructions and the questions were easily understood. On average, the

time taken to complete the L-OHIP(M) questionnaire was 20 minutes (Table 5.10).

Most of the respondents had no difficulty understanding the questions except one person

who had difficulty understanding the word “rahang”. Since only one person did not understand

the word “rahang”, it was retained. No major changes were made except for a minor change to

126

question D4 (level of education), where one category was added to the response options to

provide an option to those who felt that they had other forms of education. Since only one minor

change was made, the questionnaire was not re-tested.

Tabie 5.10. Summary of pretesting finding.

Area tested Result

Format and instruction of the questionnaire:- e Did you have any difficulty to complete

the questionnaire?

e Was the instruction easy to follow?

Wording of questions:- e Were there any questions that you found difficult to understand?

Time to complete:- e How long the questionnaire took to complete?

All respondent did not have any problem to follow the instruction.

Most respondents had no difficulty understanding the questions.

20 minutes

127

5.5) TEST OF RESPONSE OPTIONS

A total of 171 respondents completed the S-OHIP (M) questionnaire. Table 5.11 shows the

characteristics of the respondents. The mean age of the respondent was 31.6 years old, and

ranged from 18 to 100 years old. The majority of respondents were younger adults (18 to 39

years old), female and of Malay ethnicity. All respondents had at least a primary education.

Table 5.11. Characteristics of the respondents for test of response options study.

Characteristics N (%) Mean (SD) Age

18-39 122 (73.5) 31.6 (13.8) 40-59 39 (22.9) 60+ 6 (3.6) Total 166 (100.0)

Gender Male 59 (34.9) Female 110 (65.1) Total 169 (100.0)

Ethnic

Malay 122 (71.8) Chinese 24 (14.1) Indian 22 (12.9) Total 170 (100.0)

Education level Primary 2 (1.2)

Secondary 78 (46.2) University 89 (52.6) Total 169 (100.0)

Figure 5.1 shows the distribution of the response options for each of the questions. It was

observed that all ranges of options were utilized. This indicates that the response options of five-

point frequency Likert scales can be used for a Malaysian adult population.

Percent

Percent

“Mlezing ‘Once a white

Never ‘Sometime

Difficulty chewing Bad breath

Percent

Once a white Percent

Sometime Vary often

Food stuck discomfort

‘Somabina

Percent

Percent

Onca a while Sometime Never

Lass confident

‘Somakme

,

50

40

30

Py

* 2 0 z ®

: i é @ Ok peewee Mizaing (Ores 2 while Often

Maver Sometime Very ofan ‘ever Sametime Very often

Avoid eating Avoid smiling

100

80

60

40

20

5 3 i 3 8 é a ol : .

Missing: Miseing Often:

Never Semotine Vary offen Somotime Vary often,

Concentration disturbed Avoid going out

Percent

Percent

Percent

Missing ‘Onea a while Often Nover Sometime Very offen

Eating discomfort

Percent

Once awhile Sometime Vary offen

Once o white

Never Sometime

Sleep disturbed

Very often

Missing Oncea while

Sometires

Daily activites

Figure 5.1. Distribution of response options.

128

129

5.6) FIELD TESTING

5.6.1) Total Response Rate

Table 5.12 shows the response rate for the field-testing stage. A total of 979 subjects were

selected from the NOHSA sample. Of those, 694 subjects were allotted for a mail questionnaire

and 285 were allotted for an interview. All 694 subjects were sent a questionnaire (long or

short). Sixty-six questionnaires were returned uncompleted, either because of a wrong address

or because the person had moved or died. Thus, an eligible sample for the mail questionnaire

was 628 respondents. About 48% of those subjects completed the questionnaire.

Of the 285 subjects visited for an interview, 73 respondents had moved away, 40 refused

to be interviewed and 46 were not available even after two attempts were made by the

interviewers. The response rate for interviews was almost 60%.

Tabie 5.12. Response rate for field testing stage.

Interview Mail Total 1) Original Sample 285 694 979

2) Ineligible sample: 139 Moved 67 18 Wrong address 6 46 Died - 2 Total 73 66

3) Eligible sample (1-2): 212 628 840

4) Non-responders: 414 Refusal 40 NA Not available (2 visit) 46 NA Unreturned questionnaire NA 328 Total 86 328

5) Completed 126 300 426

6) Response Rate (5/3) 59.4% 47.8% 50.7%

5.6.2) Characteristics of respondents and non-respondents.

130

Table 5.13 shows the characteristics of the respondents and non-respondents. It was observed

that a higher proportion of the non-respondents were young adults (18 to 39 years old);

however, no significant difference was observed in terms of gender. More Malays than Chinese

and Indians responded. Almost 60% of the respondents were Malay.

Table 5.13. Characteristics of respondents and non-respondents.

Characteristics

Respondent Non-respondent p-value N % N %

Agegroup (year-old): 18-39 208 48.8 232 56.0 40 - 59 166 39.0 126 30.4 0.034 60+ 52 12.2 56 13.5 Total 426 100.0 414 100.0

Gender: Male 185 43.4 191 46.1 0.430 Female 241 56.6 223 53.6 Total 426 100.0 414 100.0

Ethnic:

Malay 253 59.4 158 38.2 0.000 Chinese 108 25.4 180 43.5

indian 50 11.7 73 17.6 Others 15 3.5 3 0.7 Total 426 100.0 414 100.0

Chi-sq test for comparing group differences

131

5.6.3) Quality of data

Table 5.14 and 5.15 show the percentage of missing data (blank and “don’t know” responses)

for each item. For the S-OHIP(M), the total of missing data ranged from 0.5 to 2.9 percent. The

item “had bad breath” had the highest missing values.

Table 5.14. Percentage of missing items for the S-OHIP(M)

Item Don’t know Blank Total N (%) N (%) N (%)

A1 ___ Difficult chewing any foods 2 (1.0) 1 (0.5) 3 (1.5) A4 Had bad breath cause by dental problem 5 (2.4) 4 (0.5) 6 (2.9) A1i1 Discomfort eating any foods 0 (0.0) 1 (0.5) 1 (0.5) A1i2 Had ulcer in the mouth 3 (1.5) 1 (0.5) 4 (1.9) A116 Felt discomfort due to food getting stuck. 0 (0.0) 4 (0.5) 1 (0.5)

A17_ Felt shy 4 (1.9) 0 (0.0) 4 (1.9) A24_ Had to avoid eating some foods 1 (0.5) 0 (0.0) 1 (0.5)

A27 Avoided smiling 1 (0.5) 0 (0.0) 1 (0.5) A29 Your sleep been disturbed 2 (1.0) 0 (0.0) 2 (1.0) A33 Your concentration been disturbed 3 (1.5) 0 (0.0) 3 (1.5) A35 Avoided going out 2 (1.0) 0 (0.0) 2 (1.0) A39_ Problems in carrying out daily activities 4 (0.5) 0 (0.0) 1 (0.5) A41_ Had to spend a lot of money 2 (1.0) 0 (0.0) 2 (1.0) A44__ Felt less confident of yourself 3 (1.5) 0 (0.0) 3 (1.5)

Note: N=206

The total missing items for the L-OHIP(M) ranged from 0.5 to 7.3 percent. Seven items

had more than four percent of missing data. The percentage of “don’t know” responses ranged

from 0 to 5.5 percent. Generally, a very small percentage (less than 2 percent) was left blank.

All items had a missing value of less than five percent except “felt uncomfortable with the

appearance”, which was 7.3 percent. Of that 7.3 percent, 5.5 percent of the respondents provided

a “don’t know” response.

132

Table 5.15. Percentage of missing items for the L-OHIP(M)

Ttem Don’t know Blank Total

N (%) N (%) N (%) Al Difficult chewing any foods 0 (0.0) 1 (0.5) 1 (0.5) A2 Trouble pronouncing words 1 (0.5) 1 (0.5) 2 (1.0) A3 Felt that appearance has been affected 9 (4.1) 4 (0.5) 10 (4.6) A4 Had bad breath cause by dental problem 6 (2.7) 1 (0.5) 7 (3.2) Ad Feit that foods you eat have not digested 6 (2.7) 1 (0.5) 7 (3.2)

properly.

A6 Had pain on the jaw 8 (3.8) 2 (0.9) 10 (4.5) AT Had headache due to dental problem 10 (4.5) 4 (0.5) 11 (5.0) A8& Had sensitive teeth 3 (1.4) 2 (0.9) 5 (2.3) AQ Had toothache 3 (1.4) 1 (0.5) 4 (1.8) A10_— Had painful gums 3 (1.4) 1 (0.5) 4 (1.8) Ai1 Discomfort eating any foods 2 (0.9) 2 (0.9) 4 (1.8) A1i2 Had ulcer in the mouth 6 (2.7) 1 (0.5) 7 (3.2) Ai3 Felt denture was loose 4 (1.8) 2 (0.9) 6 (2.7)

A14 Felt uncomfortable with your dentures 6 (2.7) 3 (1.4) 9 (4.1) Ai5 Felt worried 6 (2.7) 4 (1.8) 10 (4.5) Ai6 Felt discomfort due to food getting stuck. 3 (1.4) 3 (1.4) 6 (2.8) A17 Felt shy 7 (3.2) 2 (0.9) 9 (4.1) A18 Felt uncomfortable with the appearance of the 12 (5.5) 4 (1.8) 16 (7.3)

teeth, mouth or dentures

A19 Felt stressed up 5 (2.3) 4 (0.5) 6 (2.7) A20 Speech been unclear 6 (2.7) 1 (0.5) 7 (3.2) A21 People misunderstood some of your words 4 (1.8) 4 (0.5) 5 (2.3) A22 Felt food less tasty 4 (1.8) 2 (0.9) 6 (2.7) A23 Been unable to brush your teeth properly 1 (0.5) 2 (0.9) 3 (1.4) A24 Had to avoid eating some foods 0 (0.0) 1 (0.5) 1 (0.5) A25 Been unable to eat your favorite foods 3 (1.4) 2 (0.9) 5 (2.3) A26 __ Difficulty eating because of problem related to 1 (0.5) 4 (1.8) 5 (2.3)

the dentures A27 = Avoided smiling 4 (1.8) 3 (1.4) 7 (3.2) A28 Had an interrupt meals 4 (0.5) 2 (0.9) 3 (1.4) A29 Your sleep been disturbed 4 (1.8) 2 (0.9) 6 (2.7) A30 Been sad 4 (1.8) 2 (0.9) 6 (2.7) A31___ Difficult to relax 4 (1.8) 2 (0.9) 6 (2.7) A32 ~~ Felt depressed 3 (1.4) 2 (0.9) 5 (2.3) A33 Your concentration been disturbed 5 (2.3) 0 (0.0) 5 (2.3)

A34_ ‘Felt a loss of appetite to eat 3 (1.4) 0 (0.0) 3 (7.4)

A35 Avoided going out 4 (0.5) 0 (0.0) 1 (0.5) A36_— Been less tolerant of your spouse or family 2 (0.9) 0 (0.0) 2 (1.0) A37 Unable to mix around with other people 1 (0.5) 0 (0.0) 1 (0.5) A38 Got angry easily. 3 (1.4) 0 (0.0) 3 (1.4) A39 Problems in carrying out daily activities 4 (1.8) 4 (0.5) 5 (2.3) A40_ Felt unwell 2 (0.9) 4 (0.5) 3 (1.4) A41 Had to spend a lot of money 3 (1.4) 1 (0.5) 4 (1.8) A42_ Felt less happy to be in the company of others 3 (1.4) 1 (0.5) 4 (1.8) A43 Been less satisfied with life 4 (0.5) 4 (0.5) 2 (1.9)

A44 Felt less confident of yourself 0 (0.0) 1 (0.5) 4 (0.5) A45__ Been unable to work to your full capacity 3 (1.4) 4 (0.5) 4 (1.8)

Note: N=220

133

5.6.4) Comparison of mode of administration — mail vs interview.

Table 5.16 shows the characteristics of the eligible sample for mail and interview groups. Both

samples were similar in terms of age group and gender distribution. However, there was a

higher proportion of Chinese in the interview group. All interview samples were located in an

urban area, whereas almost 25 percent of the mail samples were located in a rural area.

Table 5.16. Characteristics of the eligible sample for mail and interview group.

Characteristics Mail interview

N(%) N(%)

Age 18-39 319 (50.8) 121 (57.1)

40-59 217 (34.6) 75 (35.4)

60+ 92 (14.6) 16 (7.5)

Gender

Male 288 (45.9) 88 (41.5)

Female 340 (54.1) 124 (58.5)

Ethnic

Malay 319 (50.8) 92 (43.4) Chinese 189 (30.1) 99 (46.7)

indian 105 (16.7) 18 (8.5) Others 15 (2.4) 3 (1.4)

Strata

Urban 475 (75.6) 212 (100.0)

Rural 153 (24.4) -

To compare the two modes of administration, three parameters were assessed, they were:

a) response rate,

b) percentage of incomplete data.

c) OHIP(M) score.

134

5.6.4.1) Response rate

Table 5.17 shows the response rate of mail questionnaires and interviews for the L-OHIP(M)

and the S-OHIP(M). In total, the interview response rate was significantly higher than the

response rate for the mail questionnaire. Almost 60% of respondents completed the interview.

Less than 50% of the total sample for the mail group returned the questionnaire. The type of

questionnaire used - either long or short - did not show any significant difference in the response

rate for the mail questionnaire. This suggests that the length of the questionnaire did not

determine the response rate.

Table 5.17. Response rate of mail and interview for the L-OHIP(M) and the S-OHIP(M)

L-OHIP(M) S-OHIP(M) TOTAL

Total Responded | Total Responded Total Responded

N n (%) N n (%) N n (%)

Mail 310 149 (48.1)* 318 151 (47.5)* 628 300 (47.8)*

interview 125 71 (56.8)* 87 55 (63.2)* 212 126 (59.4)*

TOTAL 435 220 (50.6) 405 206 (50.9) 840 426 (50.7)

*Chi-sq test, p<0.05

5.6.4.2) Percentage of incomplete data.

Table 5.18 shows the percentage of incomplete data. The data was regarded as incomplete when

more than twenty percent of the items were either left blank or had a “don’t know” response. In

total, less then three percent of the completed questionnaires were considered incomplete. The

percentage of incomplete data for the L-OHIP(M) and the S-OHIP(M) mail questionnaires were

four and two percent respectively. However, these differences were not significant.

Table 5.18. Percentage of the incomplete data.

135

L-OHIP(M) S-OHIP(M) N Complete incomplete N Complete incomplete

(< 20%) (>20%) (< 20%) (>20%) n (%) n (%) n (%) n (%)

Mail 149 143 (96.0) 6 (4.0) 154 148 (98.0) 3 (2.0))

Interview 71 71 (100.0) 0 (0.0) 55 55 (100.0) 0 (0.0)

Total 220 214 (97.3) 6 (2.7) 206 203 (97.1) 3 (1.5)

5.6.4.3) OHIP Score

Table 5.19 shows the mean scores (ADD-Score and SC-Score) of both types of questionnaires,

i.e., L-OHIP(M) and S-OHIP(M), for the two types of modes of administration (mail and

interview). It was observed that the mean scores - both ADD score and SC score - were lower

for interviews than for mail for both the L-OHIP(M) and the S-OHIP(M). However, the

differences were not significant.

Table 5.19. Mean ADD-Score and SC-Score

L-OHIP(M) S-OHIP(M)

ADD-Score $C-Score ADD-Score SC-Score

N Mean (SD) Mean N Mean (SD) Mean (SD) (SD)

Mail 143 28.48 (24.78) 2.71 (4.95) 146 11.25 (9.48) 1.16 (2.20)

Interview 71 25.27 (18.47) 1.89 (3.13) 54 10.15 (6.69) 0.98 (1.14)

Total 214 27.42 (22.89) 2.44 (4,44) 200 10.96 (8.81) 1.11 (2.01)

Summary for mode of administration:

1) Interview yielded a higher response rate than mail questionnaire.

2)

3)

4)

was not statistically significant.

Less missing data for the interview.

Length of questionnaire did not affect the response rate.

Score tends to be lower for interview than mail questionnaire; however, the difference

136

5.6.5) Reliability

5.6.5.]) Internal consistency

The internal consistency was assessed by Cronbach’s alpha. Table 5.20 shows the Cronbach’s

alpha values for the seven subscales of the L-OHIP(M) and the overall alpha value for the S-

OHIP(M) by mode of administration. Cronbach’s alpha was not calculated for the S-OHIP(M)

subscales. The overall Cronbach’s alpha for the seven subscales of the L-OHIP(M) ranged from

0.72 to 0.87. A higher value was observed for the mail questionnaire for all the subscales, which

ranged from 0.75 to 0.88 than the interview, which ranged from 0.66 to 0.84. However,

regardless of mode of administration all values were more than 0.60, indicating good to

excellent reliability. For the S-OHIP(M), the scale Cronbach’s alpha value was 0.89. The mail

questionnaire had a higher value than the interviews.

Table 5.20. Internal consistency for the L-OHIP(M) and the S-OHIP(M) by mode of administration.

L.-OHIP(M) S-OHIP(M)

Subscale Ni Mail Interview Total NI Mail interview Total

(N=143) (N=71) (N=214) (N=148) (N=55) (N=203)

Functional Limitation 6 0.77 0.73 0.76 ee Physical Pain 7 0.75 0.66 0.72

Psychologica! Discomfort |6 0.85 0.80 0.84

Physical Disability 9 0.88 0.84 0.87

Psychological Disability 6 0.88 0.73 0.86

Social Disability 5 0.83 0.73 0.84

Handicap 6 0.82 0.66 0.78

Scale 45 0.96 0.93 0.95 14 0.91 0.80 0.89 Note: NI = number of items

Table 5.21 shows the Cronbach’s alpha values by age groups. The mean age for the

respondents completing the L-OHIP(M) was 25.4 (sd 4.8) for the 18 to 34 years old group and

45.5 (sd 11.1) for the 35 years old and older group. For the S-OHIP(M) respondents, the mean

137

age was 26.1 (sd 5.2) for the 18 to 34 years old group and 48.5 (sd 11.5) for the 35 years old and

older group. Cronbach’s alpha values for the scale - both L-OHIP(M) and S-OHIP(M) - were

shown to have excellent internal consistency in both agegroups. For respondents aged 35 and

above who completed the L-OHIP(M), the Cronbach’s alpha values ranged from 0.72 to 0.88,

and indicated good internal consistency. However, for the 18-34 year old agegroup, the

handicap subscale showed lower Cronbach’s alpha values (less then 0.60).

Table 5.21. Internal consistency for the L-OHIP(M) and the S-OHIP(M) by agegroup.

L-OHIP(M) S-OHIP(M)

Subscale Agegroup Agegroup

NI 18-34 35+ NI 18-34 35+

(N=65) (N=149) (N=86) (N=117)

Functional Limitation 6 0.68 0.77

Physical Pain 7 0.73 0.72

Psychological Discomfort 6 0.82 0.84

Physical Disability 9 0.80 0.88

Psychological Disability 6 0.83 0.87

Social Disability 5 0.76 0.82

Handicap 6 0.56 0.82 :

Scale 45 0.94 09 | 14 088 0.90

Note: NI = number of items

5.6.5.2) Test-retest

A total of 71 and 73 respondents completed the L-OHIP(M) and the S-OHIP(M), respectively,

for the second time. For the L-OHIP(M), 42 respondents completed the mail questionnaire and

29 respondents were re-interviewed. For the S-OHIP(M), 49 respondents completed the mail

questionnaire and 24 respondents were re-interviewed (Table 5.22)

138

Table 5.22. Number of respondents who completed the questionnaire for the second time.

Mail interview Total

L-OHIP(M) 42 29 71

S-OHIP(M) 49 24 73

Table 5.23 shows the characteristics of the respondents. There was a higher proportion

of respondents in the younger adult group (aged 18-39 years old) in the S-OHIP(M) sample than

in the L-OHIP(M) sample. The L-OHIP(M) sample had three more respondents with no formal

education than the S-OHIP(M) sample. The distributions in terms of gender and ethnicity for

both samples were similar, with a higher proportion of female and Malay in both samples.

Table 5.23. Respondents’ characteristics — Test- Retest.

L-OHIP(M) S-OHIP(M Characteristics Original Retest Original Retest

(N=214) (N=71) (N=203) (N=73) n(%) n(%) n(%) n(%)

Agegroup (year-old): 18-39 91 (42.5) 35 (49.3) 114 (56.2) 46 (63.0) 40 — 59 94 (43.9) 29 (40.8) 68 (33.5) 22 (30.1) 60+ 29 (13.6) 7 (9.9) 21 (10.3) 5 (6.8)

Gender: Male 96 (44.9) 29 (40.8) 89 (43.8) 31 (42.5) Female 118 (55.71) 42 (59.2) 114 (56.2) 42 (57.5)

Ethnic: Malay 423 (57.5) 48 (67.6) 424 (61.1) 49 (67.1) Chinese 58 (27.1) 17 (23.9) 49 (24.1) 18 (24.7) indian 25 (11.7) 3 (4.2) 23 (11.3) 5 (6.8) Others 8 (3.7) 3 (4.2) 7 (3.4) 1 (1.4)

Level of education: No formal education 25 (11.7) 6 (8.5) 14 (6.9) 3 (4.1) Primary Schooi 38 (17.8) 41 (15.5) 35 (17.2) 11 (15.1) Secondary School 117 (54.7) 41 (57.7) 109 (53.7) 45 (61.6) College 14 (6.5) 5 (7.0) 20 (9.9) 8 (11.0) University 20 (9.3) 8 (11.3) 25 (12.3) 6 (8.2)

139

To assess the stability of the instrument, the Intraclass Correlation Coefficient (ICC) was

calculated for the overall scale and seven subscales. The results are shown in Table 5.24. For the

L-OHIP(M), the overall ICC ranged from 0.67 to 0.80 in seven subscales, whereas the S-

OHIP(M) ranged from 0.76 to 0.86. Overall, the short form of the questionnaire yielded better

results than the long form of the questionnaire.

Table 5.24. intraclass correlation coefficient by mode of administration.

L-OHIP(M) S-OHIP(M) Subscale Both Mail interview Both Mail Interview

(N=71) = (n=42) (n=29) (N=73) (n=49) (n=24)

Functional Limitation 0.79 0.79 0.79 0.78 0.79 0.75

Physical Pain 0.79 0.78 0.81 0.76 0.78 0.73

Psychological Discomfort 0.67 0.64 0.74 0.76 0.79 0.72

Physical Disability 0.71 0.71 0.73 0.88 0.91 0.82

Psychological Disability 0.72 0.75 0.59 0.81 0.84 0.71

Social Disability 0.77 0.84 0.63 0.76 0.78 0.45

Handicap 0.80 0.75 0.84 0.86 0.90 0.70

Scale 0.79 0.76 0.84 0.91 0.92 0.91

Table 5.25 shows the ICC values by age group. In general, the ICC values were more

than 0.7, indicating a good to excellent reliability. Psychological discomfort and physical

disability subscales of the L-OHIP(M) and the social disability subscale of the S-OHIP(M) had

an ICC lower than 0.7; however, all had an ICC above 0.6. The ICC value for the overall scale

was higher in the younger age group compared to the older adults for the L-OHIP(M), and vice

versa for the S-OHIP(M).

140

Tabie 5.25. intraclass correlation coefficient by agegroup

L-OHIP(M) S-OHIP(M)

Subscale Agegroup Agegroup

18-34 35+ 18-34 35+

(N=29) (N=42) (N=34) (N=39)

Functional Limitation 0.70 0.77 0.74 0.90

Physical Pain 0.73 0.83 0.71 0.79

Psychological Discomfort 0.74 0.64 0.67 0.82

Physical Disability 0.83 0.66 0.77 0.94

Psychological Disability 0.73 0.71 0.84 0.81

Social Disability 0.85 0.73 0.82 0.67

Handicap 0.77 0.78 0.88 0.85

Scale 0.85 0.75 0.84 0.95

5.6.6) Validity

5.6.6.1) Convergent Construct Validity.

It was observed that the mean score of the L-OHIP(M) increased as the respondents’ perceived

oral health status changed from good to poor (Table 5.26). It was also observed that there was a

significantly positive correlation between the L-OHIP(M) score (scale and all seven subscales)

and self-perceived oral health. This supported the construct validity whereby the greater the

impact on the quality of life, the poorer the perceived oral health status.

i4i

Tabie 5.26. Mean ADD-Scores and Spearman’s rank correlation coefficients for the L-OHIP(M)

Mean ADD-Score (SD)

Subscale Very good Good Fair Poor rho

(n=2) (n=103) (n=96) (n=6)

Functional Limitation 3.00 (4.24) 2.98 (3.60) 5.74 (3.97) 9.67 (1.21) 0.433*

Physical Pain 6.50 (9.79) 4.77 (3.57) 7.18 (3.87) 10.33 (1.75) 0.364*

Psychological Discomfort 5.00 (7.07) 3.53 (3.63) 6.91 (4.60) 12.17 (3.19) 0.443*

Physical Disability 6.00 (8.49) 3.21 (4.78) 6.28 (5.63) 12.50 (4.42) 0.379*

Psychological Disability 5.00 (7.07) 1.89 (2.89) 4.28 (4.15) 7.33 (4.08) 0.379*

Social Disability 2.50 (3.54) 0.83 (1.87) 1.67 (2.41) 3.17 (4.02) 0.239*

Handicap 4.00 (5.66) 4.40 (2.06) 3.34 (3.37) 5.33 (3.01) 0.361*

Scaie 32.00 (45.25) 18.61 (18.02) | 35.46 (23.37) | 60.50 (13.20) | 0.469*

* p-value < 0.001

A similar finding was also observed when the S-OHIP(M) was used. The score increased

as respondents’ perceived oral health status changed from good to poor (Table 5.27). This

significantly positive correlation indicates that there is a relationship between the two related

measures. This observation provides evidence of construct validity.

Table 5.27. Mean ADD-Scores and Spearman’s rank correlation

coefficients for the S-OHIP(M)

Perceived oral N Mean ADD-score rho

health status

Very good 11 2.27 (2.37)

Good 78 7.11 (5.27) 0.518*

Fair 107 13.56 (8.58)

Poor 7 27.86 (11.39)

p-value < 0.001

142

5.6.6.2) Discriminative Construct Validity

Table 5.28 shows the mean score of the L-OHIP(M) by global rating of perceived treatment

need and satisfaction with oral health. Those respondents who reported that they did not need

dental treatment and were satisfied with their oral health had a lower score than those who

perceived they required treatment and were not satisfied with their oral health. These differences

were statistically significant.

Table 5.28. L-OHIP mean score by perceived dental treatment need and satisfaction with orai

health.

Global rating N ADD Score SC Score Mean (SD) Mean (SD)

Perceived dental treatment need

Need treatment 135 33.45 (23.96) 3.01 (5.03) Do not need treatment 75 16.25 (16.37) 1.32 (2.89)

‘p= 0.000 "p= 0.001

Satisfaction with oral health

Yes 126 19.33 (17.70) 4.43 (3.32) No 84 39.08 (24.72) 3.83 (5.39)

"p= 0.000 "p= 0.000

'Mann-Whitney test.

Table 5.29 shows the S-OHIP(M) mean scores of perceived dental treatment need and

satisfaction with oral health. Respondents who perceived a need for dental treatment and did not

feel satisfied with their oral health had significantly higher S-OHIP(M) scores for both ADD

and SC scores.

143

Table 5.29. S-OHIP mean score by perceived dental treatment need and satisfaction with oral

health.

Global rating N ADD Score SC Score Mean (SD) Mean (SD)

Perceived dental! treatment need

Need treatment 443 13.06 ($.00) 1.14 (2.25) Do not need treatment 60 5.97 (5.60) 0.35 (0.86)

‘p= 0.000 ‘p= 0.000

Satisfaction with oral health Yes 118 7.31 (5.70) 0.45 (1.07) No 85 16.04 (9.71) 1.99 (2.62)

‘p= 0.000 ‘p= 0.000

Mann-Whitney test,

Table 5.30 shows the mean score by dental status of the L-OHIP(M) for the seven

subscales. However, for the S-OHIP(M), only the mean score for the scale was calculated. The

edentulous respondents had a higher score for both ADD and SC on the L-OHIP(M) scales than

the dentate with dentures and the dentate without dentures. On the other hand, the dentate

without dentures had the lowest score for both ADD and SC scores. However, these differences

were not statistically significant. When examined for each of the subscales, there was a

statistically significant difference in scores, both for ADD and SC, between the three groups on

the functional limitation and physical disability subscales.

When the S-OHIP(M) was used, those who were dentate with a denture had a

significantly higher mean ADD score than the edentulous and the dentate without a denture.

However, they had a lower SC score than the other two groups.

bri

JS] SUJO AA -JOYSNAN »

258°0 |

(Z9°Z) LLL

(Z€°1) 760

(60°Z) ZL'L

GZO'O |

(LE LL)

OOOL |

(ZZ) Py'EL

(98°8) 6P'OL

aj29S (

; (o9p=u)

( (09p=u)

pe=u pe=u)

enjea (g=u)

ainjuep

aunjuep eniea

(g=u) einjuep

sanjuep

~d, eyejuepy

UIA 9]e]U8Q

ou 9]e8]U8q

-d, eyejuepg

UPA 872]U8G

ou ajejueq

(IN)dIHO-S

(ag) e100s

9s ueeW

(as) e1095

Gay ues

6910 |

(L8°9) ELS

(pe's) 28°Z

(S9'&) £0°%

Zo

| (ozo)

ze°8e |

(9E°Sz) r8'0E

|

(6g°0z) L9°Sz

aje9g

geo |

(6G°0) £70

(¢¢°0) 910

(L¥'0) 80°0

2190 (Lay)

ebe (gL'€)

ee°c (GL°Z)

622 deolpuey

9so'0 |

(9S°0) 0z'0

(€€'0) 90'0

026'0 (6r'€)

£64 (16'L)

80°b (6L'Z)

LZL Ayiqesiq

jeloos

ZEL0 |

(90'L) eS°0

(vZ'0) 9L'0

(850) ZL'0

76c'0 (68°)

ZO'7 (€9'€)

GZ Z

(ye) SZ'e

Ayigesiq jea1Gojoyohsq

€00'0 |

(612) 09'F

(vs) 69°0

(L6°0) LE’O

000°0 (26°9)

196 (4S°9)

29°9 (eS'p)

Lee Ayigesig

jeaisAud

296°0 |

(LLL) 29°0

(Zr L)

620 (LE’L)

L9°0 oso

(L0°¢) 29°9

(L9°S) 29°S

(@'v) GL'S

yojwoosiq |ea1Gojoyohsq

Gz9'0 |

(ZL'L) 09°0

(160) ev'0

(98'0) LEO

Ocr'o (90'r)

Z9°S (19'€)

699 (€0'r)

LOS ule

jealshud

ozo'o |

(6r'b) 224

(60'L) 25°0

(92°0) Z€°0

200'0 (erg)

el 2

(eG'v) 8S

(9r'€) eg'€

uoneywry jeuoyoun4

(A) dIHO1

(0$}=u) (

) Ce)

(67=u) 6p=u

anjea (gp=u)

ainquep

eAnquep enjea

($L=u) einjuep

4nquep

ajeosqns

-d,. ayejuepa

UA

312]U0G ou

ajejueq -d,

ayejusps

UUM aje}UEG

ou ajejueqd

(as) 1098

DS UeaWy

(as) 81095

Gay ueeI

“snjyeys peyusp

Aq (IN)dIHO-S

24} Pue

(IN)dIHO-1 84}

JO Bs00S

UBAI] “OES

B12

145

5.6.7) Comparison of the technical properties of the L-OHIP(M) with the Australian

OHIP(A) and Canadian OHIP(C) data

To establish the measurement equivalence of the L-OHIP(M), a comparison of the technical

properties was made with the Australia data (OHIP(A)). In addition, a comparison was also

made with Canadian data (OHIP(C)). It has to be noted that in both the Australia and the

Canadian OHIP, the evaluation of the properties was done on an elderly population, whereas the

L-OHIP(M) was assessed in a population aged 18 years and above.

Table 5.31 compared the reliability (internal consistency and test-retest) of the L-

OHIP(M) with the OHIP(A) and the OHIP(C). Generally, the internal consistency of the L-

OHIP(M) was better than that of the OHIP(A), especially for the handicap domain. However,

the OHIP(A) had better internal consistency than the L-OHIP(M) for the pain domain. In

contrast, the OHIP(C) had better internal consistency for most of the subscales than the L-

OHIP(M). It has to be noted that the sample size of the Canadian sample was big. In terms of

stability, two subscales of the L-OHIP(M) had an ICC slightly lower than 0.7 compared to the

three subscales of the OHIP(A) that were lower than 0.5. This suggests that the L-OHIP(M) was

a more stable instrument than the OHIP(A).

146

Table 5.31. Comparison of the reliability of the L-OHIP(M) with the OHIP(A) and the OHIP(C)

L-OHIP(M) OHIP(A)' | OHIP(C)* Reliability 18-34 35+ 60+ 50+

(N=65) (N=149) (N=122) | (N=525)

internal Consistency (Cronbach's a ): Functional limitation 0.68 0.77 0.70 0.84

Physical pain 0.73 0.72 0.76 0.81

Psychological discomfort 0.82 0.84 0.77 0.88

Physical disability 0.80 0.88 0.82 0.88

Psychological disability 0.83 0.87 0.83 0.90 Social disability 0.76 0.82 0.73 0.88 Handicap 0.56 0.82 0.37 0.80

Test-retest:

Functional limitation 0.70 0.77 0.77 Not

Physical pain 0.73 0.83 0.42 available Psychological discomfort 0.74 0.64 0.76 Physical disability 0.83 0.66 0.72 Psychological disability 0.73 0.71 0.74 Social disability 0.85 0.73 0.08 Handicap 0.77 0.78 0.48

L-OHIP(M)=Malaysia, OHIP(A)=Australia, OHIP(C)=Canada.

Source: ‘Slade GD, Spencer AJ. Development and evaluation of the Oral Helath Impact Profile. Community

Dental Health 1994; 11:3-11.

’Secondary data from Ontario Study of the Oral Health of Older Adults database.

Construct validity of the L-OHIP(M) was compared with the OHIP(C) by assessing the

ability to distinguish between groups based on related self-reported measures (self-perceived

oral health status and self-perceived treatment needs). As shown in Table 5.32, differences in

mean ranks (DMR) obtained from Mann-Whitney tests indicated that both measurements

discriminate between groups; however, the OHIP(C) performed better in this regard. When the

odds ratio based on median splits was calculated, the L-OHIP(M) seemed to perform marginally

better than the OHIP(C). Since the database of the OHIP(A) could not be accessed, the same

analysis could not be performed on the OHIP(A).

Table 5.32. Comparison of the validity of the OHIP(M) with the OHIP(C)

147

Standardized ADD score Standardized SC score

Construct validity Median Median Malaysia Canada® Malaysia Canada*

Perceived oral health status

Very good/good 7.22 7.14 0.00 0.00 Fair/poor 18.89 16.84 4.44 2.04 p-value 0.000 <0.000 0.000 <0.0001 DMR 55 142 47 85 Odds ratio (good/poor) 5.4 5.2 7.7 2.5

Perceived dental treatment need Need treatment 14.44 13.77 2.22 2.04 Do not need treatment 5.56 7.14 0.00 0.00

p-value 0.000 0.000 0.001 0.000 DMR 52 400 27 64 Odds ratio (no/yes) 3.5 3.2 3.2 2.1

Satisfied with oral health

Yes 8.33 0.00 No 20.00 4.44 p-value 0.000 NA 0.000 NA

DMR 55 44 Odds ratio (yes/no) 49 6.6

p-values: Mann-Whitney tests DMR: difference in mean ranks between categories of grouping variables

Source: “Secondary data from Ontario Study of the Oral Health of Older Adults database.

5.6.8) Establishing functional equivalence

According to the model of equivalence (Herdman et al, 1998), functional equivalence is

achieved when all other types of equivalence in the model have been achieved. Therefore, it can

be claimed that functional equivalence has been achieved in this study since all aspects of

equivalence were established, as summarized below.

1) The semantic equivalence was ensured through a standard process of forward-backward

translation and expert panel assessment.

2) The qualitative interviews with the target population revealed that the way Malaysian

and Australian people conceptualized oral health and quality of life was similar in terms

3)

4)

5)

148

of how oral conditions affect people’s lives. The domains were the same even though

there were few other impacts identified.

Item equivalence was established by recognizing items which were not relevant to the

target population, removing some items which were causing some confusion, and adding

new items that were considered important for the target population.

Operational equivalence in terms of format of the questionnaire, response options and

method of administration were established. The format used has been shown to be

effective since most respondents either in the pretest or the field study returned a

completed questionnaire. The five-point response format can be used for the Malaysian

population. Since most people in Malaysia are literate, a mail questionnaire can be used;

however, the response rate could be compromised as with any other mail study.

Reliability and validity were reassessed and it was found that the L-OHIP(M) was

reliable and valid. It was then compared with the English version (the OHIP(A) and the

OHIP(C)), and it was revealed that the L-OHIP(M) performed equally well as the

OHIP(A) and the OHIP(C).

149

5.7) COMPARISON OF THE REGRESSION S-OHIP(M) WITH THE FREQUENCY

S-OHIP(M)

The original OHIP has a short form that was derived using a regression method. In the present

study, a short form was also developed based on the frequency method. Thus, in order to

compare these two methods, short forms based on regression and frequency were created

utilizing data from the L-OHIP(M), as described in the methodology. An evaluation in terms of

the content, score, reliability and validity of these two forms was conducted.

5.7.1) Content of the two S-OHIP(M)

Table 5.33 lists the 14 items for the regression and frequency forms of the S-OHIP(M). The two

short forms had four items in common: “had painful ulcer”, “felt shy”, “had difficulty carrying

out daily activities”, and “felt less confident of yourself’. The regression-S-OHIP(M) contained

more low frequency items than the frequency-S-OHIP(M). For the former, prevalence ranged

from 6.1 to 30.4 percent while the latter ranged from 4.2 to 55.1 percent. The frequency-S-

OHIP(M) contained seven items of more than 20 percent compared to the regression-S-

OHIP(M), which had only three items.

Table 5.33. ltem content and percent reported “very often”, “often” or “sometimes”

150

Subscale Regression-S-OHIP(M) Frequency-S-OHIP(M)

ltem % ltem %

Functional e Trouble pronouncing 12.6 Difficult chewing any 35.0 limitation wards. foods.

° Felt that foods you eat 30.4 Had bad breath cause | 33.2 have not digesied by dental problem properly.

Physical pain Had pain on the jaw. 16.4 Found it uncomfortable | 28.0 Had painful ulcer in 22.9 to eat any foods.

the mouth. Had painful ulcer in 22.9 the mouth.

Psychological ® Felt shy. 24.3 Felt shy 24.3 discomfort e Felt stressed up 15.9 Felt discomfort due to | 55.1

food stuck.

Physical disability | Been unable to eat your | 18.2 Had to avoid eating 28.5 favorite foods. some foods.

e Had an interrupt meals. | 18.2 Avoided smiling 13.6

Psychological e Been sad. 15.0 Your sleep been 16.8 disability ® Found it difficult to relax | 13.6 disturbed.

Your concentration 13.6 been affected.

Social disability ® Been less tolerant of 6.1 Avoided going out. 4.2 your spouse or family. Had difficulty 7.9

e Had difficulty carrying | 7.9 carrying out daily out daily activities. activities

Handicap e Felt unwell. 12.6 Had to spend a lot of 13.6 ® Feit less confident of 6.4 money.

yourself. Felt less confident of | 6.1 yourself.

Note: same items are bold

5.7.2) Score for both types of the S-OHIPCVY) forms

Table 5.34 shows the scores for both versions of the S-OHIP(M). The frequency-S-OHIP(M)

form had significantly higher scores than the regression-S-OHIP(M). The ADD score for

regression-S-OHIP(M) and frequency-S-OHIP(M) versions were 8.09 and 10.29, respectively.

151

Table 5.34. Mean, median and range of ADD score and SC score

Type of form ADD score SC2-score $C3-score

S-OHIP(M)- Regression Mean(SD) 8.09 (7.44) 0.59 (1.40) 2.20 (2.83) Median 6.0 0.0 1.0

Range 0-41.0 0-14 0-13

S-OHIP(M)- Frequency Mean(SD) 10.29 (7.55) 0.99 (1.56) 3.03 (2.99) Median 9.0 0.0 2.0 Range 0 - 36.0 0-710 0-13

p value* 0.000 0.000 0.000

ADD score=Additive score, SC2 Score=number of items reported as “very often and often”, SC3-score=number of

items reported as “very often, often and sometimes”

* Wilcoxon signed rank test

5.7.3) Technical properties for the two S-OHIP(M) forms.

The technical properties (reliability and validity) for both forms were tested and compared. It

was found that both forms performed equally well, as shown in Table 5.35. Both forms had a

Cronbach’s alpha of more than 0.8, which indicates excellent consistency. Both had similar ICC

values of above 0.7, indicating that both were stable measures.

The scores of both of the short versions were highly correlated with the score of the L-

OHIP(M). This indicates that both of them were able to measure equally well the same construct

as the long version. Construct validity was assessed for both of the S-OHIP(M) forms, showing

that both of them followed the hypothesis as postulated, except in discriminating between

people in terms of their dental status (dentate and edentate). The differences in mean rank,

obtained from the Mann-Whitney test, indicated that the frequency-S-OHIP(M) was marginally

better. Odds ratios based on the median splits also indicated that the frequency-S-OHIP(M) was

marginally better than the regression-S-OHIP(M).

152

Table 5.35. Comparison of the technical properties of the two forms of the S-OHIP(M)

Property §-OHIP(M) - S-OHIP(M) - Regression Frequency

RELIABILITY: Internal consistency 0.89 0.86 Test-retest 0.76 0.77

VALIDITY: Concurrent validity:

Correlation between S-OHIP(M) ADD rho = 0.965 rho = 0.964 Score with L-OHIP(M)45 score *p=0.000 *p=0.000

Convergent vaiidity: Perceived oral health status

Very good (n=2) 12.00 (16.97) 8.00 (11.31) Good (n=103) 5.08 (5.58) 7.00 (6.08) Fair (n=96) 10.46 (7.87) 13.21 (7.56) Poor (n=6) 18.50 (4.59) 20.17 (4.96)

rho= 0.440,*p=0.000 rho=0.491,*p=0.000 DMR* 51 57 Odds ratio 6.0 5.1

Discriminative validity: Perceived dental treatment need

Do not treatment (n=75) 4.41 (5.22) 6.36 (5.59) Need treatment (n=135) 40.06 (7.77) 12.44 (7.74)

‘p=0.000 ‘p=0.000 DMR 54 53 Odds ratio 3.5 4.1

Satisfaction with oral heaith

Yes (n=126) 5.56 (5.95) 7.44 (5.92) No (n=84) 11.65 (7.91) 14.38 (7.78)

‘p=0.000 ‘p=0.000 DMR 54 57 Odds ratio 46 5.6

Dental Status Deniate no denture 7.63 (6.95) 9.77 (7.10) Dentate with denture 8.43 (7.82) 10.98 (7.89) Edentate 11.60 (10.10) 13.20 (10.20)

*p =0.270 *p=0.389 DMR* 10 11 Odds ratio 1.18 1.48

* Snearman’s correlation, ‘Mann-Whitney test ,’Kruskal-Wallis test

# Categories were pooled to very good/good vs fair/poor

£ Categories were pooled to Dentate no denture vs dentate with denture/edentate

§ 95% confident interval includes 1

DMR-differences in mean rank.

153

5.8) THE IMPACT OF ORAL CONDITIONS ON THE QUALITY OF LIFE OF A

MALAYSIAN ADULT POPULATION.

The data of the L-OHIP(M) was used to described the impact of oral conditions on the quality of

life of the adult population of Malaysia and to explore variation by age group, gender, ethnicity,

level of education, and dental status.

Overall, slightly more than fifty percent of the sample had at least one impact reported as

either “very often” or “often” (Table 5.36). This suggests that a significant number of people in

this study experienced the impact on a relatively frequent basis. More than one third of the

sample frequently experienced some form of psychological discomfort due to poor oral health.

More than one quarter were bothered by the functional consequences of oral disorders and oral

pain.

Table 5.36. Percent responding “very often” or “often” to one

or more items in each subscale.

Subscale % Functional limitation 27.1

Physical pain 22.9 Psychological discomfort 37.4 Physical disability 18.7

Psychological disability 12.1

Social disability 3.7 Handicap 7A

Scale 52.8

Table 5.37 shows the response to each of the 45 OHIP items organized into seven

subscales. The response categories were reduced to three: 1) “very often” or “often”, 2)

“sometimes”, and 3) “once in a while” or “never”.

154

Table 5.37. Response to OHIP(M)

Very Sometimes Once ina often/often while/Never

“Functional. Limitation OE a iis Oe ma Difficult chewing any foods 12.1 65.0 Trouble pronouncing words 1.9 10.7 87.4 Felt that appearance has been affected 4.2 12.6 83.2 Had bad breath cause by dental problem 10.7 22.4 66.8 Felt that foods you eat have not digested properly. 12.1 18.2 69.6 Felt denture was loose 7.0 _ 5.1 87.9

“Physical Pain 0 0 ee ee EE ES a

Had pain on the jaw 3.3 13.1 83.6 Had headache due to dental problem 2.8 12.6 84.6 Had sensitive teeth 11.2 26.6 62.1 Had toothache 47 23.8 71.5 Had painful gums 3.7 16.8 79.4 Found it uncomfortable to eat any foods 9.3 18.7 72.0 Had painful ulcer in the mouth 4.7 — 18.2

“Psychological Discomfort . « Me BR SAR ah URS Felt uncomfortable with your dentures 4.7 4.2 91.1 Felt worried 16.8 21.0 62.1 Felt discomfort due to food stuck. 30.4 24.8 44.9 Felt shy 6.1 18.2 75.7 Felt uncomfortable with your appearance 6.5 14.5 79.0 Felt stressed up 5.1, 10.7 84.1 Physical Disability Lo. So fees Speech been unclear 4.2 5.6 90.2

People misunderstood some of your words 1.9 6.5 91.6 Felt food less tasty 4.2 10.7 85.0 Been unable to brush your teeth properly 7.5 10.3 82.2 Had to avoid eating some foods 10.7 17.8 715 Been unable to eat your favorite foods 5.6 12.6 81.8 Been unable to eat with your dentures 4.2 3.3 92.5 Avoided smiling A7 8.9 86.4 Had an interrupt meals . - 6.1 _ 12.4 _ 81.8 Psychological Disability ee ' Ee CO MS ge 8 a UIE ag Your sleep been disturbed 1.4 15.4 83.2

Been sad 7.0 7.9 85.0 Found it difficult to relax 2.3 11.2 86.4 Felt depressed 2.3 8.4 89.3 Your concentration been affected 1.9 11.7 86.4 Felt a loss of appetite to eat 47 13.6 818° Social Disability Ce SOR Le ae ei Tg Avoided going out 0.9 3.3 95.8 Been less tolerant of your spouse or family 0.5 5.6 93.9 Unable to mix around with other people 1.4 4,2 94.4

Got angry easily. 1.9 5.1 93.0 Had difficulty carrying out daily activities 0.9 7.0 92.1

. Handicap no CE epn Rs Ty SNS oe Ee Uta gE Felt unwell 2.3 10.3 87.4 Had to spend a lot of money 3.7 9.8 86.4 Felt less happy to be in the company of others 1.4 7.3 90.7

Felt that life in general was less satisfying 1.4 7.9 90.7

Felt less confident of yourself 14 47 93.9 Been unable to work fo your full capacity 0.9 6.4 93.0

155

Less than ten percent of the respondents responded “very often” or “often” for most of

the items except “felt discomfort due to food stuck” (30.4%), “felt worried” (16.8%), “difficulty

chewing any food” (12.1%), “felt food has not digested properly” (12.1%), “had sensitive teeth”

(11.2%), “had bad breath” (10.7%) and “avoid eating certain food” (10.7%). Of these seven

items, three belong to the functional limitation, two to the psychological discomfort and one

each to the pain and the physical disability scales.

Table 5.38 shows the percentage of respondents who answered “very often” or “often” to

one or more items in each subscale, as well the overall scores by sociodemographic

characteristics. The middle agegroup (40-59 years old) reported more impact than the younger

agegroup (18-39 years old) and older agegroup (60+ years old). More than one third of the

middle and older adults were bothered with the functional consequences of oral conditions,

compared to approximately one sixth of the younger adults. A higher proportion of middle and

older adults reported having more physical and psychological disabilities due to oral conditions

than younger adults. The younger and middle adults were more likely to have experienced some

form of psychological discomfort due to poor oral health. However, it was observed that there

was no big difference in terms of oral pain, social disability and handicap between age groups.

Both genders experienced the same amount of impact on their quality of life due to oral

conditions. When comparing among the three main ethnic groups in Malaysia, the results show

that the quality of life of Indians was most affected by poor oral health. Slightly more than

seventy percent of Indians reported having at least one of the impacts on a frequent basis. More

than fifty percent of Indians experienced some psychological discomfort due to oral conditions

as compared to only 25 percent of Malays and Chinese. Generally, Malays and Chinese had

similar impacts on all dimensions of quality of life. However, Chinese respondents experienced

more handicaps due to poor oral condition than the Malay respondents.

156

Those who had tertiary education (college or higher) reported more impacts than those

who had a lower level of education (secondary and lower). It was observed that more than one

third of the respondents having a secondary or higher education experienced some psychological

discomfort.

Almost seventy percent of the edentate respondents reported that they had at least one

impact “very often” or “often”. About one half of the edentulous respondents experienced

problems “very often” or “often” in the areas of functional limitation and physical disability.

LSI

(GL=U) ayeyuepy

: (6p=u)

299 0°02

eel 19%

L6y eee

6'te ee

@INJUSP UM

a}e]U8q O'S

ZOL 00

Zs vt

L9€ G9%

G92 (OS$L=u)

0%

eg Ov

OZ Lvl

o8e Le

Lye SiNjUSp

OU e}e]UEq

:snjeys je}ueq

9°8S 62

6z 6S

bt Ze

LL Geez

(pe=U) seyBiy

gy eBayjop ELS

v6 9%

Let SLL

Lav v2

Le? (ZL L=u)

Arepuooceg ves

€9 €9

Leb g'eZ

0'2z 9°8z

G'9e (€g=U)

JeMo] 9g A

le

UOHEONP

JO jeAa"]

9°S9 00

00 00

0'Sz Gle

GZL 0°SZ

(g=u) 4810

Oe

091 O'9L

0'8z oze

09S ozE

O'9€ (¢Z=u)

uelpuy eer

€ OL 0'0

98 0'6L

Gre v2

O6L (gg=U)

eseulyD v0g

6

ee vt

vSh ose

O27 £62

(eZ) =u) Aejey)

ou

zs G8

ve Z OL

6 OL O'6e

ZL

0'8z (QL

)=u) sjewey

Les €9

ov QFE

802 pse

0'Sz 0'92

(96=U) 9/eIN

i -depues)

© 8r 69

ve SEL

92Z L102

L0Z ole

(6z=U)+09 v'9S

vl Ze

6'r1L or

Gly Llé

O've (v6=U)6S-OP

g'0g Ll

VV 88

Z El G'8e

ZL

GOL (L6=U)

6E-81L :dnowbaby

Ayyiqesip Ayiqesip

Aynqesip HOJWUODSIP

ujed UOHe HUT

snjejs je}U8q

ayeos desipuey

jelsos jeaiBojoyoAsyg

jeaisAud jearBojoysAsy

= jesisAug

jeuonoung

jaiudesGowapols0sg

snjels

je}usp pue

soiyesBowapolsos

Aq sajeasqns

yore ul

suey! SJOLWU

JO BUO

O} , UdYO,,

JO ,,Uayo

Aiaa,, Buipuodsas

Jusdied "ge's

aiqeL

158

5.9) COMPARISON OF THE ORAL HEALTH-RELATED QUALITY OF LIFE

BETWEEN THE MALAYSIAN POPULATION AND THE CANADIAN

POPULATION AGED 50 AND ABOVE.

Table 5.39 shows the background of the Malaysian and Canadian data. The Malaysian data were

collected in 2002, whereas the Canadian data were collected in 1991 and 1992. For the purpose

of this comparison, the age of the respondents was standardized to 50 years old and above. The

total sample size for the Malaysian sample was 56 and for the Canadian sample was 519. The

Malaysian sample had a higher proportion of male respondents compared to the Canadian

sample. Both samples had a higher proportion of dentate respondents than edentulous.

Table 5.39. Background of the Malaysian and Canadian data

Malaysia Canada

Year of data collection 2002 1991/92

Total sample size 56 519

Gender

Male 29 (51.8) 227 (43.7)

Female 27 (48.2) 292 (56.3)

Dental status

Dentate 44 (78.6) 428 (82.5)

Edentulous 12 1.4) 91 (17.5)

(percent)

Figure 5.2 compares the impacts of oral health on the quality of life between a Malaysian

population and a Canadian population aged 50 and above. Overall, the Malaysian sample had a

slightly higher impact compared to the Canadian sample. Generally, the Malaysian sample had

higher impacts on all dimensions of quality of life than the Canadian sample. Functional

consequences were most affected by the oral health conditions for both samples. Nevertheless, It

159

was observed that physical aspects were more affected than social aspects of life in both

samples.

60 “%o

f Malaysia 50

Canada

EE PSUS BES ie SS Functional Psychological Physical Psychological Social Handicap Scale

Limitation discomfort disability disability disability Score

Figure 5.2. Comparison of the percent responding “very often” or “often”

to one or more items in each subscale between the Malaysian sample and the Canadian sample

Table 5.40 compares the oral health impact on the quality of life between the two

populations by gender. Generally, both genders in the Malaysian sample reported more impact

than in the Canadian sample. The male respondents had a slightly higher impact than the female

respondents in the Malaysian sample, whereas in the Canadian sample the female respondents

reported slightly more impacts than the male respondents. The female respondents of the

Canadian sample reported higher impacts than the male respondents on all subscales except

functional limitation. In the Malaysian sample, the male respondents reported higher impacts on

pain, psychological discomfort, and psychological disability, whereas the female respondents

160

reported more impacts on functional limitation, physical disability, social disability, and

handicap.

Table 5.40. Comparison of the percent responding “very often” or “often” to one

or more items in each subscale between the Malaysian sample and the Canadian

sample by gender

Malaysia Canada

Subscales Male Female Male Female

N=29 N=27 N=227 N=292

% % % %

Functional limitation 37.9 44.4 38.8 38.0

Physical pain 27.6 18.5 15.0 19.9

Psychological discomfort 44.8 25.9 10.1 14.0

Physical disability 27.6 29.6 9.7 10.6

Psychological disability 24.1 11.1 3.4 5.8

Social disability 3.4 3.7 1.8 2.1

Handicap 6.9 14.8 3.5 7.2

Scale 55.2 51.9 44.1 45.5

Table 5.41 compares the impact of oral health on the quality of life between the two

populations by dental status. Overall, edentulous respondents in both populations had a higher

impact compared to dentate respondents. It was also observed that the edentulous respondents

had a higher score on almost all domains of quality of life for both samples.

161

Table 5.41. Comparison of the percent responding “very often” or “often” to one

or more items in each subscale between the Malaysian sample and the Canadian

sample by dental status

Malaysia Canada

Subscales Dentate | Edentulous | Dentate | Edentulous

N=44 N=12 N=428 N=91

% % % %

Functional limitation 36.4 58.3 37.4 42.9

Physical pain 22.7 25.0 15.9 26.4

Psychological discomfort 36.4 33.3 11.2 17.6

Physical disability 25.0 41.7 9.7 10.6

Psychological disability 15.9 25.0 3.0 12.1

Social disability 2.3 8.3 0.9 6.6

Handicap 6.8 25.0 4.9 8.8

Scale 50.0 66.7 43.7 50.5

CHAPTER 6

DISCUSSION AND CONCLUSION

The main aim of this project was to develop an OHRQoL measure for a Malaysian adult

population aged 18 and above. The reason for choosing 18 years old as the cut-off point is

because 18 year olds have completed secondary school and are beginning to enter work force or

the higher education system. Thus, they are no longer under the systematic school oral health

care program, which means that they have to obtain oral health care themselves.

Since there are a number of OHRQoL measures that have been developed and tested, the

investigator decided to adapt one of those measures rather than to develop a new one. This

would not only facilitate the development of the measure, but also reduce the need for

developing new measures that have the same purpose. In addition, it would also allow some

direct comparison to be made across cultures.

The review of the OHRQoL measures revealed that the Oral Health Impact Profile

(OHIP) is the most sophisticated measure currently available. Since the items were selected

from an interview with the patients, the items included in the instrument could be considered to

be important to the patient. As such, to a certain extent, items included in this measure not only

capture aspects of life that have been affected but also the importance of these aspects to the

respondents. The OHIP’s psychometric properties have been assessed in several countries. In

addition, it covers a broad range of oral health-related quality of life domains. Thus, the OHIP

was the most suitable measure to be adapted for the Malaysian population.

According to Hunt (1995), if the adaptation of a measure to another language or culture

is performed with no intention to carry out a cross-cultural comparison, then the original

measure is used merely as a template or guide for the production of a culturally appropriate set

162

163

of items. However, if the intention of adaptation is to compare or combine data cross-culturally,

then it is fraught with difficulties of a conceptual, technical and ethical nature. Although the

main aim of this project was not intended to carry out cross-cultural comparison, it is desirable

to be able to do so. Therefore, proper adaptation was carried out based on the framework

proposed by Herdman et al (1998). Herdman et al’s framework provides the clearest way to

adapt a measure. The concept of equivalence is a fundamental aspect in cross-cultural

adaptation, as discussed in the literature review (page 71). Herdman’s model proposes that six

types of equivalence need to be assessed. Thus, to establish these equivalences, a multistage

study was designed as described in the methodology section of this thesis.

6.1) STUDY LIMITATIONS

As in any study, the present study has several limitations. Thus, interpretation of the data was

based on these limitations.

Translation and adaptation: Some researchers recommended that the translators translate to

their mother tongue (Beaton et al, 2000; Guillemin et al, 1993) and be familiar with both

cultures (Carlson, 2000; Leplege and Verdier, 1995; Geisinger, 1994). However, it was

extremely difficult to find translators who meet these criteria for this study. Nevertheless, the

involvement of the evaluators - who are English-speaking people - in the process of translation

increased the confidence in the quality of the translation.

Another limitation in the process of adaptation in this study was that the allocation of the

two methods of administration of the questionnaire was not random. Ideally, a crossover study

design would have been more appropriate. Thus, any comparison between these two modes of

administration must be interpreted within this limitation.

164

Comparison between two short forms: Ideally, to compare two different types of measure, these

measures need to be administered to the same sample independently. In this study we compared

two different forms of short version of OHIP(M). However, we utilized an existing long form of

OHIP(M) data to create and compare these two measures. As such, this comparison entails some

limitations.

Comparison between two cultures: The data used to compare the impact of oral conditions

between Malaysian and Canadian populations were obtained from two separate studies

conducted independently at different time using different methods. Thus, the comparison must

be treated with caution.

6.2) TRANSLATION

The decision to translate OHIP into Malay language was made because the Malay language is

the national language of Malaysia. Thus, it was assumed that all Malaysians understand the

language. However, it must be noted that the Malay language itself is not homogenous in nature.

Although there is a standard Malay language, there are several Malay dialects spoken

throughout Malaysia. For example, the word “segan” has inconsistencies in meaning from

dialect to dialect, as discussed in the result section (see page 114). Hence, one of the problems in

the process of translation of the OHIP into Malay is to ensure that no particular dialect word is

used, but at the same time to make sure that all words will be understood by all people.

Although a dictionary is a valuable resource, its language is often not the language of the people

(Sechrest et al, 1972). For example, the word “sogok” is an equivalent word for “self conscious”

given by a dictionary but its use is very limited among the people in Malaysia. Thus, such words

should be avoided.

165

There are several ways to translate an existing HRQoL measure as discussed in the

literature review (see page 78). In this study, a forward-backward translation technique was

applied since this technique had been widely used and had been empirically proven to be

effective in achieving semantic equivalence (Kim and Lim, 1999). In this study, such a process

allowed the original developer and other evaluator, whom are English speakers, to be directly

involved in the process of ensuring that the translation achieved semantic equivalence. As such,

there is confidence that the translation achieved an acceptable semantic equivalence.

There were some difficulties encountered in the process of translation, as discussed in

the result section (see page 113). Some of these difficulties could be due to the way in which the

original questionnaire was written. According to the developer of the original questionnaire, the

questions were written based on the patients’ own language (communication via email). Hence,

some of the items used colloquial words (for example “sore jaw”), which were difficult to

translate into Malay language. One of the suggestions put forward by Brislin et al (1973) to help

in writing translatable English is to avoid metaphors and colloquialisms. Perhaps the original

questionnaire was developed with no expectation that it would be translated to other languages.

Because the focus of translation is to transfer meaning from the source language to the target

language, the translator first needs to understand the meaning of the statement before it can be

translated into the target language (Esposito, 2001). Thus techniques such as “‘decentering” and

“explain around” were used (Brislin, 1973) and have been shown to be helpful.

It is evident that the socio-emotional items were the most difficult items to translate. In

some cases, two different levels of emotional states in the English language were translated into

the same word in the Malay language, for example, “miserable” and “upset”. Limited

vocabularies for describing emotional states reflect the culture of the Malay society, which

could be considered to be a collectivist society. A similar situation was also observed by Hunt et

166

al (1986). Hunt stated that “the socio-emotional items are likely to cause the most trouble and

items as what are perhaps more universal experiences, like pain, sleep and impaired physical

mobility, the least.” Nevertheless, most of the items in the original questionnaire were

considered by the translators to be easy to translate.

6.3) CULTURAL ADAPTATION

Translation alone is not sufficient to ensure the validity of the measurement, since culture has a

significant impact on the way concepts are expressed. Thus, the measure also needs to be

adapted to the culture and lifestyle of the target population, as discussed in the literature review

(page 82).

In this study, qualitative interviews with patients were conducted to assess conceptual

and items equivalence. A purposive sample was used. It has to be noted that the purpose of

sampling was not to get a representative sample. Rather, it was based on getting a broad view on

how oral diseases or disorders affect people’s lives across age groups and ethnicity in Malaysia.

Although Malaysian society is heterogeneous with people of different cultures and religious

backgrounds, they have integrated into the so called “Malaysian” culture. Thus, language in

particular is no longer a barrier between different ethnic groups in Malaysia (Jaafar, 1995). As

such, language was not a barrier in the interviews.

From the interview, we found that most of the impacts of oral diseases or disorders were

similar across ethnicity in Malaysia’s population. We also observed that some impacts were

more commonly reported by the younger adults than the older adults, for example, “self-

confident”.

It was concluded that the way Malaysians and Australians conceptualize oral health was

similar in terms of the impact of oral disease on daily activities. It has to be noted that the items

167

of the original questionnaire were selected from a pool of items, derived from interviews with an

elderly population. Thus, items such as “self confident”, which were found to be important

among the younger adults, were not included in the original questionnaire. Since this adapted

questionnaire will be used in an adult population (18 years old and above), such items were

included.

In order to make the content culturally sensitive to the target population some items were

deleted, rephrased or added, as discussed in the result section (page 122). However, most of the

items were retained because of their universality, at least for these two populations (Malaysian

and Australian), as shown from the results of the qualitative study. A total of 45 questions were

finally included in the OHIP(M), compared to 49 questions in the original questionnaire.

Because of the cost incurred in administering a long questionnaire, most measures develop a

shorter version. Thus, based on the qualitative interviews, a short version was also developed,

containing only 14 items.

It is recommended to pretest adapted questionnaires to assess the face validity. This

ensures that the level of language used is appropriate to the target population (Leplége and

Verdier, 1995). In addition, it also enables investigators to establish whether or not respondents

understand the concept or task (Collins, 2003). A convenience sample of twenty participants

was used to assess face validity for this study. Aday (1996) mentioned that a nonprobability

sample is appropriate to be used in the early stages of designing a standardized survey. In this

study, a small sample size was used at this stage because it was qualitative in nature. This pre-

testing demonstrated that the format, instruction and language used were appropriate across

ethnicity and agegroup for the target population.

One important aspect of a standardized questionnaire is the response format. There are

several types of response format available, such as Visual Analog Scale, Likert Scale, etc. Some

168

of them are too complicated to be used in some population groups. Thus, it is important to

assess the suitability of the response format for the target population. A non-appropriate

response format will result in either a loss of sensitivity or will create a “noise” on the

instrument, causing a response-bias. In this study, the suitability of the five-point frequency

Likert scale response format was assessed. Because Malaysian people are very moderate when

providing responses, it is necessary to test whether or not they use the entire spectrum of

options. However, in this study it was observed that the response format could be used in the

Malaysian adult population. Perhaps, the questions in the questionnaire were not threatening:

there is evidence that the way that people respond to a question will depend on how threatening

they perceive the question to be (Aday, 1996).

Field testing was conducted in order to assess the suitability of the mode of

administration and to re-establish the psychometric properties of the adapted OHIP. Because the

intention was to use this measure on the general Malaysian population, a population-based

sample was used to test its psychometric properties. With the time and resource constraints, the

sample was limited only to one state of Malaysia, namely Selangor. Nevertheless, the sample

was selected based on probability sampling. Thus, the results could be generalized to Selangor

and, to a certain extent, to Peninsular Malaysia. However, it could not be generalized to the East

Malaysia population because their culture and lifestyle is quite different from the population of

Peninsular Malaysia. In addition, preparation of the questionnaire was based only on the main

ethnic groups of Peninsular Malaysia, namely Malay, Chinese and Indian. The main ethnic

group of East Malaysia was not included.

As mentioned in the methodology, the sample for this stage was obtained from the

NOHSA, where a complex design was used; this has resulted in what is known as “design

effect”. To overcome this problem, the Stata statistical program was used to deal with the design

169

effect. Nevertheless, according to Dr. Slade’s opinion (personal communication), for the

purpose of testing reliability and validity, a complex design could be treated as a simple random

sample since these tests only show correlations or associations existing within the sample - not

within the population. However, this effect becomes important and has to be taken into account

when estimating the means or proportion of the population.

The question of how this questionnaire was going to be administered was re-addressed.

There are several methods of administering questionnaires. The methods commonly used are

face-to-face interviews, telephone interviews and by mail. Computer administered

questionnaires have recently become popular. Most researchers are concerned with obtaining a

high response rate because it is an indirect indication of the extent of a non-response bias. It has

been shown that some methods of administration are superior to others in terms of getting a high

response rate; however, the cost is higher.

In this study, only two methods of administration were assessed: a face-to-face interview

and a mail questionnaire. Although the original OHIP could be administered through a

telephone interview, it is not appropriate for use in Malaysia at this time because not all

households have a telephone. Since the original questionnaire was not designed for computer-

assisted administration, this method was not assessed in this study. To assess this, the sample

was divided into two groups: mail and interviews. However, it was not randomly allocated due

to administrative reasons as mentioned earlier. As such, the comparison between these two

modes of administration may be biased. Thus, interpretation of the results was made within this

limitation.

In most surveys, mail is preferred over interviews and other methods mainly because it is

cheaper. However, mail administration is not appropriate when there is a low literacy rate. In the

case of the Malaysian population, statistics have shown that the literacy rate is high (see

170

introduction page 6). Despite the high literacy rate in Malaysia, the response rate for mail

questionnaire in this study was low (slightly lower than fifty percent) even when respondents

were given two reminders. Perhaps the response rate could have been improved by increasing

the number of reminders. However, Locker and Miller (1994a) concluded that the four-wave

design tends to be inefficient, with relatively few respondents responding to the fourth mailing.

It also has to be noted that a low response rate could have been due partly to using a sample

which had been used by several other studies prior to the present study. Furthermore, due to

financial constraints this study did not offer any incentives for participation. Another reason

could be due to using only a Malay language questionnaire. This could have led to a lower

response rate among the non-Malays. Thus, perhaps a multilingual questionnaire is needed to

increase the response rate among the non-Malay population, especially if a self-completed

questionnaire is administered.

On the other hand, the interview yielded better response rate, as expected. However, a

face-to-face interview is time-consuming and costly. In this study, the process of finding the

homes of the respondents was very time consuming, since the map was not very reliable.

Furthermore, the interview could only be done on weekends because most of the respondents

work on weekdays. In addition, the interview needed to be done at a very specific time in order

to obtain a better response. For example, most people have lunch from 12:00 — 2:00 pm and do

not like to be disturbed, and after 2:00 pm most families take an afternoon nap. Thus, perhaps

the most appropriate time for interviews is from 10:00 am to 12:00 pm and after 5:00pm.

About half of the non-respondents refused to be interviewed. The main reason was

because they had been involved in many studies prior to the present study. Slightly more than

half of the non-respondents could not be interviewed because they were not available. Most of

171

them did not respond to the appointment card left at their house. This could have indicated that

they were not interested in participating in the study.

Incomplete data is another source of non-response bias (Streiner and Norman, 1995). In

this study, the questionnaire was considered to be unusable when more than twenty percent of

the items were left blank or had “don’t know” responses. As anticipated, the mail questionnaire

had a higher percentage of incomplete data than the interview questionnaire both for the L-

OHIP(M) and the S-OHIP(M). It was also observed that the length of the questionnaire

influenced the completeness of data.

Another parameter used to compare the two modes of administration was the OHIP(M)

score. It was observed that the mail questionnaire had a slightly higher score than the interview;

however, the differences were not significant. A crossover study conducted by Slade et al (1992)

examined the effects of data collection methods for a self-reported instrument and concluded

that “mail questionnaire responses to oral health impact questions are less prone to subject

acquiescence or response bias.”

Therefore, potential users of the OHIP(M) have to decide which method they want to

employ. On one hand, the mail questionnaire compromised the response rate. On the other hand,

it reduced the response-bias as well as the cost and time, compared to interview. It is also

noteworthy that it is much easier to deal with a non-response bias than a response bias. Thus,

potential users have to draw a line between response rate, cost and response bias. It is also

recommended that only one method of administration should be used.

It was important to re-establish the psychometric properties of the adapted measures

since some changes had been made. Moreover, reliability and validity are situation-dependent.

Thus, the reliability and validity of the L-OHIP(M) were reestablished and the S-OHIP(M) was

tested as presented in the results section. In general, the findings suggest that both versions of

172

the OHIP(M) demonstrated good reliability and validity. A detail of the S-OHIP(M)

psychometric properties will be discussed in the next section.

Although in psychometrics testing a value of 0.7 is considered to be an acceptable level

for assessing homogeneity, this study used a slightly lower value of 0.6 to indicate good

reliability. This is because OHIP is an outcome measure rather than a measure of traits such as

intelligence or anxiety, whereby the subscales of this measure contain more than one concept.

Thus, the items are not expected to be homogenous. It has also been suggested that there is no

need for more than a moderate internal consistency for such measurements (Fitzpatrick et al.,

1998). This study suggests that the L-OHIP(M) had a good internal consistency for all subscales

and it was slightly better than the OHIP(A).

Test-retest reliability is a necessary property for discriminative patient-based outcome

measures. Estimates of the test-retest reliability of the L-OHIP(M) are slightly better than the

OHIP(A). However, it must be noted that these differences could be due to: 1) the differences in

the population studied (adult population aged 18+ for the present study and the elderly in the

Australian study), and 2) could be due to the time between the administration of the

questionnaire (two weeks for the present study and three months for the Australian study).

However, in most instances, two weeks is a sufficient time frame, especially for a long

questionnaire to test for stability.

Reliability testing alone is not sufficient to establish the usefulness of a measure

(Streiner and Norman, 1995). This is because reliability testing only tells us about the

reproducibility of the measure but does not ensure that it measures what it was intended to

measure - in other words, the scale’s validity. Since the L-OHIP(M) scores obtained from mail

and interviews were not significantly different, the data were combined for validity analysis.

This was done to increase the number of respondents in each category of independent variables.

173

Since there was no “gold standard”, construct validity was assessed. As expected, it was

observed that the poorer the perception of oral health status, the higher the score of the L-

OHIP(M) and that they were significantly correlated. In addition, the Malaysian version of the

OHIP was also able to discriminate between groups. These findings provide evidence to support

the validity of the L-OHIP(M).

Locker and Slade (1994) found that there were very weak correlations between clinical

measures and the OHIP score. This is consistent with the conceptual model used to develop the

OHIP, which suggests that functional and psychosocial impacts of disease are modified by

sociodemographic and other variables. Therefore, no attempt was made to correlate or associate

clinical indicators with the OHIP(M), except for dental status (dentate and edentate). The reason

for using dental status to validate the measure is because studies have found that the edentulous

reported more impact than the dentate especially in terms of functioning (Slade and Spencer,

1994(b)). Moreover, this data can be easily collected using a questionnaire. As expected,

edentulous and dentate with dentures had significantly higher scores for functional limitation

and physical disability subscales. Although the overall score did not show significant

differences, there was a trend that those who were edentulous and dentate with a denture had a

higher overall score than dentate without a denture. However, further analysis looking at the

association between the numbers of teeth retained and the score is needed to obtain a true

picture of the situation.

This study demonstrated that the Malaysian version of the OHIP works equally well as

the English version. Hence, it can be concluded that it has achieved functional equivalence as

one of the criteria in cross-cultural adaptation proposed by Herdman, et al (1998).

174

6.4) RELIABILITY AND VALIDITY OF THE S-OHIP(M)

The short form of the OHIP(M) was developed and it is designated as S-OHIP(M). As

mentioned in the methodology (page 108), the main reason for developing a short form is to

provide an alternative, cheaper and less time-consuming means of data collection based on the

premise that a long questionnaire is not permitted to be used in some research settings and

clinical practices even though it provides more comprehensive data. For example, a measure

that takes a long time to complete and score may not be feasible in a clinical setting because of

the burden placed on patients or clinicians (Locker and Allen, 2002b).

The S-OHIP(M) was developed simultaneously with the adaptation of the long version

of the OHIP where the reliability and validity of the adapted L-OHIP(M) were not yet

established. Thus, the L-OHIP(M) cannot be considered as a gold standard (Coste et al, 1997).

Consequently, the expert-based approach, where patients were the experts, was employed in this

study. The selection of the items for the S-OHIP(M) was made based on an assumption that the

items were considered as important when they were frequently reported by the patients

interviewed in the qualitative interview stage. As such, it reflects the patients’ collective opinion

of the importance of the items selected. In this way, the method could be considered as an

“expert-based’ approach. Since a large proportion of items were deleted from the L-OHIP(M),

the measurement properties were not the same as the L-OHIP(M). Hence, the S-OHIP(M) was

treated as a new measure and the reliability and validity were tested on the new and independent

sample of the target population as recommended by Coste, et al (1997).

This study demonstrated that the S-OHIP(M) is reliable, as shown by the ICC and

Cronbach’s alpha. The ICC of more than 0.9 indicated that the measure is stable. Similar ICC

values were observed for mail and interview questionnaires. This suggests that the S-OHIPCM)

175

is stable regardless of how it is administered. A high Cronbach’s alpha value suggests that the

items are homogenous in terms of measuring oral health-related quality of life.

The way that the items were selected to create the S-OHIP(M), where the two most

frequent items reported from each domain were selected, ensures the content validity in terms of

coverage. The results presented on page 141 demonstrate that the S-OHIP(M) was valid.

Although the S-OHIP(M) was developed to be used in surveys as a descriptive and

discriminative measure, it is also desired to be used in clinical trials and in clinical practice as an

evaluative measure. However, in this study, the responsiveness of this measure was not tested

due to time and resources constraints. Therefore, the responsiveness will be tested in future

research.

6.5) COMPARISON FREQUENCY S-OHIP(M) WITH REGRESSION S-OHIP(™)

Basically, there are three ways in which a measure can be shortened: 1) statistical approach

(regression, internal consistency or factor analysis), 2) expert-based approach, and 3) item-

impact method (Locker and Allen 2002b). The original OHIP was shortened using the

regression statistical approach (Slade, 1997c), whereas the Malaysian short version was

developed based on the most frequent items reported by patients, as discussed earlier. The

question as to whether these two approaches yielded different sets of measures was investigated

in this study. To make this comparison feasible, the same sample was used, as reported in the

methodology (page 108). However, it has to be noted that the comparison was not made with

Slade’s short form due to the changes that had occurred during the process of adaptation.

The results of this study found that a different method of shortening produced a different

set of items. This study found that only four items were identical. Similar findings were also

reported by Locker and Allen (2002b) in their attempt to shorten the OHIP using an item-impact

176

method. Even using the same method in a different culture produced a different set of items as

reported by Wong et al (2002a) when producing the Chinese short version of OHIP using the

same method as the original short version. They found that only five items were identical in

both the Chinese short form and the original short form. Although they have a different set of

items, they were found to be reliable and valid. This may suggest that the methods in producing

a short version are not stable. The issue of which methods are more stable is a matter of debate.

As Locker and Allen (2002b) stated, “probably the method of developing a short form

instrument is not as important as its content.” They also suggested that different short forms may

be needed according to the purpose for which they are used. If the measure is to be used as a

descriptive measure in a survey to document population oral health-related quality of life, then

the aim should be to maximize the score. A measure consisting of low-prevalence items failed to

document the full extent of the oral health impact (Locker and Allen, 2002b). This study found

that the frequency short form had more items whose prevalence exceeded 20 percent and had

higher scores compared to the regression short form, indicating that it identified more oral

health impacts. Therefore, in a case where a measure is used as a descriptive measure, then

frequency S-OHIP(M) would be the choice.

In a case where a measure is to be used as a discriminative instrument, items affecting

most patients will fail to distinguish between those who are and are not severely compromised

(Locker and Allen, 2002b). As such, the inclusion of low-frequency, relatively severe items will

maximize the ability of the measure to discriminate between groups. On the other hand, if the

measure is to be used as an evaluative measure, high frequency items which are more likely to

demonstrate change as a result of health care interventions should be included in the instrument

(Locker and Allen, 2002b; Guyatt et al, 1986). Nevertheless, according to Juniper et al (1997),

177

the ultimate approach is to test the properties of the instruments based on their purpose, as

described in the literature reviews (page 24)

This study found that both short forms showed excellent internal consistency reliability

and good test-retest reliability when used for this population. High correlation between scores

from both short forms, and scores from the L-OHIP(M), indicated that both forms had good

concurrent validity. Both short forms performed equally well when tested for discriminative

validity. However, the responsiveness of these measures was not tested in this present study.

This could be one of the future research areas to determine which methods will be more superior

in terms of detecting change.

6.6) THE IMPACT OF ORAL HEALTH TO QUALITY OF LIFE OF AN ADULT

POPULATION

This study was also aimed to provide preliminary data on the impact of oral conditions on the

quality of life of the adult population of Malaysia. The L-OHIP(M) data was utilized. The data

obtained by mail and interview was combined since the difference in scores was not statistically

significant. Because the sample of this study was selected based on a probability sampling, the

findings could be generalized at least for the Selangor population. However, due to a small

sample size, only a descriptive analysis was performed.

The response rate of the present study was low in spite of the response-enhancement

strategies that were employed to a certain extent; for example, a letter of introduction was sent

prior to the interview and two reminders were sent to the mail respondents. Perhaps the response

rate could have been increased if a higher number of callbacks and visits to the respondents’

houses were made. Due to time and resources constraints, this was not done. A high non-

response rate in this study could have caused a non-response bias. However, Locker et al (1990)

178

reported that “when the responders and non-responders to surveys do not differ, the response

rate has no effect on prevalence estimates: high participation rates merely serve to make these

estimate more precise.” In the case of the present study, the respondents and non-respondents

did differ in terms of age groups and ethnicity. As such, the prevalence of the OHRQoL reported

here may be biased. However, the direction and the magnitude of the bias could not be

determined in this study. A study by Patten et al (2003) examined the effect of giving an

incentive on response rates in a community-based irritable bowel syndrome (IBS) survey and

found that the prevalence of IBS was higher in the group that was offered no incentive. This

may suggest that persons with IBS may be more likely to participate in such survey. If a same

phenomenon applies to the present study, then the result would be overestimated. Thus, to

ensure that a future study addresses this issue, it is suggested that more money is put to obtain

information on non-respondents rather than attempts to increase the response (Locker et al,

1990). Nevertheless, caution has to be used when interpreting the findings from this study.

The results revealed that a substantial proportion of the adult population in this sample

experienced some very frequent impacts associated with their oral condition. More than one

fourth of the sample reported functional problems, about one fifth reported pain, more than one

third experienced some form of psychological discomfort, and one fifth reported that they were

disabled in some way because of poor oral health.

Many studies on oral health-related quality of life have focused on older populations

with the assumption that they will be more likely to perceive a greater impact on their quality of

life because of a lifetime’s experience of oral ill health. However, the result of this study showed

that the younger age group perceived slightly greater impacts than the older generation. A

similar observation was also seen by Srisilapanan and Sheiham (2001) whereby based on the

OIDP measure, the younger Thai adult population (35-44 years old) had higher impacts than the

179

older people of Thailand. This could be due to the fact that the older generation had adapted to

the situation.

Although McGrath and Bedi (2000) concluded that there are gender variations in the

social and psychological impacts of oral health, this study observed that both genders perceived

similar impacts. The preliminary report of NOHSA found a similar result using a non-

standardized questionnaire in collecting data on the psychosocial impacts (Oral Health Division

Ministry of Health Malaysia, 2001b). This suggests that gender did not affect how oral health is

perceived among adult populations of Malaysia.

In this study, it was found that those of Indian ethnicity experienced the greatest impact

on a relatively frequent basis for almost all domains, compared to the Malay and the Chinese.

However, it must be noted that the sample size of Indians in this study was too small to make a

valid comparison. Further research is needed using a bigger sample size to confirm this

observation.

It was observed that those who attained a higher education reported more impacts related

to oral conditions. It was also observed that respondents who achieved a lower education

reported more impact on the functional limitation, compared to other domains. However, among

those who had a higher education, psychological discomfort was most prevalent. This finding

was similar with the NOHSA preliminary result, where the level III subjects (middle secondary

level and below) reported a significantly higher impact on functional limitation (Oral Health

Division Ministry of Health Malaysia, 2001b).

This study observed that edentulous persons reported more social impacts than dentate

persons on a more frequent basis. However, it must be noted that the sample size for the

edentulous is too small. Slade and Spencer (1994b) also reported that edentulous persons aged

60 and older in South Australia had more social impacts than dentate persons. According to

180

them, the result was not surprising since most of the items in the OHIP related to chewing or

eating.

6.7) VARIATION IN THE ORAL HEALTH IMPACT BETWEEN MALAYSIAN AND

CANADIAN POPULATIONS

This study also aimed to determine whether or not there was variation in terms of oral health

impacts on the quality of life between two different cultures - Malaysian and Canadian - using

the Oral Health Impact Profile (OHIP) measure. OHIP has been adapted for the Malaysian

population. In the process of adaptation, changes to the content were made. Nevertheless, the

Malaysian version of OHIP measures the same underlying concepts as the original OHIP. Thus,

a comparison was only made between each domain rather than between individual items.

Because of the limitations of this study (as noted earlier) it has to be noted that this comparison

only provides some insight rather than making a valid comparison.

From the findings, it was determined that Malaysian older adults were more affected by

their oral conditions than the Canadian older adults. More than half of the Malaysian

respondents experienced very frequent impacts associated with their oral conditions. Were these

high impacts reported by the Malaysian respondents due to their poor oral health status as

measured by the clinical indicators or due to their perception or expectation of oral health? As

Locker et al (1991) observed, respondents who had worse clinical oral health status were not

reflected by their self-reported indicators. It is interesting to note that although Malaysian

respondents reported a higher impact on each subscale, the pattern of impact was almost similar:

physical aspects were most affected and social aspects were the least affected.

When looking at the differences of the impacts between genders, both Malaysian men

and women were more affected than Canadian men and women. Malaysian men had a slightly

181

higher impact than Malaysian women. In contrast, Canadian women had a higher impact than

Canadian men. However, the differences were small in both samples.

In both populations, edentulous respondents reported more impacts than dentate

respondents. However, Malaysian dentate and edentulous groups had higher impacts than the

Canadian population, respectively.

The differences observed in this study could be due to the social and cultural factors

between these two populations. Due to the nature of the study design for both studies we are

unable to determine which factors contributed to the differences observed in this comparison.

However, it provides some evidence that Malaysian adults were more affected than Canadian

adults. Further study with a more appropriate study design is needed to determine what factors

have contributed to these differences.

6.8) CONCLUSION

1) Two versions of the OHIP were developed: L-OHIP(M) and S-OHIP(M). Both versions

were found to be valid and reliable regardless of the mode of adminstration. However,

this study only provides initial evidence for reliability and validity of these measures.

Further study is recommended to collect more evidence to support the validity and

reliability of these measures.

2) Two types of short forms were developed using different methods: frequency and

regression. Both forms performed equally well when tested for discriminative validity.

However, the responsiveness of these measures was not tested in the present study. This

could be one of the future research areas to determine which version will be superior in

terms of detecting change.

3)

4)

182

The preliminary results revealed that a substantial proportion of the sample included in

this study experienced frequent psychosocial impacts associated with oral conditions.

The Malaysian population aged 50 and above had slightly higher impacts than the

Canadian population. However, the pattern of impacts was similar, where physical

consequences were most affected and social aspects were the least affected.

6.9) FUTURE RESEARCH

1)

2)

3)

To translate the Malay version of OHIP(M) to other main languages of Malaysia;

namely English, Tamil and Mandarin, and to test its validity.

Rationale: in this study it was found that the response rate among the non-Malays was

low. This could be due to the difficulty in answering the Malay questionnaire among the

non-Malays. Perhaps, a multilingual questionnaire could improve the response rate

among the non-Malay population.

To test its reliability and validity to Sabah and Sarawak people.

Rationale: The adaptation of the questionnaire in this study was only based on people of

Peninsular Malaysia. Since the culture and lifestyle of Sabah and Sarawak people are

quite different from people of Peninsular Malaysia, the reliability and validity of these

measures need to be reassessed before the questionnaire can be used for Sabah and

Sarawak populations.

To assess the evaluative properties of these measures.

Rationale: The aim of this study was to develop an OHRQoL measure for descriptive

and discriminative purposes. There is a need to develop an evaluative measure to use in

program evaluation, clinical trials, and clinical practice. Thus, the responsiveness

property needs to be established.

183

4) To establish the normative values for the Malaysian population.

Rationale: In order to make a meaningful interpretation of the OHRQoL results, a local

reference of what is ‘normal’ for the population is required (McGrath and Bedi, 2002).

REFERENCES

Aday LA (1996). Designing and conducting health surveys: A comprehensive guide. Second edition. San Francisco: Jossey-Bass Publishers.

Adulyanon S, Sheiham A (1996). A new socio-dental indicators on daily performances.

Journal of Dental Research; 75 (IADR abstracts): abstract # 1711.

Adulyanon S, Sheiham A (1997). Oral Impacts on Daily Performances. In: Slade GD (Ed).

Measuring oral health and quality of life. Chapel Hill, North Carolina: Department of Dental

Ecology, School of Dentistry, University of North Carolina.

Adulyanon S, Vourapukjaru J, Sheiham A (1996). Oral impacts affecting daily performances

in a low dental disease Thai population. Community Dentistry and Oral Epidemiology; 24: 385-389.

Allen PF, Locker D (1997). Do item weights matter? An assessment using the oral health impact profile. Community Dental Health; 14: 133-138.

Allen PF, Mc Millan AS, Walshaw D, Locker D (1999). A comparison of the validity of generic-and disease-specific measures in the assessment of oral health-related quality of life.

Community Dentistry and Oral Epidemiology; 27: 344-352.

Allen PF, McMillan AS (1998). The impact of tooth loss in a denture wearing population: an assessment using the Oral Health Impact Profile. Community Dental Health; 16: 176-180.

Allen PF, McMillan AS, Locker D (2001a). An assessment of sensitivity to change of the Oral Health Impact Profile in a clinical trial. Community Dentistry and Oral Epidemiology; 29: 175-182.

Allen PF, McMillan AS, Walshaw D (2001b). A patient-based assessment of implant- stabilized and conventional complete dentures. Journal of Prosthetic Dentistry; 85: 141-147.

Allison P, Locker D, Jokovic A, Slade G (1999). A cross-cultural study of oral health values.

Journal of Dental Research; 78: 643-649.

Asmah Haji Omar (1985). Language and the worldview of the Malay peasants. In: Osman MT

(Ed). Malaysian world-view. Singapore: Institute of Southeast Asian Studies.

Atchison KA, Der-Martirosian C, Gift H (1998). Components of self-reported oral health and general health in racial and ethnic groups. Journal of Public Health Dentistry; 58: 301-308.

Atchison KA, Dolan TA (1990). Development of the Geriatric Oral Health Assessment Index.

Journal of Dental Education; 54:670-687.

184

185

Awad MA, Locker D, Kormmer-Bitensky N, Feine JS (2000). Measuring the effect of intra-oral implant rehabilitation of health-related quality of life in a randomized controlled clinical trial.

Journal of Dental Research; 79: 1659-1663.

Badia X, Herdman M, Mearin F, Perez I (2000). Adaptation into Spanish of the IBSQOL

questionnaire for the measurement of quality of life in patients with irritable bowel syndrome. Rev. Esp Enfern Dig; 92: 644-650.

Beaton DE, Bombardier C, Guillemin F, Ferraz MB (2000). Guidelines for the process of cross-cultural adaptation of self-report measures. Spine; 25: 3186-3191.

Behling O, Law KS (2000). Translating questionnaires and other research instruments:

problems and solution series, quantitative application in the social sciences. Thousand Oaks: Sage Publications Inc.

Brislin RW, Lonner WJ, Thorndike RM (1973). Cross-cultural Research methods. New York: John Wiley & sons.

Broder HL, Slade G, Caine R, Reisine S (2000). Perceived impact of oral health conditions among minority adolescents. Journal of Public Health Dentistry; 60: 189-192.

Buck D, Jacoby A, Baker GA, Ley H, Steen N (1999). Cross-cultural differences in health-

related quality of life of people with epilepsy: Findings from a European study. Quality of Life Research; 8: 675-685.

Bullinger M, Anderson R, Cella D, Aaronson N (1993). Developing and evaluating cross-

cultural instruments from minimum requirements to optimal models. Quality of Life Research; 2: 451-459.

Bumiputra Commerce (2003). Special economic issue: Summary of Bank Negara 2002 annual report. Economic Research Services; 2:.

Calabrese JM, Friedman PK, Rose LM, Jones JA (1999). Using the GOHAI to assess oral

health status of frail homebound elders: reliability, sensitivity and specificity. Special Care in Dentistry; 19: 214-219.

Carlson ED (2000). A case study in translation methodology using the Health-Promotion

Lifestyle Profile Ul. Public Health Nursing; 17 : 61-70.

Carmines EG, Zeller RA (1979). Reliability and validity assessment. Beverly Hills: Sage Publications.

Cella DF, Lloyd SR, Wright BD (1996). Cross-cultural instruments equating: current research

and future directions. In Spilker B (Ed.). Quality of life and pharmacoeconomics in clinical

trials. Second edition. Philadelphia: Lippincott-Raven.

186

Central Bank of Malaysia (2002). Release of the 2002 Bank Negara Malaysia Annual Report:

Malaysian economy in 2002. http:/www.bnm.gov.my/en/Documents/News/2003.

Coates E, Slade GD, Goss AN et al (1996). Oral conditions and their social impact among HIV dental patients. Australian Dental Journal; 41: 33-36.

Cohen LK, Jago JD (1976). Toward the formulation of sociodental indicators. International Journal of Health Services; 6(4): 681-699.

Collins D (2003). Pretesting survey instruments: An overview of cognitive methods. Quality of Life Research; 12: 229-238.

Coste J, Guillemin F, Pouchot J, Fermanian J (1997). Methodological approaches to shortening composite measurement scales. Journal of Clinical Epidemiology; 50: 247-252.

Cushing AM, Shetham A, Maizels J (1986). Developing socio-dental indicators. The social

impact of dental disease. Community Dental Health; 3:3-17.

Dato Asmah Haji Omar (2000). Oxford Fajar advanced learner’s English-Malay dictionary/ A.S Hornby. Shah Alam: Fajar Bakti.

Del Greco L, Walop W, Eastridge L (1987). Questionnaire development 3: Translation.

Canadian Medical Association; 136:817-818.

Department of Statistics Malaysia (2001). Population and housing census of Malaysia 2000:

Population distribution and basic demographic characteristics. Malaysia: Department of Statistics Malaysia.

Department of Statistics Malaysia (2002). Population and housing census of Malaysia 2000: Education and social characteristics of the population. Malaysia: Department of Statistics Malaysia.

DeVellis RF (1991). Scale development: theory and applications. New Delhi: Sage

Publications.

Deyo RA, Diehr P, Patrick DL (1991). Reproducibility and responsiveness of health status

measures. Controlled Clinical Trials; 12(Suppl): 142-158

Dillman DA (1978). Mail and telephone survey: the total design method. New York: Wiley.

Dolan TA (1997). The sensitivity of the Geriatric Oral Health Assessment Index to dental care.

Journal of Dental Education; 61: 37-46.

Donner A, Eliasziw M (1987). Sample size requirements for reliability studies. Statistics in Medicine 1987; 6: 441-448.

187

Esposito N (2001). From meaning to meaning: the influence of translation techniques on non- English focus group research. Qualitative Health Research;11: 568-579.

Fayers PM, Machin D (2000). Quality of life: Assessment, analysis and interpretation.

England: John Wiley & Sons Ltd.

Fitzpatrick R, Davey C, Buxton MJ, Jones DR (1998). Evaluating patient-based outcome

measures for use in clinical trials. Helath Technology Assessment; 2(14): 1-74.

Flaherty JA, Gaviria FM, Pathak D et al (1988). Developing instruments for cross-cultural psychiatric research. The Journal of Nervous and Mental Disease; 176: 257-263.

French DJ, Caroll A, Christie MJ (1998). Health-related quality of life in Australian Children

with asthma: lessons for the cross-cultural use of quality of life instruments. Quality of Life Research; 7: 409-419.

Geisinger KF (1994). Cross-cultural normative assessment: Translation and adaptation issues

influencing the normative interpretation of assessment instruments. Psychological Assessment; 6: 304-312.

Gift HC, Redford M (1992). Oral health and the quality of life. Clinics in Geriatric Medicine; 8:673-683.

Gill TM, Feinstein AR (1994). A critical appraisal of the quality of quality-of-life

measurements. Journal of American Medical Association; 272: 619-626.

Gilmer JS, Tripp-Reimer T, Buckwalter KC et al (1995). Technical notes: Translation and validation issues for a multidimensional elderly self-assessment instrument. Western Journal of Nursing Research; 17: 220-226.

Gooch BF, Dolan TA, Bourque LB (1989). Correlates of self-reported dental health status

upon enrollment in the RAND health insurance experiment. Journal of Dental Education; 53:629-637.

Green LW, Lewis FM (1986). Measurement and evaluation in health education and health

promotion. California: Mayfield Publishing Company.

Guamaccia PJ (1996). Anthropological perspectives: the importance of culture in the

assessment of quality of life. In Spilker B (Ed.). Quality of life and pharmacoeconomics in

clinical trials. Second edition. Philadelphia: Lippincott-Raven.

Guillemin F, Bombardier C, Beaton D (1993). Cross-cultural adaptation of health-related

quality of life measures: literature review and proposed guidelines. Journal of Clinical

Epidemiology; 46:1417-1432.

188

Guillemin F, Bombardier C, Beaton D (1993). Cross-cultural adaptation of health-related quality of life measures. Literature review and proposed guidelines. J. Clinical Epidemiology; 46: 1417-1432.

Guyatt GH, Bombardier C, Tugwell PX (1986). Measuring disease specific quality of life in clinical trials. The Canadian Medical Association Journal; 134: 889-895.

Guyatt GH, Kirshner B, Jaeschke R (1992). Measuring Health Status. What are the necessary

measurement properties? Journal of Clinical Epidemiology; 45: 1341-1345.

Hanley A, Schott J (1999). Culture, religion and patient care in a multi-ethnic society: A

handbook for professionals. London: Age Concern England.

Hays RD, Anderson R, Revicki DA (1995). Psychometric evaluation and interpretation of

health-related quality of life data. In: Shumaker SA, Berzon R (Eds). The International

Assessment of Health-Related Quality of Life: theory, translation, measurement and analysis.

Oxford: Rapid Communications.

Helman CG (1990). Culture, health and illness. Second edition. London: Wright.

Herdman M, Fox-Rushby J, Badia X (1997). ‘Equivalence’ and translation and adaptation of

health-related quality of life questionnaires. Quality of Life Research; 6:237-247.

Herdman M, Fox-Rushby J, Badia X (1998). A model of equivalence in the cultural adaptation of HRQoL instriments: the universalist approach. Quality of Life Research; 7: 323-335

Higginson IJ, Carr AJ (2001) Measuring quality of life: Using quality of life measure in

clinical setting. British Medical Journal; 322: 1297-1300.

Hunt RJ, Slade GD, Strauss RP (1995). Differences between Racial Groups in the Impact of

Oral Disorders among older Adults in North Carolina. Journal of Public Health Dentistry;

§5(4): 205-209.

Hunt SM (1995). Cross-cultural comparability of quality of life measures. In: Guggenmoos- Holzmann I, Bloofield K, Brenner H, Flick U (Eds). Quality of life and health: Concepts,

methods and applications. Berlin: Blackwell Wissenschafts-Verlag.

Hunt SM, McEwen J, McKenna SP (1986). Measuring health status. London: Croom Helm

Ltd.

Jaafar N (1995). Oral health behaviour in a multi-ethnic society. Symposium: Developing

methods to assess cultural dimensions of oral health and disease.73™ Genaral Session and

Scientific Meeting, International Association for Dental Research, Singapore.

189

John MT, Patrick DL, Slade GD (2002). The German version of the Oral Health Impact

Profile-translation and psychometric properties. European Journal of Oral Sciences; 110: 425- 433.

Jui-meng C (1998). A message from the Minister of health of Malaysia.

http://dph.gov.my/messages/minister/index.htm 1998.

Juniper EF, Guyatt GH, Jaeschke R. (1996). How to develop and validate a new health-related

quality of life instrument. In: Spilker B (ed.). Quality of life and pharmacoeconomics in clinical trials. 2"* edition. Philadelphia: Lippincott-Raven.

Juniper EF, Guyatt GH, Streiner DL, King DR (1997). Clinical impact versus factor analysis for quality of life questionnaire construction. Journal of Clinical Epidemiology; 50: 233-238.

Kim A, Lim EY (1999). How critical is back translation in cross-cultural adaptation of attitue

measures? Paper presented at the annual meeting of the American Educational Research Association, Montreal, Canada.

Kirshner B, Guyatt G (1985). A methodological framework for assessing health indices. Journal of Chronic Disease; 38: 27-36.

Kline P (2000). Handbook of psychological testing. Second edition. London: Routledge.

Kressin N, Spiro A, Bosse R, Garcia R, Kazis L (1996). Assessing oral health-related quality of life: findings from the normative aging study. Medical Care; 34:416-427.

Kressin NR, Atchison KA, Miller DR (1997). Comparing the impact of oral disease in two

populations of older adults: Application of the Geriatric Oral Health Oral Health Assessment

Index. Journal of Public Health Dentistry; 57: 224-232.

Leao A, Sheiham A (1995). Relation between clinical dental status and subjective impacts on daily living. Journal of Dental. Research; 74:1408-1413.

Leplége A, Verdier A (1995). The adaptation of health status measures: methodological aspects of the translation procedure. In: Shumaker SA, Berzon R (Eds). The International

Assessment of Health-Related Quality of Life: theory, translation, measurement and analysis.

Oxford: Rapid Communications.

Locker D (1988). Measuring oral health: A conceptual framework. Community Dental Health;

5:3-18.

Locker D (1996). Applications of self-reported assessment of oral health outcomes. Journal of

Dental Education; 60: 494-500.

190

Locker D (1997a). Concepts of oral health, disease and the quality of life. In: Slade GD.

Measuring oral health and quality of life. Chapel Hill, North Carolina: Department of Dental

Ecology, School of Dentistry, University of North Carolina.

Locker D (1997b). Subjective Oral Health Status Indicators. In: Slade GD (Ed). Measuring

oral health and quality of life. Chapel Hill, North Carolina: Department of Dental Ecology,

School of Dentistry, University of North Carolina.

Locker D (1997c). Clinical correlates of changes in self-perceived oral health in older adults.

Community Dentistry and Oral Epidemiology; 25: 199-203.

Locker D (1998). Issues in measuring change in self-perceived oral health status. Community Dentistry and Oral Epidemiology; 26: 41-47.

Locker D, Matear D, Stephens M, Jokovic A (2002a). Oral health-related quality of life of a

population of medically compromised elderly people. Community Dental Health; 19: 90-97.

Locker D, Allen PF (2002b). Developing short-form measures of oral health-related quality of

life. Journal of Public Health Dentistry; 62: 13-20.

Locker D, Clarke M, Payne B (2000). Self-perceived oral health status, psychological well-

being and life satisfaction in an older adult population. Journal of Dental Research; 79: 970-

975.

Locker D, Jokovic A (1996). Using subjective oral health status indicators to screen for dental

care needs in older adults. Community Dentistry and Oral Epidemiology; 24: 398-402.

Locker D, Jokovic A (1997). Three-year changes in self-perceived oral health status in an

older Canadian population. Journal of Dental Research; 76:1292-1297.

Locker D, Miller Y (1994a). Evaluation of subjective oral health status indicators. Journal of

Public Health Dentistry; 54(3): 167-176.

Locker D, Miller Y (1994b). Subjectively reported oral health status in an adult population. Community Dentistry and Oral Epidemiology; 22:425-430.

Locker D, Slade G (1993). Oral Health and the Quality of Life Among older adults: The Oral

Health Impact Profile. Journal of Canadian Dental Association; 59: 830-838. (cont. Ref. On

page 844).

Locker D, Slade G (1994). Association between clinical and subjective indicators of oral

health status in an older adult population. Gerodontology; 11: 108-114.

Locker D, Leake JL, Hamilton M, Hicks T, Lee J, Main PA (1991). The oral health status of

older adults in four Ontario communities. Journal of Canadian Dental Association; 57: 727-

732.

191

Locker D, Slade GD, Leake JL (1990). The response rate problem in oral health surveys of

older adults in Ontario. Canadian Journal of Public Health; 81: 210-214.

Lwanga SK, Lemeshow S (1991). Sample size determination in health studies: A practical

Manual. Geneva: World Health Organization.

MacLachlan M (1997). Culture and health. Toronto: John Wiley & Sons.

Masalu JR, Astrom AN (2003). Applicability of an abbreviated version of the Oral Impacts on Daily Performances (OIDP) scale for use among Tanzanian students. Community Dentistry

and Oral Epidemiology; 31: 7-14.

Mascarenhas AK (1999). A comparison of oral health in elderly populations seeking and not seeking dental care. Special Care in Dentistry; 19: 248-253.

McDowell I, Newell C (1996). Measuring health: A guide to rating scales and questionnaires. Second edition. New York: Oxford University Press.

McGrath C, Bedi R (2000). Gender variations in the social impact of oral health. Journal of the

Irish Dental Association; 46: 87-91.

McGrath C, Bedi R (2001). An evaluation of a new measure of oral health related quality of

life - OHQoL-UK(W)° . Community Dental Health; 18: 138-143.

McGrath C, Bedi R (2002). Population based norming of the UK oral health related quality of

life measure [OHQoL-UK(W)*! British Dental journal; 193: 521-524.

Melas F, Marcenes W, Wright PS (2001). Oral health impact on daily performances in patients

with implant-stabilized overdentures and patients with conventional complete dentures. International Journal of Oral and Maxillofacial Implants; 16: 700-712.

Ministry of Education (2003). Sekolah. http://www.moe.gov.my/sekolah/frinsckolah. htm

Morigama IM (1968). Problems in the measurement of health status. In: Shelton EB, Moore W

(Eds). Indicators of social change: Concepts and measurements. New York: Russell Sage.

Moroldo MB, De Cunto C, Hiibscher O et al (1998). Cross-cultural adaptation and validation of an Argentine Spanish version of the Stanford Childhood Health Assessment Questionnaire.

Arthritis Care and Research; 11(5): 382-390

Newton JT, Corrigan M, Gibbons DE, Locker D (2003). The self-assessed oral health status of

individuals from White, Indian, Chinese, and Black Caribbean communities in South-east

England. Community Dentistry and Oral Epidemiology; 31: 192-199.

NHHES’96 Team (1999). National household health expenditure survey 1996 (NHHES’96) Malaysia: Final report. Kuala Lumpur: Department of Social and Preventive Medicine,

Faculty of Medicine, University of Malaya.

192

Oral Health Division, Ministry of Health Malaysia (2001a). Oral healthcare in Malaysia.

Malaysia: Oral Health Division, Ministry of Health Malaysia.

Oral Health Division, Ministry of Health Malaysia (2001b). Oral health status, impacts and

treatment needs of Malaysian adults. Kuala Lumpur: Oral Health Division, Ministry of Health Malaysia.

Patrick DL, Deyo RA (1989). Generic and disease-specific measures in assessing health status and quality of life. Medical Care; 27: $217-S232.

Patten SB, Li FX, Cook T, Hilsden RJ, Sutherland LR (2003). Irritable bowel syndrome: are

incentives useful for improving survey response rates? Journal of Clinical Epidemiology; 56:

256-261.

Phillips LR, Luna-de Hernandez I, Torres de Ardon E (1994). Strategies for achieving cultural

equivalence. Research in Nursing and Health; 17:149-154.

Ramli R (2001). Challenges and prospects of oral health promotion — The Malaysian

experience. Paper presented at the FDI conference Kuala Lumpur.

Reisine ST, Fertig J, Waber J, Leder S (1989). Impact of dental conditions on patients’ quality

of life. Community Dentistry and Oral Epidemiology; 17:7-10.

Robinson PG, Gibson B, Khan FA, Birnbaum W (2001). A comparison of OHIP 14 and OIJDP

as interviews ans questionnaires. Community Dental Health; 18: 144-149.

Saunders MJ, Paunovich ED, Cornell JE (1995). Oral health quality of life inventory: development and validation. Journal of Dental Research; 74 (AADR abstracts): abstract #

1253.

Sechrest L, Fay TL, Zaidi SMH (1972). Problems of translation in cross-cultural research.

Journal of Cross-Cultural Psychology; 3: 41-56.

Sheiham A, Steele JG, Marcenes W, Tsakos G, Finch S, Walls AWG (2001). Prevalence of

impacts of dental and oral disorders and their effects on eating among older people; a national

survey in Great Britain. Community Dentistry and Oral Epidemiology; 29: 195-203.

Slade GD (Ed) (1997a). Measuring oral health and quality of life. Chapel Hill, North Carolina:

Department of Dental Ecology, School of Dentistry, University of North Carolina.

Slade GD (1997b). The Oral Health Impact Profile. In: Slade GD (Ed). Measuring oral health

and quality of life. Chapel Hill, North Carolina: Department of Dental Ecology, School of

Dentistry, University of North Carolina.

193

Slade GD (1997c). Derivation and validation of a short-form oral health impact profile. Community Dentistry and Oral Epidemiology; 25: 284-296.

Slade GD (1998). Assessing change in quality of life using the Oral Health Impact Profile.

Community Dentistry and Oral Epidemiology; 26: 52-61.

Slade GD, Hoskin GW, Spencer AJ (1996a). Trends and fluctuations in the impact of oral conditions among older adults during a one year period. Community Dentistry Oral

Epidemiology; 24: 317-321.

Slade GD, Spencer AJ, Locker D et al (1996b). Variations in the Social Impact of oral

Conditions Among Older Adults in South Australia, Ontario, and North Carolina. Journal of Dental Research; 75: 1439-1450.

Slade GD, Spencer AJ (1994a). Development and evaluation of the Oral Health Impact Profile. Community Dental Health; 11:3-11.

Slade GD, Spenser AJ (1994b). Social impact of oral conditions among older adults.

Australian Dental Journal; 39: 358-364.

Slade GD, Spencer AJ, Keily P (1992). Effects of data collection methods on self-reported oral

health impact. Journal of Dental Research; 71: 978 Abstract 15.

Srisilapanan P, Shetham A (2001a). The prevalence of dental impacts on daily performances in

older people in Northern Thailand. Gerodontology; 18: 102-108.

Srisilapanan P, Sheiham A (2001b). Assessing the difference between sociodental and

normative approaches to assessing prosthetic dental treatment needs in dentate older people..

Gerodontology; 18: 25-34.

Strauss RP, Hunt RJ (1993). Understanding the value of teeth to older adults: influences on the quality of life. Journal of American Dental Association; 124:105-110.

Streiner DL, Norman GR (1995). Health measurement scales: A practical guide to their development and use. Second edition. Oxford: Oxford University Press.

Szklo M, Nieto FJ (2000). Epidemiology: beyond the basic. Maryland: An Aspen Publication.

Tewolde S (1999). Malaysia: A cultural profile. University of Toronto, Toronto: Anti-Racism,

Multiculturalism and Native Issues(AMNIJ) Centre, Faculty of Social Work.

Tickle M, Craven R, Blinkhorn AS (1997a). An evaluation of a measure of subjective oral

health status in the UK. Community Dental Health; 14: 175-180.

194

Tickle M, Craven R, Worthington HV (1997b). A comparison of subjective oral health status

of older adults from deprived and affluent communities. Community Dentistry and Oral Epidemiology; 25: 217-222.

Touw-Otten F, Meadows K (1996). Cross-cultural issues in outcome measurement. In: Hutchinson A (Ed). Health outcome measures in primary and out-patient care. The Netherlands: Harwood Academic Publishers.

Tsakos G, Marcenes W, Sheiham A (2001a). Evaluation of a modified version of the index of

Oral Impacts on Daily Performances (OIDP) in elderly populations in two European countries. Gerodontology; 18: 121-130.

Tsakos G, Marcenes W, Shieham A (2001b). Cross-cultural differences in oral impacts on daily performances between Greek and British older adults. Community Dental Health; 18: 209-213.

Tubert-Jeannin S, Riordan PJ, Morel-Papernot A, Porcheray S, Saby-Collet S (2003).

Validation of an oral health quality of life index (GOHAI) in France. Community Dentistry and Oral Epidemiology; 31: 275-284.

van de Vijver F, Kwok Leong (1997). Methods and data Analysis for Cross-cultural research.

Thousand Oaks: Sage Publication Inc.

Wilkin D, Hallam L, Doggett MA (1992). Measures of need and outcome for primary health

care, Oxford: Oxford University Press.

Wilson IB, Cleary PD (1995). Linking clinical variables with health-related Quality of life.

Journal of American Medical Association; 273: 59-65.

Wong MC, Lo EC, McMillan AS (2002a). Validation of a Chinese version of the Oral Health

Impact Profile (OHIP). Community Dentistry and Oral Epidemiology; 30: 423-430.

Wong MCM, Liu JKS, Lo ECM (2002b). Translation and validation of the Chinese version of GOHAL. Journal of Public Health Dentistry; 62: 78-83.

World Health Organization (1948). Primary health care. Report of the international

conference on primary health care. Alma-Ata, U.S.S.R., Geneva: World Health Organization.

APPENDIX A

Oral Health Impact Profile Questionnaire — Original

195

-t.

Xe, oO {w3q 40°

The University of North Carolina at Chapel Hill

School of Dentistry

Oral Health Impact Profile

Page 1

INSTRUCTIONS THE QUESTIONNAIRE. This questionnaire asks how troubles with your teeth, mouth or dentures may have caused problems in your daily life. We would like you to complete the questionnaire even if you have good dental health. We would like to know how often you have had each of the 49 listed problems during the LAST YEAR.

HOW TO ANSWER THE QUESTIONS. Each question on the left hand side of the page asks you about a particular dental problem. You should think about each question in turn, and circle the answer to the right of the question, to indicate how often you have had the problem during the last year.

If you occasionally had sore spots in your mouth, you would circle the answer as shown in this example.

Q16.Have you had any sore spots in VERY FAIRLY /O HARDLY NEVER DON'T your mouth? OFTEN OFTEN EVER KNOW

If you have never had the problem during the last year, circle "NEVER" as follows. Q16.Have you had any sore spots in VERY FAIRLY OCCAS- HARDLY DON'T

your mouth? OFTEN OFTEN IONALLY EVER KNOW

WHAT IF THE QUESTION DOES NOT APPLY? Question 3 applies only to people who have all or some of their own teeth. If the question does not apply to you then you would answer by checking the box as follows. EXAMPLE

Q13.Have you had a toothache? VERY FAIRLY OCCAS- HARDLY NEVER DON'T OFTEN OFTEN IONALLY EVER KNOW

Does not apply - | do not have my own teeth [4

ANY QUESTIONS? If you would like to speak to one of the researchers, please phone Kevin Moss at the School of Dentistry, phone (919) 966-2791. Once again, thank you for your help!

Page 2

HOW OFTEN have you had the problem during the last year? (circle your answer)

Q1. Have you had difficulty chewing any foods because of problems VERY FAIRLY OCCAS- HARDLY NEVER DON’T

with your teeth, mouth or OFTEN OFTEN JONALLY EVER KNOW dentures?

Q2. Have you had trouble pronouncing any words because VERY FAIRLY OCCAS- HARDLY NEVER DON’T

of problems with your teeth, OFTEN OFTEN IONALLY EVER KNOW mouth or dentures?

Q3. Have you noticed a tooth which VERY FAIRLY OCCAS- HARDLY NEVER DON’T

doesn't look right? OFTEN OFTEN {ONALLY EVER KNOW

Q4. Have you felt that your appearance has been affected VERY FAIRLY OCCAS- HARDLY NEVER DON’T

because of problems with your OFTEN OFTEN JONALLY EVER KNOW teeth, mouth or dentures?

Q5. Have you felt that your breath has been stale because of problems VERY FAIRLY OCCAS- HARDLY NEVER DON'T with your teeth, mouth or OFTEN OFTEN IONALLY EVER KNOW

dentures?

Q6. Have you felt that your sense of taste has worsened because of VERY FAIRLY OCCAS- HARDLY NEVER DON’T

problems with your teeth, mouth OFTEN OFTEN IONALLY EVER KNOW or dentures?

Q7. Have you had food catching in VERY FAIRLY OCCAS- HARDLY NEVER DON’T your teeth or dentures? OFTEN OFTEN IONALLY EVER KNOW

Q8. Have you felt that your digestion has worsened because of VERY FAIRLY OCCAS- HARDLY NEVER DON'T

problems with your teeth, mouth OFTEN OFTEN IONALLY EVER KNOW or dentures?

Q9. Have you had painful aching in VERY FAIRLY OCCAS- HARDLY NEVER DON’T your mouth? OFTEN OFTEN IONALLY EVER KNOW

Q10. Have you had a sore jaw? VERY FAIRLY OCCAS- HARDLY NEVER DON’T

OFTEN OFTEN JONALLY EVER KNOW Qii. Have you had headaches

because of problems with your VERY FAIRLY OCCAS- HARDLY NEVER DON’T

teeth, mouth or dentures? OFTEN OFTEN JIONALLY EVER KNOW

Page 3

HOW OFTEN have you had the problem during the last year? (circle your answer)

Q12. Have you had sensitive teeth, for example, due to hot or cold foods VERY FAIRLY OCCAS- HARDLY NEVER DON'T or drinks? OFTEN OFTEN JIONALLY EVER KNOW

Does not apply - | do not have my own teeth

Q13. Have you had toothache? VERY FAIRLY OCCAS- HARDLY NEVER DON'T OFTEN OFTEN IONALLY EVER KNOW

Does not apply - | do not have my own teeth

Q14. Have you had painful gums? VERY FAIRLY OCCAS- HARDLY NEVER DON’T OFTEN OFTEN IONALLY EVER KNOW

Q15. Have you found it uncomfortable to eat any foods because of VERY FAIRLY OCCAS- HARDLY NEVER DON’T

problems with your teeth, mouth OFTEN OFTEN JIONALLY EVER KNOW or dentures?

Q16. Have you had sore spots in your VERY FAIRLY OCCAS- HARDLY NEVER DON'T

mouth? OFTEN OFTEN IONALLY EVER KNOW

Q17. Have you felt that your dentures VERY FAIRLY OCCAS- HARDLY NEVER DON'T have not been fitting properly? OFTEN OFTEN IONALLY EVER KNOW

Does not apply - | do not have dentures

Q18. Have you had uncomfortable VERY FAIRLY OCCAS- HARDLY NEVER DON'T

dentures? OFTEN OFTEN IONALLY EVER KNOW

Does not apply - | do not have dentures

Q19. Have you been worried by dental VERY FAIRLY OCCAS- HARDLY NEVER DON'T problems? OFTEN OFTEN JONALLY EVER KNOW

Q20. Have you been seif conscious because of your teeth, mouth or VERY FAIRLY OCCAS- HARDLY NEVER DON'T dentures? OFTEN OFTEN IONALLY EVER KNOW

Q21. Have dental problems made you VERY FAIRLY OCCAS- HARDLY NEVER DON'T miserable? OFTEN OFTEN JONALLY EVER KNOW

Q22. Have you felt uncomfortable about the appearance of your VERY FAIRLY OCCAS- HARDLY NEVER DON'T

teeth, mouth or dentures? OFTEN OFTEN IONALLY EVER KNOW

Page 4

HOW OFTEN have you had the problem during the last year?

(circle your answer)

Q23. Have you felt tense because of VERY FAIRLY OCCAS- HARDLY NEVER DON’T problems with your teeth, mouth OFTEN OFTEN JONALLY EVER KNOW or dentures?

Q24. Has your speech been unclear because of problems with your VERY FAIRLY OCCAS- HARDLY NEVER DON’T teeth, mouth or dentures? OFTEN OFTEN IONALLY EVER KNOW

Q25. Have people misunderstood some of your words because of VERY FAIRLY OCCAS- HARDLY NEVER DON’T

problems with your teeth, mouth OFTEN OFTEN JIONALLY EVER KNOW or dentures?

Q26. Have you felt that there has been less flavor in your food because VERY FAIRLY OCCAS- HARDLY NEVER DON’T

of problems with your teeth, OFTEN OFTEN JIONALLY EVER KNOW mouth or dentures?

Q27. Have you been unable to brush your teeth properly because of VERY FAIRLY OCCAS- HARDLY NEVER DON’T problems with your teeth, mouth OFTEN OFTEN IONALLY EVER KNOW or dentures?

Does not apply - | do not have my own teeth

Q28. Have you had to avoid eating some foods because of problems VERY FAIRLY OCCAS- HARDLY NEVER DON’T

with your teeth, mouth or OFTEN OFTEN {ONALLY EVER KNOW dentures?

Q29. Has your diet been unsatisfactory

because of problems with your VERY FAIRLY OCCAS- HARDLY NEVER DON'T

teeth, mouth or dentures? OFTEN OFTEN JONALLY EVER KNOW

Q30. Have you been unable to eat with your dentures because of VERY FAIRLY OCCAS- HARDLY NEVER DON'T problems with them? OFTEN OFTEN IONALLY EVER KNOW

Does not apply - | do not have dentures

Q31. Have you avoided smiling because of problems with your VERY FAIRLY OCCAS- HARDLY NEVER DON’T

teeth, mouth or dentures? OFTEN OFTEN JONALLY EVER KNOW

Page 5

HOW OFTEN have you had the problem during the last year? (circle your answer)

Q32. Have you had to interrupt meals

because of problems with your VERY FAIRLY OCCAS- HARDLY NEVER DON’T

teeth, mouth or dentures? OFTEN OFTEN IONALLY EVER KNOW

Q33. Has your sleep been interrupted

because of problems with your VERY FAIRLY OCCAS- HARDLY NEVER DON’T

teeth, mouth or dentures? OFTEN OFTEN JONALLY EVER KNOW

Q34. Have you been upset because of problems with your teeth, mouth VERY FAIRLY OCCAS- HARDLY NEVER DON’T or dentures? OFTEN OFTEN IONALLY EVER KNOW

Q35. Have you found it difficult to relax because of problems with your VERY FAIRLY OCCAS- HARDLY NEVER DON’T teeth, mouth or dentures? OFTEN OFTEN JONALLY EVER KNOW

Q36. Have you felt depressed because of problems with your teeth, VERY FAIRLY OCCAS- HARDLY NEVER DON'T mouth or dentures? OFTEN OFTEN JIONALLY EVER KNOW

Q37. Has your concentration been affected because of problems VERY FAIRLY OCCAS- HARDLY NEVER DON’T with your teeth, mouth or OFTEN OFTEN IONALLY EVER KNOW dentures?

Q38. Have you been a bit embarrassed because of problems with your VERY FAIRLY OCCAS- HARDLY NEVER DON'T

teeth, mouth or dentures? OFTEN OFTEN JONALLY EVER KNOW

Q39. Have you avoided going out because of problems with your VERY FAIRLY OCCAS- HARDLY NEVER DON'T

teeth, mouth or dentures? OFTEN OFTEN JONALLY EVER KNOW

Q40. Have you been less tolerant of your partner or family because of VERY FAIRLY OCCAS- HARDLY NEVER DON'T

problems with your teeth, mouth OFTEN OFTEN JONALLY EVER KNOW

or dentures?

Q41. Have you had trouble getting along with other people because VERY FAIRLY OCCAS- HARDLY NEVER DON'T

of problems with your teeth, OFTEN OFTEN JIONALLY EVER KNOW mouth or dentures?

Page 6

HOW OFTEN have you had the problem during the last year? (circle your answer)

Q42. Have you been a bit irritable with other people because of VERY FAIRLY OCCAS- HARDLY NEVER DON'T

problems with your teeth, mouth OFTEN OFTEN IONALLY EVER KNOW or dentures?

Q43. Have you had difficulty doing your usual jobs because of problems VERY FAIRLY OCCAS- HARDLY NEVER DON'T with your teeth, mouth or OFTEN OFTEN IONALLY EVER KNOW dentures?

Q44. Have you felt that your general health has worsened because of VERY FAIRLY OCCAS- HARDLY NEVER DON’T

problems with your teeth, mouth OFTEN OFTEN IONALLY EVER KNOW or dentures?

Q45. Have you suffered any financial loss because of probiems with VERY FAIRLY OCCAS- HARDLY NEVER DON'T

your teeth, mouth or dentures? OFTEN OFTEN IONALLY EVER KNOW

Q46. Have you been unable to enjoy other people's company as much VERY FAIRLY OCCAS- HARDLY NEVER DON’T

because of problems with your OFTEN OFTEN {ONALLY EVER KNOW teeth, mouth or dentures?

Q47. Have you felt that life in general was less satisfying because of VERY FAIRLY OCCAS- HARDLY NEVER DON’T

problems with your teeth, mouth OFTEN OFTEN JONALLY EVER KNOW or dentures?

Q48. Have you been totally unable to function because of problems VERY FAIRLY OCCAS- HARDLY NEVER DON'T

with your teeth, mouth or OFTEN OFTEN JONALLY EVER KNOW dentures?

Q49. Have you been unable to work to

your full capacity because of VERY FAIRLY OCCAS- HARDLY NEVER DON’T

problems with your teeth, mouth OFTEN OFTEN JONALLY EVER KNOW

or dentures?

Q50. Please write today’s date /_ J month day year

Page 7

APPENDIX B

Qualitative Interview Statement

196

UNIVERSITY MALAYA

JABATAN PERGIGIAN MASYARAKAT FAKULTI PERGIGIAN, UNIVERSITI MALAYA 50603 KUALA LUMPUR MALAYSIA

Department of Community Dentistry

Faculty of Dentistry, University of Malaya Tel: 603-7967 4805 50603 Kuala Lumpur Malaysia Fax: 603-7967 4532

Development of an Oral Health-Related Quality of Life Measure for the Malaysian Population: Cross-cultural Adaptation of the Oral Health Impact Profile

INFORMATION SHEET

(in-depth interview study)

What are we doing? As part of my PhD research I am developing a questionnaire that will assess the

problems that dental and oral diseases create for people in their day to day life. This questionnaire can be used in many ways — to help to determine the amount and type of dental care needed by the Malaysian population, or to tell us if the treatment we provide to patients does in fact improve their overall health and well-being. A questionnaire of this type, called the Oral Health Impact Profile, was developed in Australia and part of our work is to see if it will be useful for us to use here in Malaysia. We expect that the questionnaire will need to be modified in several ways to make it appropriate for the Malaysian population.

What would we like you to do? First, we would like to talk to you about the dental or oral problems that have brought

you to the Faculty today and the way these problems affect your daily life. We would like to tape record this interview to ensure that we get as much detailed information as possible. Then we would like to show you a copy of the Oral Health Impact Profile and have your opinions on the relevance and clarity of the questions it contains. Also we would like to know if your dental or oral condition causes you any problems that are not covered by these questions. The interview will take between 30 and 40 minutes to complete.

Your participation in this study is entirely voluntary and will not affect the services you receive from the Faculty of Dentistry. You may request that your interview is not recorded and you can decline to answer any of the questions you are asked. You are free to withdraw from the interview at any time.

What happens next? The questionnaire will be revised based on the comments of the people who take part

in this phase of the study. We will then do another study to see if the questionnaire is easy to complete and acceptable to the Malaysian population on whom it will be used. We will then use it in further studies to make sure that it does measure how dental and oral problems affect

the well-being of Malaysians.

Investigator: Dr Roslan Saub, Department of Community Dentistry, Faculty of Dentistry, University of Malaya.

Phone number: 7967 4551/4805/4856 Email: roslans@um.edu.iny

N:\Windows NT 5.0 Workstation Profile\My Documents\Appendix\Appendix B -interviewinfo-English.doc

UNIVERSITY MALAYA

JABATAN PERGIGIAN MASYARAKAT FAKULTI PERGIGIAN, UNIVERSITI MALAYA

50603 KUALA LUMPUR MALAYSIA

Department of Community Dentistry

Faculty of Dentistry, University of Malaya Tel: 603-7967 4805 530603 Kuala Lumpur Malaysia Fax: 603-7967 4932

“Development of an Oral Health-Related Quality of Life Measure for the Malaysian Population: Cross-cultural Adaptation of the Oral Health Impact Profile”

MAKLUMAT KAJIAN (Kajian Temu Ramah)

Apa yang kami buat? Saya sedang menghasilkan satu borang kajiselidik untuk menilai kesan masalah pergigian ke

atas kehidupan seharian seseorang sebagai sebahagian dari projek PhD saya. Borang kajiselidik ini boleh digunakan dalam berbagai cara — membantu di dalam menentukan kuantiti dan jenis penjagaan pergigian yang diperlukan oleh masyarakat Malaysia, atau memberitahu kita samada rawatan yang

diberikan kepada pesakit telah meningkatkan mutu keseluruahan kesihatan dan kesejahteran hidup. Borang kejiselidik seperti ini yang dikenali sebagai “Profile kesan kesihatan mulut” yang telah dibentuk di Australia dan sebahagian dari kerja saya ialah untuk mengkaji samada borang kajiselidik ini sesuai untuk digunakan di Malaysia. Saya menjangkakan borang ini perlu divbahsui dalam beberapa cara untuk disesuikan bagi kegunaan di Malaysia.

Apa yang kami hendak anda lakukan?

Pertamanya kami ingin menemu ramah anda tentang masalah pergigian yang anda alami yang telah membawa anda ke Fakulti Pergigian ini dan juga kami ingin mengetahui bagaimana masalah tersebut memberi kesan ke atas kehidupan seharian anda. Kami ingin merakamkan temu ramah ini untuk memastikan kami dapat semua maklumat secara terperinci yang mungkin. Kemudian kami akan memberikan kepada anda borang kajiselidik “Profile Kesan Kesihatan Mulut” untuk mendapatkan pendapat anda samada soalan yang terkandung di dalam borang tersebut releven dan jelas. Kami juga ingin mengetahui jika ada masalah yang anda hadapi disebabkan masalah pergigian anda yang tidak terdapat di dalam borang kajiselidik ini. Temu ramah ini akan mengambil masa selama lebih kurang 30 —40 minit.

Penglibatan anda di dalam kajian ini adalah secara sukarela dan ia tidak akan menjejaskan perkhidmatan yang anda terima di Fakulti Pergigian. Anda boleh meminta supaya temu ramah anda tidak dirakamkan dan anda boleh untuk tidak menjawab mana-mana soalan yang ditanya. Anda boleh untuk menarik diri pada bila-bila masa sahaja.

Apa yang berlaku kemudian? Borang kajiselidik tersebut akan dikaji semula berpandukan kepada komen yang diberikan

oleh mereka yang terlibat di dalam temu ramah ini. Kami kemudian akan menjalankan satu lagi kajian untuk melihat samada borang kajisedik yang telah dihasilkan mudah untuk dijawab dan sesuai untuk masyarakt Malaysia. Kemudian kami akan menjalankan kajiselidik seterusnya dengan menggunakan borang kajiselidik ini untuk memastikan ia dapat mengukur kesan masalah pergigian keatas kesejahteran hidup seseorang.

Penyelidik: Dr Rosian Saub, Jabatan Pergigian Masyarakat, Fakulti Pergigian, Universiti Malaya. Phone number: 7967 455 1/4805/4856 Email: raslans@uim.edu.my

N:\Windows NT 5.0 Workstation Profile\My Documents\Appendix\Appendix B-interviewinfo-bahasa.doc

APPENDIX C

Consent Form — Qualitative Interview

197

Appendix C

UNIVERSITY MALAYA

JABATAN PERGIGIAN MASYARAKAT FAKULTI PERGIGIAN, UNIVERSITI MALAYA $0603 KUALA LUMPUR MALAYSIA

Department of Community Dentistry

Faculty of Dentistry, University of Malaya me pone tes 50603 Kuala Lumpur Malaysia “oe

CONSENT FORM

(In-depth interview study)

This consent form is to indicate my agreement to participate in the study “Development of an oral health-related quality of life measure for the Malaysian population; Cross-cultural adaptation of the Oral Health Impact Profile”.

I have been informed that the study is part of the PhD project of Dr Roslan Saub and aims to develop a questionnaire which can measure the way in which oral diseases and disorders affect people in their everyday life. I have been informed that I will be asked to participate in an interview of approximately 30 to 40 minutes in which I will be asked

questions concerning my dental and oral problems and the way in which they affect my daily

activities such as eating, communicating with others and how I feel about myself. I have been informed that this interview will be tape-recorded. I have also been informed that I will be asked to examine an existing questionnaire (the Oral Health Impact Profile) to see if it does or

does not describe adequately things that happen to me because of oral problems. I will be asked to comment on the content of the questionnaire and the relevance and clarity of the questions that it contains.

I understand that my participation is entirely voluntary and I can either refuse to take part, or if I do take part to refuse to answer any questions that I might be asked. I understand that I can also withdraw at any time during the interview and also request that the tape recorder is not used to record my comments. I understand that a refusal to take part or to answer any questions will not affect my receipt of any services provided by the Faculty of Dentistry, University of Malaysia. I have been assured that all information I give will remain completely confidential and my name will not appear on any forms or documents other than this consent form or appear in any papers or reports based on the study. I understand that I will not benefit from taking part in the study.

Name Signature

Date

Witness

[1

N:\Windows NT 5.0 Workstation Profile\My Documents\Appendix\Appendix C- interviewconsent-English.doc

UNIVERSITY MALAYA

JABATAN PERGIGIAN MASYARAKAT FAKULTI PERGIGIAN, UNIVERSITI MALAYA 50603 KUALA LUMPUR MALAYSIA

Department of Community Dentistry

Faculty of Dentistry, University of Malaya Tel. 603-7967 4805

50603 Kuala Lumpur Malaysia Fax: 603-7967 4532

SURAT PERAKUAN (Kajian Temu Ramah)

‘Surat perakuan ini menandakan yang saya bersetuju untuk melibatserta di dalam kajian yang bertajuk “Development of an oral health-related quality of life measure for the Malaysian population; Cross-cultural adaptation of the Oral Health Impact Profile”.

Saya telah diberitahu bahawa kajian ini adalah sebahagian daripada Projek PhD Dr Roslan Saub yang bertujuan untuk menghasilkan satu borang soalselidik yang boleh digunakan untuk mengukur kesan masalah pergigian keatas kehidupan seharian seseorang. Saya diberitahu bahawa saya dikehendaki terlibat di dalam satu temu ramah yang akan mengambil masa selama lebihkurang 30 - 40 minit di mana saya akan ditanya soalan berkaitan dengan masalah mulut dan gigi dan kesannya keatas kehidupan seharian saya seperti makan, berkomunikasi dengan orang lain dan bagaimana saya merasakan tentang diri saya sendiri. Saya diberitahu bahawa temu ramah ini akan dirakamkan. Saya juga dikehendaki meneliti borang kajiselidik yang sediaada (Profile Kesan Kesihatan Mulut) untuk melihat samada borang ini mengambarkan sepenuhnya perkara yang berlaku ke atas diri saya yang disebabkan oleh masalah mulut dan gigi. Saya juga dikehendaki untuk memberikan komen tentang kandungan borang kajiselidik tersebut samada soalan—soalan yang terkandung relevan dan jelas.

Saya faham bahawa penglibatan saya di dalam kajian ini adalah sukarela dan saya boleh untuk tidak terlibat di dalam kajian tersebut dan jika saya terlibat saya boleh untuk tidak menjawab mana- mana soalan yang ditanya kepada saya. Saya faham bahawa saya juga boleh menarik diri pada bila- bila masa semasa temu ramah dijalankan dan saya juga boleh meminta supaya komen saya tidak dirakamkan. Saya juga faham bahawa jika saya tidak mahu terlibat di dalam kajian tersebut atau menjawab mana-mana soalan yang diajukan kepada saya tidak akan menjejaskan perkhidmatan yang diberikan oleh Fakulti Pergigian Universiti Malaya kepada saya. Saya telah diberi jaminan bahawa segala maklumat yang saya berikan akan dianggap sulit dan nama saya tidak akan digunakan di dalam apa-apa borang atau dokumen melainkan borang perakuan ini atau dalam mana-mana laporan berkaitan dengan kajian ini. Saya faham saya tidak mempunyai sebarang keuntungan dari menyertai kajian ini.

Nama Tandatangan

Tarikh

Saksi

LT]

N:\Windows NT 5.0 Workstation Profile\My Documents\Appendix\Appendix C- interviewconsent-bahasa.doc

APPENDIX D

Guidelines - Qualitative Interview

198

CHECKLIST FOR INTERVIEWS

What is your reason for coming today?

How long you have this............ (the problem)?

Can you tell me little bit more about your problem?

How does this............ (problem) affect your daily life?

Eating?

Difficulty chewing food?

Avoiding some food?

The taste of the food?

Speech?

Appearance?

Work?

Relationship with family, friends etc.?

Mood change?

Sleep?

Pain and discomfort?

Oral care?

Do you wear dentures or any prosthesis? If “Yes”

What type of denture do you wear?

Can you tell me your experience wearing dentures?

Demography information:

N:\Windows NT 5.0 Workstation Profile\My Documents \Appendix\Appendix D -interview guideline.doc

SENARAI SEMAK UNTUK TEMU RAMAH

Apakah tujuan kehadiran anda hari ini?

Berapa lama anda telah mengalami........... (masalah)?

Bolehkah anda beritahu saya lebih lanjut berkaitan dengan.......

(masalah)?

Bagaimana......... (masalah) memberi kesan ke atas kehidupan seharian

anda?

Makan?

Kesukaran mengunyah?

Cegah sesetengah makanan?

Rasa makanan?

Pertuturan?

Rupa-paras?

Kerja?

Perhubungan dengan keluarga dan rakan.?

Peruabhan mood?

Tidur?

Kesakitan dan ketidak selesaan?

Penjagaan mulut?

Adakah anda memakai gigi palsu atau sebarang prostesis? Jika “Ya”

Apakah jenis gigi palsu anda pakai?

Bolehkah anda beritahu saya pengalaman anda memakai gigi palsu?

informasi demografi: COP OORT ee OTE ORE EE E O ERHO E e

N:\Windows NT 5.0 Workstation Profile\My Documents \Appendix\ Appendix D -interview guideline.doc

APPENDIX E

Malaysian Oral Health Impact Profile Questionnaire — Long Form

L-OHIPC™)

199

UNIVERSITI MALAYA

JABATAN PERGIGIAN MASYARAKAT

FAKULTI PERGIGIAN, UNIVERSITI MALAYA

50603 KUALA LUMPUR MALAYSIA

Department of Community Dentistry

Faculty of Dentistry, University of Malaya

50603 Kuala Lumpur Malaysia

BORANG SOAL SELIDIK PROFIL KESAN KESIHATAN MULUT

L-OHIP(M)

ID:CILILIC

KAM! AMAT MENGHARGAI JASA BAIK ANDA MELUANGKAN SEDIKIT MASA UNTUK MENJAWAB BORANG SOAL SELIDIK INI.

BAHAGIAN A Bahagian ini ingin mengetahui sejauh mana masalah berkaitan dengan gigi, mulut atau gigi palsu mengganggu kehidupan seharian anda. Kami ingin mengetahui berapa kerap anda

pemah mengalami masalah yang tersenarai sepanjang SETAHUN YANG LEPAS. Apa yang anda perlu lakukan ialah menanda (V) pada kotak yang disediakan untuk setiap soalan- soalan berikut.

BERAPA KERAPKAH ANDA MENGALAMI MASALAH-MASALAH YANG TERSENARAI SEPANJANG SETAHUN YANG LEPAS.

(Tandakan ¥ untuk jawapan anda)

Pejabat

Ai) Pernahkah anda mengalami kesukaran mengunyah sebarang makanan [at disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT [| /KERAP [_|KADANG- [| |SEKALI- [| |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

A2) Pernahkah anda mengalami kesukaran menyebut sebarang perkataan | ta2 disebabkan masalah gigi, mulut atau gigi palsu anda?

[JSANGAT [| |KERAP | |KADANG- | |SEKALI- [| |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

A3) Pernahkan anda merasakan wajah anda terjejas disebabkan masalah gigi, [las mulut atau gigi palsu anda?

[ISANGAT [| |KERAP | |KADANG- [| |SEKALI- [ |TIDAK [_ ]TIDAK KERAP KADANG SEKALA PERNAH TAHU

A4) Pernahkah anda merasakan yang masalah gigi, mulut atau gigi palsu anda [jaa menyebabkan nafas anda berbau?

[|ISANGAT [| |KERAP [ |KADANG- [ |SEKAL- [ |TIDAK []TIDAK KERAP KADANG SEKALA PERNAH TAHU

A5) Pernahkah anda merasakan makanan yang dimakan tidak hadam dengan | [_|as baik disebabkan masalah gigi, mulut atau gigi palsu anda?

[]SANGAT [| |KERAP | |KADANG- [| |SEKALI- [ /TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

iD: OOF 2

BERAPA KERAPKAH ANDA MENGALAMI MASALAH-MASALAH YANG TERSENARAI SEPANJANG SETAHUN YANG LEPAS.

(Tandakan ¥ untuk jawapan anda)

AB)

A7)

A8)

AQ)

A10)

A11)

A12)

Pernahkah anda mengailami sakit rahang?

[ |ISANGAT | |KERAP [| |KADANG- | |SEKALI- [ |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengalami masaiah gigi, mulut atau gigi palsu yang menyebabkan anda sakit kepala?

[|SANGAT | |KERAP | |KADANG- | |SEKALI- [| /TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengalami rasa ngilu pada gigi, contohnya apabila terkena air atau makanan panas atau sejuk?

| _|TIDAK BERKAITAN - SAYA TIDAK MEMPUNYAI GIGI SENDIRI.

[JSANGAT | IKERAP | |KADANG- [| |SEKALI- | /TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengalami sakit gigi?

[_|TIDAK BERKAITAN - SAYA TIDAK MEMPUNYAI GIG! SENDIRI.

[ISANGAT [| |KERAP [| |KADANG- [| |SEKALI- | |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengalami sakit gusi?

[_ISANGAT | |KERAP | |KADANG- | |SEKALI- | |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengalami rasa tidak selesa untuk makan sebarang makanan disebabkan masalah gigi, mulut atau gigi palsu anda?

[]SANGAT [| |KERAP [| |KADANG- [ |SEKALL- [ |TIDAK [_|TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengalami tompok-tompok putih yang pedih (Ulser) di dalam mulut?

[ISANGAT [| |KERAP [| |KADANG- [| |SEKALI- [| |TIDAK [ |TIDAK KERAP KADANG SEKALA PERNAH TAHU

ID: CILIC I

Pejabat

| |a6

[| Aq

[las

[las

[ato

[Jan

| dan

BERAPA KERAPKAH ANDA MENGALAMI MASALAH-MASALAH YANG TERSENARAI SEPANJANG SETAHUN YANG LEPAS.

(Tandakan V untuk jawapan anda)

A13)

A14)

A15)

A16)

A17)

A18)

Pernahkah anda merasakan yang gigi palsu anda longgar semasa anda memakainya?

|_| TIDAK BERKAITAN - SAYA TIDAK MEMAKAI GIGI PALSU.

[ISANGAT [| |KERAP [| |KADANG- [| |SEKALI- [| ITIDAK [ |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda merasa kurang selesa dengan gigi palsu anda?

|_|TIDAK BERKAITAN - SAYA TIDAK MEMAKAI GIGI PALSU.

[ISANGAT | |KERAP [| |KADANG- | |SEKALI- | |TIDAK [ |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda merasa risau dengan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT [ |KERAP [| |KADANG- | |SEKALI- [| |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda merasa tidak selesa disebabkan makanan terlekat di celah

gigi atau gigi palsu anda?

[-JSANGAT [| |KERAP [| |KADANG- [| |SEKALI- | /TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda merasa malu disebabkan masalah gigi, mulut atau gigi palsu anda?

(_ISANGAT | |KERAP [| |KADANG- | |SEKALI- [| |TIDAK |_|TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda merasa tidak selesa dengan penampilan gigi, mulut atau gigi palsu anda?

[]SANGAT [ |KERAP [.|KADANG- [ /SEKALI- | |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

ID: ILICIL]

Pejabat

[dan

[lai

[jas

late

[law

[las

BERAPA KERAPKAH ANDA MENGALAMI MASALAH-MASALAH YANG TERSENARAI SEPANJANG SETAHUN YANG LEPAS.

(Tandakan V untuk jawapan anda)

A19)

A20)

A21)

A22)

A23)

A24)

A25)

Pernahkah anda merasa tertekan disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT [|_|KERAP [| |KADANG- | |SEKALI- | /TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah pertuturan anda menjadi tidak jelas disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT [ |KERAP | |KADANG- [| |SEKALI- | [TIDAK [ |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah orang tersalah faham sesetengah perkataan yang anda sebutkan disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT [| |KERAP | |KADANG- | [SEKALI- | |TIDAK | ]TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda merasa makanan yang dimakan kurang sedap disebabkan masalah gigi, mulut atau gigi palsu anda?

[_ISANGAT [ |KERAP [| |KADANG- | |SEKALI- [ |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengalami kesukaran untuk memberus gigi dengan sempurna disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT [| |KERAP [| |KADANG- [ |SEKALL- [_ |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengelak daripada memakan makanan tertentu disebabkan masalah gigi, mulut atau gigi palsu anda?

[]SANGAT | |KERAP [ |KADANG- [ |SEKALI- [_|TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda tidak dapat makan makanan kesukaan anda disebabkan

masalah gigi, mulut atau gigi palsu anda?

[|ISANGAT [| |KERAP [| |KADANG- [| |SEKALI- [_|TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

DIO

Pejabat

| las

|_| a20

[a2

[| a2s

[laze

|| aos

BERAPA KERAPKAH ANDA MENGALAMI MASALAH-MASALAH YANG TERSENARAI SEPANJANG SETAHUN YANG LEPAS.

(Tandakan V untuk jawapan anda)

A26)

A27)

A28)

A29)

A30)

A31)

Pernahkah anda mengalami kesukaran untuk makan dengan menggunakan gigi palsu anda disebabkan masalah padanya?

[| TIDAK BERKAITAN - SAYA TIDAK MEMAKAI GiGi PALSU.

[-|SANGAT | |KERAP [ |KADANG- [| |SEKALI- [| |TIDAK |_ |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengelak daripada senyum disebabkan masalah gigi, mulut

atau gigi palsu anda?

[ISANGAT [| |KERAP [| |KADANG- [| |SEKALI- [| |TIDAK [_|TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah makan anda terganggu seketika disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT [| |KERAP [| |KADANG- [| |SEKALI- (| /TIDAK [_|TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah tidur anda terganggu disebabkan masalah gigi, mulut atau gigi palsu anda?

[ |SANGAT [| |KERAP | |KADANG- | |SEKALI- | /TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda merasa sedih disebabkan masalah gigi, mulut atau gigi palsu anda?

[]SANGAT [| |KERAP | |KADANG- [ |SEKALI- [ (TIDAK |_|TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda berasa sukar untuk relaks disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT [| |KERAP [ /KADANG- [ |SEKALL [ |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

ID:CILILIL)

Pejabat

| are

[a7

[| ars

| | a29

[| A30

[Jas

BERAPA KERAPKAH ANDA MENGALAMI MASALAH-MASALAH YANG TERSENARAI SEPANJANG SETAHUN YANG LEPAS.

(Tandakan V untuk jawapan anda)

A32)

A33)

A34)

A35)

A36)

A37)

Pernahkah anda merasa murung disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT | |KERAP [| |KADANG- | |SEKALI- [ |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah tumpuan anda terganggu disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT | |KERAP | |KADANG- | |SEKALI- [ |TIDAK [ |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda merasa kurang selera makan disebabkan masalah gigi, muiut atau gigi palsu anda?

[ISANGAT | |KERAP | |KADANG- [| |SEKAL- | /TIDAK [ ]TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengelak daripada keluar berjalan-jalan disebabkan masalah gigi, mulut atau gigi palsu anda?

[-JSANGAT [| |KERAP | |KADANG- [ |SEKALI- [ |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda menjadi kurang bertolak-ansur dengan pasangan atau

keluarga anda disebabkan masalah gigi, mulut atau gigi palsu anda?

[]SANGAT [ |KERAP [| |KADANG- [ |SEKALI- [| /TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda tidak dapat bergaul dengan orang lain disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT [| |KERAP [ |KADANG- [| |SEKALL- [| |TIDAK ( |TIDAK KERAP KADANG SEKALA PERNAH TAHU

IDS ICILILI

Pejabat

| | a32

[| A33

[| A34

[| A35

[| a36

das

BERAPA KERAPKAH ANDA MENGALAMI MASALAH-MASALAH YANG TERSENARAI SEPANJANG SETAHUN YANG LEPAS.

(Tandakan V untuk jawapan anda)

A38)

A39)

A40)

A41)

A42)

A43)

Pernahkah anda menjadi cepat marah terhadap orang lain disebabkan masalah gigi, mulut atau gigi palsu anda?

[|ISANGAT [| |KERAP [| |KADANG- [ |SEKALI- [ |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengalami masalah untuk menjalankan kerja-kerja harian anda disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT [| |KERAP [| |KADANG- [| |SEKALI- [_ |TIDAK [ |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah masaiah gigi, mulut atau gigi palsu anda menyebabkan anda merasa tidak sihat?

[|ISANGAT [| |KERAP [| |KADANG- | |SEKALI- | |TIDAK | \TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda terpaksa mengeluarkan perbelanjaan yang tinggi

disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT | |KERAP [| |KADANG- [ |SEKALI- [| |TIDAK []TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda merasa kurang gembira berada bersama dengan orang lain disebabkan masaiah gigi, mulut atau gigi palsu anda?

[]SANGAT [| |KERAP [| |KADANG- | |SEKALI- | |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda merasa kurang berpuashati dengan kehidupan anda disebabkan masalah gigi, mulut atau gigi palsu anda?

[JSANGAT [ |KERAP [| |KADANG- [| |SEKALI- [| |TIDAK [ |TIDAK KERAP KADANG SEKALA PERNAH TAHU

IDs CILIC)

Pejabat

| lass

[lass

[| A40

[| Adl

[| Aq

[| A43

BERAPA KERAPKAH ANDA MENGALAMI MASALAH-MASALAH YANG TERSENARAI SEPANJANG SETAHUN YANG LEPAS.

(Tandakan V untuk jawapan anda)

A44) Pernahkah anda merasa kurang yakin dengan diri anda disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT | |KERAP | /KADANG- | |SEKALI- [| TIDAK (_ |TIDAK KERAP KADANG SEKALA PERNAH TAHU

A45) Pernahkah anda menghadapi masalah untuk melaksanakan kerja sepenuh kemampuan anda disebabkan masalah gigi, mulut atau gigi palsu anda?

[ISANGAT [| |KERAP | |KADANG- [ |SEKALI- | /|TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

BAHAGIAN B

Pejabat

[| Add

[| A45

Bahagian ini adalah bertujuan untuk mengetahui pendapat anda secara keseluruhan berkaitan dengan kesihatan mulut anda. Apa yang anda perlu lakukan ialah hanya

menandakan (V) pada kotak yang disediakan untuk setiap soalan dibawah ini.

Bi) Pada pendapat anda, kesihatan mulut anda pada tahap?

| |SANGAT | |BAIK | _|SEDERHANA | ]BURUK BAIK

B2) Adakah anda merasakan yang anda memeriukan rawatan pergigian?

[ |YA | |TIDAK

B3) Adakah anda berpuashati dengan kesihatan mulut anda sekarang?

| |YA | |TIDAK

ID:CICILIT

Pejabat

(|p

[|e

[lps

BAHAGIAN C Bahagian ini adalah berkenaan gigi dan gigi palsu anda. Anda hanya perlu tandakan (7) pada kotak yang disediakan untuk soalan-soalan berikut.

C1) Adakah anda telah kehilangan kesemua gigi bahagian atas? a

[_]YA | |TIDAK

C2) Adakah anda telah kehilangan kesemua gigi bahagian bawah? [le

[YA | |TIDAK

C3) Adakah anda memakai gigi palsu atas? lle

[_]YA | |TIDAK

C4) Adakah anda memakai gigi palsu bawah? | ea

_|YA | |TIDAK

BAHAGIAN D Bahagian ini adalah berkenaan latarbelakang anda. Anda hanya perlu isikan tempat kosong

yang disediakan atau tandakan (Vv) yang mana berkenaan pada soalan-soalan berikut.

Pejabat

D1) Tarikh lahir anda: / { [ip Hari bulan Tahun

D2) Jantina: | |Lelaki | |Perempuan | _|pe

D3) Bangsa: [ |Melayu [ |Cina | lindia [ |Lain-lain [lps (Nyatakan:....... ee )

D4) Tahap pendidikan tertinggi: | |Tidak bersekolah [_|pa

| \Sekolah Rendah

| _|Sekolah Menengah

| Universiti

[ |Lain-lain, (Nyatakan:........0000c6cccccceeces ) KAMI MENGUCAPKAN RIBUAN TERIMA KASIH.

IDC ICILIC] 10

APPENDIX F

Malaysian Oral Health Impact Profile Questionnaire — Short Form

S-OHIP(M)

200

UNIVERSITI MALAYA

JABATAN PERGIGIAN MASYARAKAT

FAKULTI PERGIGIAN, UNIVERSITI MALAYA

50603 KUALA LUMPUR MALAYSIA Department of Community Dentistry

Faculty of Dentistry, University of Malaya

50603 Kuala Lumpur Malaysia

BORANG SOAL SELIDIK PROFIL KESAN KESIHATAN MULUT

S-OHIP(M)

mL ICICI]

KAM! AMAT MENGHARGAI JASA BAIK ANDA MELUANGKAN SEDIKIT MASA UNTUK MENJAWAB BORANG SOAL SELIDIK INI.

BAHAGIAN A Bahagian ini ingin mengetahui sejauh mana masalah berkaitan dengan gigi, mulut atau gigi palsu mengganggu kehidupan seharian anda. Kami ingin mengetahui berapa kerap anda pernah mengalami masalah yang tersenarai sepanjang SETAHUN YANG LEPAS. Apa yang

anda perlu lakukan ialah menanda (Vv) pada kotak yang disediakan untuk setiap soalan- soalan berikut.

BERAPA KERAPKAH ANDA MENGALAMI MASALAH-MASALAH YANG TERSENARAI SEPANJANG SETAHUN YANG LEPAS.

(Tandakan ¥ untuk jawapan anda)

Pejabat

A1) Pernahkah anda mengalami kesukaran mengunyah sebarang makanan Plat disebabkan masalah gigi, mulut atau gigi palsu anda?

| |ISANGAT [ |KERAP [| |KADANG- [| |SEKALL- [ |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

A2) Pernahkah anda merasakan yang masalah gigi, mulut atau gigi palsu anda [laa menyebabkan nafas anda berbau?

| |ISANGAT [ |KERAP | |KADANG- [_|SEKALI- [ |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

A3) Pernahkah anda mengalami rasa tidak selesa untuk makan sebarang (dan

makanan disebabkan masalah gigi, mulut atau gigi palsu anda?

[ ]SANGAT [ |KERAP [ |KADANG- [ |SEKALI- [ |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

AA) Pernahkah anda mengalami tompok-tompok putih yang pedih (Ulser) di | lar

dalam mulut?

[ |SANGAT [| |KERAP = [|KADANG- [| |SEKALI- [ |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

A5) Pernahkah anda merasa tidak selesa disebabkan makanan terlekat di celah [Als gigi atau gigi paisu anda?

| ISANGAT [ |/KERAP [ |KADANG- | |SEKAL-E [| |TIDAK [ |TIDAK KERAP KADANG SEKALA PERNAH TAHU

wm OOOo 2

BERAPA KERAPKAH ANDA MENGALAMI MASALAH-MASALAH YANG TERSENARAI SEPANJANG SETAHUN YANG LEPAS.

(Tandakan V untuk jawapan anda)

A6)

A7)

A8)

AQ)

A10)

A11)

Pernahkah anda merasa malu disebabkan masalah gigi, mulut atau gigi palsu anda?

| |SANGAT | |KERAP [_|KADANG- | |SEKALI- [ |TIDAK | ITIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengelak daripada memakan makanan tertentu disebabkan

masalah gigi, mulut atau gigi palsu anda?

[ JSANGAT [| |KERAP [| |KADANG- [| |SEKALI- | |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengelak daripada senyum disebabkan masalah gigi, mulut

atau gigi paisu anda?

[-|SANGAT [| |KERAP [| |KADANG- | |SEKALI- [_|TIDAK [ |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah tidur anda terganggu disebabkan masaiah gigi, mulut atau gigi palsu anda?

[-|SANGAT | |KERAP [_|KADANG- [| |SEKALI- [| |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah tumpuan anda terganggu disebabkan masalah gigi, mulut atau

gigi palsu anda?

[ ISANGAT | /KERAP [| |KADANG- [| |SEKALI- [ |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda mengelak daripada keluar berjalan-jalan disebabkan masalah gigi, mulut atau gigi palsu anda?

[ |SANGAT [_|KERAP [| /KADANG- | |SEKAL- [| |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

wri JC ILL)

Pejabat

[jai

[jaz

[aor

[| azo

[| A33

[| A3S

BERAPA KERAPKAH ANDA MENGALAMI MASALAH-MASALAH YANG TERSENARAI SEPANJANG SETAHUN YANG LEPAS.

(Tandakan ¥ untuk jawapan anda)

A12)

A13)

A14)

Pernahkah anda mengalami masalah untuk menjalankan kerja-kerja harian

anda disebabkan masalah gigi, mulut atau gigi palsu anda?

| |SANGAT | |KERAP [| |KADANG- [ |SEKALI- [ |TIDAK | |TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda terpaksa mengeluarkan perbelanjaan yang tinggi disebabkan masalah gigi, mulut atau gigi palsu anda?

[ ]|SANGAT | |KERAP [| |KADANG- [ |SEKALI- [ |TIDAK [_|TIDAK KERAP KADANG SEKALA PERNAH TAHU

Pernahkah anda merasa kurang yakin dengan diri anda disebabkan masalah gigi, mulut atau gigi palsu anda?

| |SANGAT [| |KERAP [| |KADANG- [| |SEKALI [| |TIDAK | \TIDAK KERAP KADANG SEKALA PERNAH TAHU

BAHAGIAN B Bahagian ini adalah bertujuan untuk mengetahui pendapat anda secara keseluruhan berkaitan dengan kesihatan mulut anda. Apa yang anda perlu lakukan ialah hanya menandakan (V) pada kotak yang disediakan untuk setiap soalan dibawah ini.

B1)

B2)

B3)

Pada pendapat anda, kesihatan mulut anda pada tahap?

[_]SANGAT ( |BAIK |_|SEDERHANA [_ |BURUK BAIK

Adakah anda merasakan yang anda memerlukan rawatan pergigian?

[YA | |TIDAK

Adakah anda berpuashati dengan kesihatan mulut anda sekarang?

[_IYA | |TIDAK

wm: IL ILI]

Pejabat

[__]a39

[lai

[| aaa

Pejabat

[|p

| |p

| lps

BAHAGIAN C

Bahagian ini adalah berkenaan gigi dan gigi palsu anda. Anda hanya periu tandakan (V) pada kotak yang disediakan untuk soalan-soalan berikut.

C1) Adakah anda telah kehilangan kesemua gigi bahagian atas? Ma

[ \YA [_ |TIDAK

C2) Adakah anda telah kehilangan kesemua gigi bahagian bawah? | le

[ |YA |_|TIDAK

C3) Adakah anda memakai gigi palsu atas? lle

[ lYA | |TIDAK

C4) Adakah anda memakai gigi palsu bawah? ( Jea

| YA | |TIDAK

BAHAGIAN D Bahagian ini adalah berkenaan latarbelakang anda. Anda hanya perlu isikan tempat kosong yang disediakan atau tandakan (V) yang mana berkenaan pada soalan-soalan berikut.

Pejabat

D1) Tarikh lahir anda: / / [|p Hari bulan Tahun

D2) Jantina: | |Lelaki | |Perempuan [|p

D3) Bangsa: [ |Melayu [| /Cina | lindia [|Lain-lain [Ips (Nyatakan:.........c ccc )

D4) Tahap pendidikan tertinggi: |_|Tidak bersekolah [ |ps

| Sekolah Rendah

|__|Sekolah Menengah

| |Universiti KAM! MENGUCAPKAN RIBUAN TERIMA KASIH.

wi JOO 5

APPENDIX G

Introductory Letter - Mail Respondents

201

UNIVERSITY MALAYA

JABATAN PERGIGIAN MASYARAKAT

FAKULTI PERGIGIAN, UNIVERSITI MALAYA 50603 KUALA LUMPUR MALAYSIA

Department of Community Dentistry

Faculty of Dentistry, University of Malaya Tel: 603-7967 4805 50603 Kuala Lumpur Malaysia Fax: 603-7967 4532

Tarikh: ..... 0c... eee ee

Tuan/Puan,

Kaji selidik kesan kesihatan mulut masyarakat Malaysia.

Saya sedang menjalankan kajian untuk mengetahui sejauh mana penyakit pergigian memberi kesan ke atas kualiti hidup seseorang. Ini adalah sejajar dengan matlamat Kementerian Kesihatan untuk melahirkan masyarakat yang sihat dan mempunyai kualiti hidup yang tinggi. Oleh itu maklumat berkenaan dengan kesan penyakit pergigian adalah perlu bagi membolehkan kerajaan merancang perkhidmatan pergigian Malaysia dengan lebih berkesan dan seterusnya melahirkan masyarakat sihat.

Maka dengan itu saya meminta jasabaik tuan/puan untuk mengisi borang kajiselidik yang dikepilkan dan mengembalikan borang yang telah lengkap diisi kepada saya dengan mengunakan sampul surat yang disertakan bersama. Jawapan dari tuan/puan adalah penting untuk memastikan satu gambaran sebenar berkenaan perkara tersebut.

Di sini saya memberi jaminan kepada tuan/puan bahawa segala maklumat yang tuan/puan berikan adalah rahsia. Nama tuan/puan tidak akan digunakakn didalam sebarang terbitan hasil dari kajiselidik ini.

Saya sertakan juga sebatang pen sebagai tanda terima kasih saya yang tidak terhingga kepada tuan/puan yang telah membantu saya di dalam projek ini.

Sekian, terlma kasih.

Yang benar,

Dr Rosian Saub

Email: roslans@um.edu.my

Tel(off) : 03-7967 4551

PENCEGAHAN PUNCA KESTHATAN

APPENDIX H

Reminder Postcard — Mail Respondents

202

Tarikh:

Tuan/Puan,

Borang Kaji Selidik Profil Kesan Kesihatan Mulut

Saya ingin mengucapkan ribuan terima kasih kerana telah memulangkan borang kaji

selidik di atas. Jika tuan/puan belum berbuat demikian, saya sangat berharap jasa baik

tuan/puan untuk mengisi dan memulangkannya kepada saya dengan seberapa segera

kerana jawapan tuan/puan adalah amat penting.

Jika tuan/puan telah kehilangan borang tersebut, tuan/puan boleh menghubungi

Jururawat Pauline atau Puan Nani di talian berikut 03-7967 4856, 03-7967 4876 atau

emailkan kepada Dr Roslan Saub di alamat os . dan kami akan kirimkan

kepada tuan/puan borang yang baru.

Sekian, terima Kasih

Yang benar

Dr Roslan Saub

Setem Kepada

Jabatan Pergigian Masyarakat

Fakulti Pergigian Universiti Malaya

Lembah Pantai

50603 Kuala Lumpur

APPENDIX I

Second Reminder Letter — Mail Respondents

203

UNIVERSITY MALAYA

JABATAN PERGIGIAN MASYARAKAT

FAKULTI PERGIGIAN, UNIVERSITI MALAYA

50603 KDALA LUMPUR MALAYSIA

Department of Community Dentistry

Faculty of Dentistry, University of Malaya Tel: 603-7967 4805 50603 Kuala Lumpur Malaysia Fax: 603-7967 4532

Tuan/Puan,

Kaji selidik kesan kesihatan mulut masyarakat Malaysia.

Lebih kurang tiga minggu yang lepas saya ada menghantar borang kajiselidik kepada tuan/puan untuk mengetahui sejauh mana penyakit mulut memberi kesan ke atas kualiti hidup tuan/puan. Sehingga setakat ini saya belum lagi menerima borang jawaban dari tuan/puan.

Demi untuk mencapai matlamat Kementerian Kesihatan Malaysia untuk melahirkan masyarakat yang sihat dan mempunyai kualiti hidup yang tinggi, kerajaan perlu merancang perkhidmatan pergigian dengan lebih berkesan. Untuk tujuan tersebut, maklumat berkenaan dengan kesan penyakit pergigian adalah amat perlu. Maka dengan itu saya meminta jasabaik tuan/puan untuk mengisi borang kajiselidik tersebut kerana setiap jawapan dari tuan/puan adalah amat penting untuk kajian ini.

Bersama ini saya kepilkan sekali lagi borang kaji selidik tersebut. Saya sertakan juga sampul surat bersetem bagi memudahkan tuan/puan untuk mengembalikannya kepada saya.

Di sini saya memberi jaminan kepada tuan/puan bahawa segala maklumat yang tuan/puan berikan adalah rahsia. Nama tuan/puan tidak akan digunakan di dalam sebarang terbitan hasil dari kajiselidik ini.

Kerjasama tuan/puan amat saya hargai.

Sekian, terima kasih.

Yang benar,

Dr Roslan Saub

Penyelidik

Email: roslans@um.edu.my Tel(off) 03-7967 4551/4805

PENCEGAHAN PUNCA KESTHATAN

APPENDIX J

Identification Letters — The Interviewers

204

UNIVERSITY MALAYA

JABATAN PERGIGIAN MASYARAKAT FAKULTI PERGIGIAN, UNIVERSITI MALAYA 50603 KUALA LUMPUR MALAYSIA

Department of Community Dentistry

Faculty of Dentistry, University of Malaya Tel: 603-7967 4805 50603 Kuala Lumpur Malaysia Fax: 603-7967 4532

SURAT PENGENALAN

Tuan/ Puan,

KAJISELIDIK KESAN KESIHATAN MULUT KE ATAS KEHIDUPAN HARIAN MASYARAKAT MALAYSIA.

Pembawa surat ini adalah seorang pelajar di Fakulti Pergigian, Universiti Malaya, yang telah dilantik dan dilatih oleh saya untuk menjalankan temubual dengan tuan/puan bagi tujuan kaji selidik di atas. Nama dan alamat beliau adalah seperti berikut:

Nama: 00... cece ceeeeecceneeceeneceeasenes

Alamat: ..0 000.000 cece ceeec ee ccaceceeesseeesesecesnecevaeeenenaeess

No. telefon: .........0.......0008

2) Jika tuan/puan mempunyai sebarang kemusykilan, tuan/puan bolehlah menghubungi saya seperti berikut:

Nama: Dr Roslan Saub Alamat tempat kerja: Jabatan Pergigian Masyarakat,

Fakulti Pergigian, Universiti Malaya

50603, Kuala Lumpur No. telefon tempat kerja: 03-7967455 1/4805 Email: roslans@um.edu.my

Sekian, terima kasih

Yang benar,

Dr Roslan Saub

Penyelidik Email: roslans@um.edu.my

Tel(off — direct) : 603-7967 4551

PENCEGAHAN PUNCA KESIHATAN

APPENDIX k

Appointment Card — Interview Respondents

205

Tarikh:

Tuan/Puan,

Kajian Kesan Kesihatan Mulut Ke Atas Kehidupan Seharian Masyarakat Malaysia

Saya telah hadir di rumah tuan/puan pada ........ fiw. ee , pukul ........ (pagi/petang) untuk menemuramah tuan/puan berkenaan kajian di atas tetapi tuan/puan tiada di rumah. Saya berharap tuan/puan dapat menghubungji saya di talian......................08 bagi membolehkan saya membuat temujanji dengan tuan/puan.

Sebarang pertanyaan berkenaan kaji selidik ini bolehlah dikemukan kepada Dr Roslan Saub (penyelidik) di talian 03-79674551/4805 atau emailkan kepada beliau di alamat

Sekian, terima Kasih

Yang benar

( ) Penemuramah

Setem Kepada

Jabatan Pergigian Masyarakat

Fakulti Pergigian

Universiti Malaya

Lembah Pantai

50603 Kuala Lumpur

APPENDIX L

Letter to the Police

206

UNIVERSITY MALAYA

JABATAN PERGIGIAN MASYARAKAT FAKULTI PERGIGIAN, UNIVERSITI MALAYA 50603 KUALA LUMPUR MALAYSIA

Department of Community Dentistry

° : . Tel: 603-7967 4805 Faculty of Dentistry, University of Malaya Fax: 603.7967 4532

50603 Kuala Lumpur Malaysia

24 Mei, 2002.

Ketua,

Polis Daerah Petaling

Tbu Pejabat Polis Petaling Jaya

46050 Petaling Jaya Selangor

Tuan,

KAJI SELIDIK KESAN PENYAKIT MULUT KE ATAS KEHIDUPAN SEHARIAN MASYARAKAT MALAYSIA.

Adalah dengan segala hormatnya saya merujuk kepada perkara tersebut di atas.

Dimaklumkan bahawa saya Dr Roslan Saub dari Jabatan Pergigian Masyarakat, Fakulti Pergigian, Universiti Malaya akan menjalankan satu kaji selidik seperti di atas. Kaji selidik

ini akan melibatkan penghuni-penghuni rumah yang berusia 18 tahun ke atas yang telah dipilih dalam kawasan tuan. Saya atau penemurah yang telah dilantik oleh saya akan membuat lawatan dari rumah ke rumah untuk menemubual dengan penghuni yang terpilih dalam proses pengumpulan data. Projek ini akan bermula pada pertengahan bulan Jun 2002 sehingga Ogos 2002.

Sekian dimaklumkan. Terima kasih

Yang benar,

Dr Roslan Saub

Email: roslans@um.edu.my I a eee t ERE SEE e

Tel(off) : 603-7967 4551/4805

PENCEGAHAN PUNCA KESTHATAN

UNIVERSITY MALAYA

JABATAN PERGIGIAN MASYARAKAT

FAKULTI PERGIGIAN, UNIVERSITI MALAYA

50603 KUALA LUMPUR MALAYSIA

Department of Community Dentisiry

Faculty of Dentistry, University of Malaya Tel: 603-7967 4805 50603 Kuala Lumpur Malaysia Fax: 603-7967 4532

24 Mei, 2002.

Ketua,

Polis Daerah Shah Alam

Polis Di Raja Malaysia Persiaran Damai

Sekyen 11 40000 Shah Alam

Selangor

Tuan,

KAJI SELIDIK KESAN PENYAKIT MUOLUT KE ATAS KEHIDUPAN SEHARIAN MASYARAKAT MALAYSIA.

Adalah dengan segala hormatnya saya merujuk kepada perkara tersebut di atas.

Dimaklumkan bahawa saya Dr Roslan Saub dari Jabatan Pergigian Masyarakat, Fakulti Pergigian, Universiti Malaya akan menjalankan satu kaji selidik seperti di atas. Kaji selidik ini akan melibatkan penghuni-penghuni rumah yang berusia 18 tabun ke atas yang telah

dipilih dalam kawasan tuan. Saya atau penemurah yang telah dilantik oleh saya akan membuat lawatan dari rumah ke rumah untuk menemubual dengan penghuni yang terpilih dalam proses pengumpulan data. Projek ini akan bermula pada pertengahan bulan Jun 2002

sehingga Ogos 2002.

Sekian dimaklumkan. Terima kasih

Yang benar,

Dr Roslan Saub Email: roslans@um.edumy

Tel(off) : 603-7967 4551/4805

PENCEGAHAN PUNCA KESIHATAN

APPENDIX M

Introductory Letter From The Dean — Interview Respondents

207

UNIVERSITI MALAYA Pejabat Dekan/Office of the Dean Fakulti Pergigian/Faculn’ of Dentistry Universiti Malava

50603 Kuala Lumpur Malaysia

Tel No: 6-03-79674800

6-03-7956 5143 (DL) Fax : 6-03-7967 4809

e-mail: ishakar@um.edu.my

Dean: Professor Dr. Ishak Abdul Razak, BDS (Malava), DDPHRCS (England), MSe (London), PhD (Malaya) FICD

Bil Tuan/Your ref:

Bil. Kami/Our ref:

Tarikh

Tuan/Puan,

Kajian Kesan Penyakit Mulut ke atas Kehidupan Seharian Masyarakat Malaysia.

Saya ingin memaklumkan kepada tuan/puan bahawa Dr. Roslan Saub dari Jabatan Pergigian Masyarakat, Fakulti Pergigian, Universiti Malaya, sedang menjalankan kajian tersebut di atas untuk mengukur sejauh mana penyakit pergigian memberi kesan ke atas kehidupan seharian tuan/puan. Maklumat ini kelak dapat digunakan didalam merancang perkhidmatan pergigian di Malaysia dengan lebih berkesan dan saksama. Kajian ini adalah merupakan sebahagian dari pengajian Doktor Falsafah (PhD) beliau.

Beliau dan pasukan beliau akan hadir di kawasan tuan/puan untuk menemuramah dengan tuan/puan. Projek ini akan bermula dari bulan Jun 2002 hingga Ogos 2002.

Dengan itu saya amat berharap agar tuan/puan dapat memberikan kerjasama kepada beliau.

Sekian, terima kasih

Yang benar

Fakulti Pergigian Univesiti Malaya.

APPENDIX N

Impact Of Oral Disorders On Daily Lives According To The Domains

208

Impacts of oral disorders on daily lives according to the domains

Domain Individual impacts

Functional Limitation Difficult to chew foods (especially hard foods: nuts, chicken bone, sugar cane)

Can’t bite hard foods.

Can’t pronounce certain words (eg: “s”

Face look ugly “like old person”

Look awkward

Loose denture

Unsatisfied with the look of the denture

Physical Pain Cause headache

Cause pain on the other part of the body (eg. Ear, hand, neck)

Toothache

Pain on the gums

Ulcer

Sensitive to cold water

Denture pain

Psychological Discomfort Feel shy

Worry to laugh or talk.

Feel uncomfortable due to food stuck in between the

teeth or dentures.

Scared that it might worsen

Scared that the tooth next to it will be affected

Bad breath

Feel worry about it

Feel very unhappy

Uncomfortable denture

Physical Disability Avoid certain foods (eg. Sweets, cold drink, hard foods)

Avoid chew on the pain side.

Avoid talking

Avoid laughing Difficult to brush teeth.

Unable to eat favorite foods

Have to cover the mouth when smiling

Have to blend the foods.

Food feel less tasty

Difficult to clean

Have to brush more frequently

Avoid using that side.

Tend to close the mouth Psychological Disability Disturb sleep (have to take pain killer)

Sleep all day long

Unable to concentrate (eg: to other people conversation,

103

study, work).

Lazy to do other works besides sleeping.

Easy to get angry

Loss appetite.

Can’t enjoy the foods

Social Disability Preferred to stay at home.

Tend to be quite

Tend to stand far from people

Handicap Have to take medical certificate (mc)

Financial burden.

Less confident to talk to people