M

C

T

R

l

c

0

d

Development of an instrument to measure seniors’ patient safety

health beliefs: The Seniors Empowerment and Advocacy

in Patient Safety (SEAPS) survey§

Nancy C. Elder a,*, Saundra L. Regan a,1, Harini Pallerla a,1,Linda Levin b,2, Douglas Post c,3, Donald J. Cegela d,4

a Department of Family Medicine, University of Cincinnati, Cincinnati, United Statesb Department of Environmental Health, University of Cincinnati, Cincinnati, United States

c Department of Family Medicine, Ohio State University, Columbus, United Statesd School of Communication and Department of Family Medicine, Ohio State University, Columbus, United States

Received 20 February 2007; received in revised form 18 July 2007; accepted 22 July 2007

www.elsevier.com/locate/pateducou

Patient Education and Counseling 69 (2007) 100–107

Abstract

Objective: To develop a survey to measure seniors’ embracement of ambulatory patient safety self-advocacy behaviors, the Senior Empowerment

and Advocacy in Patient Safety (SEAPS) survey.

Methods: Content was developed by review of published recommendations combined with interviews and focus groups with community

members; items were generated for subscales based on the health belief model (HBM). Psychometric characteristics were assessed by cluster and

correlation analyses on a pilot test of 143 community dwelling seniors; the ability of the subscales and demographic variables to predict reported

behavior was investigated by multiple regression.

Results: The four subscales of the SEAPS were outcome efficacy (OE), attitudes (ATT), self-efficacy (SE) and behaviors (BEH). Cronbach alphas

were 0.74 for ATT, 0.79 for BEH, and 0.91 for OE and SE. Analysis of variance showed that there were no differences in any subscale score by race,

education level or frequency of doctor visits, but women were noted to have significantly higher scores ( p < .01) on the ATT and SE subscales and

for the total of all the scales. Multiple regressions showed that SE significantly predicted self-reported behavior ( p < .001). OE was a significant

predictor for whites ( p < .001) but not for African-Americans ( p = .24).

Conclusions: We have developed a short, 21-item self-administered survey to assess seniors’ views about their participation in safety tasks.

Practice implications: We believe the SEAPS shows promise to be a tool for evaluating interventions and training programs aimed at improving

seniors’ self-advocacy skills. Effective interventions may improve the involvement of patients in their own safety in the clinical setting.

# 2007 Elsevier Ireland Ltd. All rights reserved.

Keywords: Patient safety; Patient communication; Health belief; Ambulatory care

§ This work was supported by a Dean’s Discovery Grant from the College of

edicine, University of Cincinnati.

* Corresponding author at: Department of Family Medicine, University of

incinnati, PO Box 670582, Cincinnati, OH 45267-0582, United States.

el.: +1 513 558 1430; fax: +1 513 558 3266.

E-mail addresses: eldernc@fammed.uc.edu (N.C. Elder),

egansl@fammed.uc.edu (S.L. Regan), Pallerh@fammed.uc.edu (H. Pallerla),

evinls@UCMAIL.UC.EDU (L. Levin), Doug.Post@osumc.edu (D. Post),

egala.1@osu.edu (D.J. Cegela).1 Tel.: +1 513 558 1430.2 Tel.: +1 513 558 4838.3 Tel.: +1 614 293 8087.4 Tel.: +1 614 292 3675.

738-3991/$ – see front matter # 2007 Elsevier Ireland Ltd. All rights reserved.

oi:10.1016/j.pec.2007.07.007

1. Introduction

Utilization of strategies to increase patients’ involvement in

their health care can improve health outcomes [1]. Teaching

patients to effectively communicate with clinicians can

enhance patient participation. Findings suggest that patient

communication training improves important clinical indicators,

including chronic disease outcomes, functional status and

adherence to treatment recommendations [2–4].

Patient safety, the freedom from accidental injury in health

care, has also become an issue of importance for patients [5–7].

Recommendations for patients to be more involved in the safety

of their own care [8–10] have been utilized by patients to varying

N.C. Elder et al. / Patient Education and Counseling 69 (2007) 100–107 101

degrees [5,7,11]. In the ambulatory setting, where most health

care in America takes place [12], patients’ activities can

influence clinical outcomes, and this has been postulated for

patient safety, too [13]. For this reason, recommendations to

patients flow freely [8,9] and educators and researchers are

developing interventions to assist patients in understanding and

participating in safety oriented health behaviors [14]. However,

evaluating the effectiveness of such interventions can be difficult.

There are several theories that try to explain the health

behaviors of individuals [15]. The health belief model (HBM)

[16,17] posits that several variables are necessary prior to

individuals participating in specific health behaviors. These

include being able to overcome the barriers and perceiving

benefit from the recommended behaviors as well as having

confidence in one’s ability to effectively take action [7,17]. This

last point refers to having self-efficacy, which several authors

have added to the original HBM [7,18–20].

As patients are being instructed to take a more active role in

improving the safety of their ambulatory care, there is a need to

objectively evaluate their activities. Seniors are especially at

risk for errors and safety lapses, both because they use more

health care and because they are less likely to participate in

active medical decision making [21–24]. However, it is not

even clear what the most important safety behaviors are for

elderly patients in the ambulatory setting, as most published

safety recommendations have been developed with little patient

input and focus mainly on hospital care [14]. While some

programs have tried training patients to follow some published

safety recommendations, their evaluation methods are often site

specific and not applicable to large groups of patients [13].

Since seniors are the largest consumers of health care [25], we

designed a survey based on the HBM geared to this population,

developed with patient as well as expert input and focused on

safety issues of importance in the ambulatory setting. We

believe such an instrument could supplement difficult to

measure objective outcomes of patient focused ambulatory

safety interventions and education programs [26]. This paper

describes the development and pilot testing of a survey

instrument to measure seniors’ participation and embracement

of ambulatory patient safety self-advocacy behaviors.

2. Methods

To develop the Seniors Empowerment and Advocacy for

Patient Safety (SEAPS) survey, we used a multi-step process to

address both content and construct validity. We elected to focus

on three subscales:

� O

utcome efficacy (OE) the belief that the actions will be a

benefit to one’s health,

� A

ttitudes (ATT) concerns about barriers to participating in the

actions,

� S

Fig. 1. Steps in the development of the SEAPS survey.

elf-efficacy (SE) confidence in one’s ability to effectively

take action.

We also included a self-report proxy for the performance of

actions called behaviors (BEH). This allows predictors of

behavior to be determined and also provides the ability to assess

self-reported changes in patient safety behaviors which may

result from future interventions. This research received approval

from the University of Cincinnati Institutional Review Board.

2.1. Item generation and selection

Since there is not a ‘‘gold standard’’ for patient self-

advocacy behaviors for ambulatory care, it was necessary for us

to develop the content from multiple sources in order to achieve

content validity (Fig. 1).

2.1.1. Initial safety recommendations

In a previous study, we interviewed 24 primary care patients

who had experienced medical error or preventable problems in

their health care; these patients described what health care

behaviors they now used to improve their chances for quality

and safe care [5]. At this same time, we reviewed published

recommendations from organizations such as the National

Patient Safety Foundation (NPSF) [27], the Agency for

Healthcare Quality and Safety (AHRQ) [8], the Joint

N.C. Elder et al. / Patient Education and Counseling 69 (2007) 100–107102

Commission for the Accreditation of Healthcare Organizations

(JCAHO) [9] and others [10,28,29]. In addition, to ensure that

the recommendations to patients were applicable to important

medical problems, we also reviewed the medical literature

describing common errors and problems in primary care

[26,30–32]. Combining these sources, we collated an initial list

of safety recommendations applicable to ambulatory care.

2.1.2. Final safety recommendations

In order to solicit further patient input into the recommenda-

tions, we conducted eight focus groups with volunteers from

the community. Participants were solicited for these groups

through flyers posted in primary care physicians’ offices, in

community, senior and recreation centers and on community

bulletin boards around town. All ages of adults were invited,

and 79 people (mean age = 45, range 20–78, 18% over age 60)

participated and discussed their experiences with using the

recommendations. From this information the list of safety

recommendations for patients was refined, and after additional

medical literature review, we collated a final list of 10 primary

care safety recommendations (Table 1).

2.1.3. Development of SEAPS 40 question draft

The recommendations were formulated into questions for

the subscales adapted from the health belief model: outcome

efficacy, attitudes, self-efficacy and behaviors. The questions

were then reviewed via e-mail by a small group of primary care,

patient safety and patient education researchers and advocates.

2.2. Psychometrics

2.2.1. Pretest

The SEAPS 40 was pretested on two groups of independent

living elderly individuals. Participants completed the survey

and participated in a ‘‘think aloud’’ focus group discussion

about the survey. Group 1 (nine African-American women,

mean age 75.4 years) met at an urban senior center, while group

2 (eight white women and four white men, mean age 84.7) met

Table 1

Initial safety recommendations for patients

Patients educate themselves on their own health concerns and medications

Patients give providers a complete and thorough story of their health

problems and concerns

Patients ask questions of their providers until they understand medical

diagnoses, medications prescribed, tests ordered and future plans

Patients speak up about problems with office staff and office procedures

that interfere with the ability to access care

Patients keep a list of their medications, dietary supplements and drug

allergies

Patients double check medication prescriptions at the pharmacy to assure

they have received the name and doses of the medicine they expect

Patients double check on the results of laboratory or diagnostic testing to

assure they have received results for all tests done

Patients access a second opinion when needed

Patients are willing to change providers if they feel health concerns are

not being met

Patients bring a friend or family member to doctors’ visits and for

procedures and tests to serve as an advocate and helper

at a suburban retirement community. The average time to

complete the survey was 12.4 min for group 1 and 13.8 min for

group 2. After review of participant comments and survey

results, specific changes made at this point included re-wording

of the outcome efficacy questions and simplification of

language.

2.2.2. Pilot test

The SEAPS 40 was then pilot tested with a sample of

community dwelling elderly to assess baseline response

characteristics, item reproducibility and to decrease the total

number of questions in the survey.

2.2.2.1. Cohort description. Participants were solicited at

multiple locations to insure a range of ages, race (white and

African-American (AA)) and socioeconomic classes. These

locations included waiting rooms in the University Family

Medicine health centers, senior centers, and social functions at

several churches. Inclusion criteria included age over 59, living

independently and English speaking. Exclusion criteria

included significant cognitive impairment. For those with

low literacy or visual problems, the survey was read by research

assistants.

Two hundred seniors were approached and 150 (75%)

completed the survey, 50 (25%) refused; of the 150, 143

answered enough questions to be used for analysis. All

participants were recruited in the Cincinnati metropolitan area

(urban and suburban). The mean age of the participants was 70

(range 60–97) years, 71% were female, 62% were white, and

73% were at least high school graduates. Over 99% of the

participants had medical coverage (Medicare, Medicaid,

private insurance or a combination) and visited a doctor at

least once a year.

2.2.2.2. Descriptive statistics, survey accuracy and reliabili-

ty. Each survey question was answered on a Likert-type scale

from zero to three. Negatively worded questions were reverse

coded, so that three indicated increased self-efficacy, increased

outcome efficacy, positive attitudes and more frequent

behaviors. Missing responses were replaced by the subject’s

modal response to answered questions within the same

subscale. We calculated the mean and standard deviation of

each question across subjects to compare averages and ranges.

Subscale means for all subjects were calculated by averaging

question responses for each subject, then averaging the

subject-specific means across subjects. Since a normal

approximation to the distribution of subscale means was

reasonable, we calculated the percent accuracy of each

subscale as 1.96 times the standard error of the subscale

average (or one-half the width of the 95% confidence interval

of the mean). We used Cronbach alphas to estimate the

reliability of questions for each subscale.

Subscale accuracies ranged from 4.6 to 6.0%, indicating that

95% of the time the error of measurement (difference between

observed and true subscale means) was within 4.6–6.0% of the

observed subscale mean. Cronbach alphas were 0.74 for ATT,

0.79 for BEH, and 0.91 for OE and SE.

N.C. Elder et al. / Patient Education and Counseling 69 (2007) 100–107 103

2.2.2.3. Questionnaire reduction. Subscale-specific question

reduction was implemented by performing cluster analyses in

which clusters of questions were developed that could be

represented by a single variable. Seventy-five percent of the

variability of all questions in each subscale was explained by

five and/or six clusters. These analyses were followed by

hierarchical cluster analyses, involving nine stages (the

number of original questions minus one) in which the two

‘‘nearest’’ questions or clusters of questions were combined

sequentially based on the Spearman similarity index as a

distance measure representing clustering. Clusters were

similar to those obtained by the initial analyses. Graphical

representations of hierarchical clusters were obtained that

facilitated the selection of questions which represented the

original sets of clusters. Medians (minimum, maximum)

values of the resulting (nine) distances indices for each

subscale were: ATT 0.10 (0.00, 0.34), OE 0.37 (0.05, 0.51),

BEL 0.26 (0.06, 0.45), and BEH 0.13 (0.01, 0.45). Cluster

analyses were performed using the SAS procedure PROC

VARCLUS and the S-PLUS function hclust.

Clustering of questions on each subscale was also assessed

by race and gender, since a priori it was judged important to

keep those questions that contained information specific to

race and gender differences. Based on the magnitude of

Spearman indices across race–gender subgroups, with

priority given to the results for all subjects, three different

subsets of five or six questions were selected from each

subscale. The final subset of questions from each subscale

Table 2

Means for 21-question final SEAPS for health belief model subscales and total SE

Category of (N) Means and 95% confidence intervals

BEH ATT

Gender

Female (101) 1.6 (1.5, 1.8) 2.0 (1.9, 2.1)

Male (42) 1.4 (1.3, 1.6) 1.7 (1.5, 1.9)

p-Value .06 <.01

Race

AA (50) 1.7 (1.5, 1.8) 1.9 (1.8, 2.1)

White (93) 1.5 (1.4, 1.7) 1.9 (1.7, 2.0)

p-Value .25 .65

Education

<HS (39) 1.5 (1.3, 1.7) 1.9 (1.7, 2.1)

HS (53) 1.6 (1.4, 1.7) 1.8 (1.7, 2.0)

<College (25) 1.8 (1.5, 2.0) 2.0 (1.8, 2.3)

College (26) 1.6 (1.4, 1.9) 1.9 (1.7, 2.1)

p-Value .25 .54

Doctor office visits

Occas. (100) 1.6 (1.5, 1.7) 1.9 (1.8, 2.0)

Frequent (43) 1.6 (1.4, 1.8) 1.9 (1.7, 2.0)

p-Value .71 .90

All (143) 1.6 (1.5, 1.7) 1.9 (1.8, 2.0)

Means, 95% confidence intervals and p-values by analysis of variance. N = num

SE = self-efficacy subscales; A = African-American; HS = High school; Occas. = s

was chosen as follows: race–gender means by subscale were

calculated for each subset of questions. The subset of

questions from each subscale whose means were on average

closest to the race–gender means calculated from all 10

questions was chosen for the analyses. Spearman correlations

between subscale means ranged from 0.4 to 0.6. The

correlations between BEH and the other subscales were:

ATT 0.42, SE 0.38, and OE 0.51.

2.2.2.4. Comparison of original SEAPS 40 question survey

and final SEAPS. Nine of the original 10 safety recommenda-

tions for patients remained in the final SEAPS. The

recommendation for patients double checking their prescrip-

tions from the pharmacy was removed for two reasons—in all

subcategories, it received the highest mean scores from survey

participants, indicating that they already were comfortable with

and doing this activity, and published medical literature noted

that elderly patients considered this activity to be an

unnecessary recommendation [25].

For each subscale, we performed analysis of variance of the

final SEAPS data to test differences between race, gender,

education level, and frequency of doctor visits. Although not

shown, these analyses were also performed for the original

SEAPS 40 question survey. We tested the equality of means

between the two versions of the survey by judging the overlap

of 95% confidence intervals. Both subscale averages and 95%

confidence intervals for the SEAPS 40 were in good agreement

with the final SEAPS survey.

APS score by demographic characteristics

OE SE Total

1.9 (1.8, 2.1) 2.3 (2.1, 2.4) 1.9 (1.8, 2.0)

1.8 (1.6, 1.9) 1.9 (1.7, 2.1) 1.7 (1.6, 1.9)

.36 .01 <.01

2.0 (1.8, 2.2) 2.2 (2.0, 2.4) 1.9 (1.8, 2.0)

1.9 (1.8, 2.0) 2.1 (2.0, 2.3) 1.8 (1.7, 1.9)

.37 .65 .33

2.0 (1.8, 2.2) 2.0 (1.8, 2.3) 1.8 (1.7, 2.0)

1.8 (1.7, 2.0) 2.1 (1.9, 2.3) 1.8 (1.7, 2.0)

2.0 (1.7, 2.2) 2.4 (2.2, 2.6) 2.0 (1.8, 2.2)

1.9 (1.7, 2.2) 2.2 (2.0,2.5) 1.9 (1.7, 2.1)

.67 .25 .35

1.9 (1.8, 2.0) 2.2 (2.0, 2.3) 1.9 (1.8, 2.0)

2.0 (1.8, 2.2) 2.2 (2.0, 2.4) 1.9 (1.8, 2.0)

.18 .97 .59

1.9 (1.8, 2.0) 2.2 (2.0, 2.3) 1.9 (1.8, 2.0)

ber of subjects; ATT = attitudes; BEH = behaviors; OE = outcome efficacy;

everal times a year or less; Frequent = once a month or more.

Table 3

Multiple regression analysis of the association between health belief model subscales and demographic characteristics as independent variables, and self-reported

behavior as dependent variable

Predictor variable Magnitude and direction

of effect of predictor variable on BEHa

p-Value measuring significance

of predictor variable

Percent model variability

explained by predictor variable

SE +0.3 <.001 22.7

OE

White +0.4 <.001 28.3

AA �0.1 .24 5.3

ATT +0.1 .15 4.1

Race: AA vs. white +1.1 <.001 32.9

Gender: female vs. male +0.02 .79 0.1

Education: <HS vs. �HS �0.2 .08 6.3

Frequency of office visits: occ. vs. freq. +0.04 .70 0.3

AA = African-American;<HS = less than high school graduation;�HS = high school graduation or greater; occ. = several times a year or less; freq. = once a month

or more.a For SE, OE, and ATT the magnitude of the effect indicates the change in behavior score when the independent variable increases by +1 unit. The effect of OE

differs by race; there were no significant modifying effects of demographic variables for any other subscales. For demographic variables the magnitude of the effect

indicates the difference in mean behavior of subjects in the first category compared to the second category (e.g. AA compared to white or female compared to male).

N.C. Elder et al. / Patient Education and Counseling 69 (2007) 100–107104

2.2.3. Correlation of behavior and other subscales

in the final SEAPS

We determined the ability to predict BEH from other

subscale scores, adjusted for race, gender, education and

frequency of doctor visits by multiple regression analysis.

Regression coefficients were interpreted by calculating their

effect on BEH when a subscale score increased by 1 unit, or

when a categorical variable was equal to 1 (white, male, high

school graduate or greater, frequent doctor visits) versus 0

(African-American, female, less than high school graduation or

occasional doctor visits), controlling for other variables. The

contribution to the variability of BEH scores that was explained

by the model (R2) and the percents of the variability of BEH

explained by each predictor variable were obtained.

In order to investigate the robustness of the regression

analysis for predicting behavior from the other (correlated)

subscale measurements (r = 0.42–0.51), the analysis was

repeated by replacing the three subscale measurements of

each subject by (i) a single variable equal to the first principal

component (PC), which explained 68% of the variation of

subscale measurements, and (ii) by the first two principal

components which explained 90% of the variability. The same

covariates, i.e. race, gender, education, and frequency of

doctor visits were included in each model. The coefficients of

determination (R2) were 0.39, 0.36, and 0.38 when all

subscales, the first PC, and the first two PCs were modeled,

respectively. Predicted values of behavior were almost

identical for the PC analyses and the original analysis. All

analyses were performed using SAS and S-Plus statistical

software.

3. Results

Table 2 shows the total and subscale averages and 95%

confidence intervals for the final SEAPS survey (see

Appendix A). Analysis of variance revealed that there were

no differences in any subscale score by race, education level or

frequency of physician visits, but women were noted to have

significantly higher scores on the ATT and SE scales.

We analyzed the ability of the HBM subscales and

demographic variables to predict self-reported behavior using

multiple regression (Table 3). The SE subscale, measuring an

individual’s confidence they could do a task, was positively and

significantly predictive of self-reported behavior. Race was

found to be a predictor of behavior both in that it modified the

effect of outcome efficacy (which was a significant predictor for

whites but not for African-Americans) and also as an

independent predictor of behavior. The attitude scale, while

positively associated with behavior, was not significant.

Gender, education and frequency of doctor visits were not

significantly related to self-reported behavior.

4. Discussion and conclusion

4.1. Discussion

From the care of chronic health problems to preventing

disease, many patients are taking an active role in their own health

care, and generally benefiting from such [23,33]. Following this

trend, patients in America today are also being called on to be

active participants in their own safety [8–10,34,35]. Yet the few

studies that measure patients’ participation in safety tasks show

mixed results [6,25,36], and these focus primarily on the hospital

setting. There are no validated instruments to assess patients’

beliefs about participating in safety activities. We have created

the Seniors Empowerment and Advocacy in Patient Safety

survey, a short self-administered survey that can assess elderly

persons’ beliefs about being their own patient safety advocate in

ambulatory care. We believe the strength of this survey lies in the

rigor with which we generated the initial questions using a multi-

step process with active patient input. The demographic

correlations we report are consistent with research in similar

areas, such as participation in medical decision making [23,37–

39], adding to the strength of our survey. Thus, along with

N.C. Elder et al. / Patient Education and Counseling 69 (2007) 100–107 105

objective measures of patient behavior, this survey can assist

researchers and educators in assessing patients’ beliefs and

attitudes, and allow assessments of changes after interventions.

Two weaknesses of prior recommendations to patients about

how to improve their own safety are that they were developed

with little direct patient input and have a strong focus on

hospital care [14]. However, most medical care in America is

provided in physicians’ offices, especially primary care offices,

and not hospitals [12]. We used both published recommenda-

tions and direct patient input to develop our survey content, and

focused it on ambulatory care. A series of individual interviews

[5], focus groups and pretesting ensured that the content of our

SEAPS survey was strongly grounded in the experience of

patients, especially elderly patients. Moving patient safety

research out of the hospital to the community is an important

step in improving medical quality and safety [40].

Our survey shows applicability for both African-Americans

and whites, as well as both genders and all levels of educational

attainment and frequency of doctor visits. While we did find

significant differences between women and men in their

attitudes and perceptions of self-efficacy on univariate analysis,

this gender finding is consistent with research on active

decision making and participation in health care, in which

women are regularly found to be more active participants in

health care [23,24,33,38,39].

The three HBM subscales were all positively predictive of

self-reported behavior, although only self-efficacy was sig-

nificant for all participants, consistent with other research on

the HBM and patient safety [6,7]. Race was a significant

modifier for outcome efficacy and was also an independent

predictor of behavior. This is consistent with research showing

the decision to be an active participant in health care is complex

and depends on the interplay of personal, physician and

contextual factors [23,24,37,38,41]. Our findings suggest that

comfort with certain tasks and self-confidence in one’s ability

to perform a task may account for some of these differences

[7,17]. The exact role race, education, gender and illness burden

play when elderly people make decisions about health care is

still uncertain. In published studies, demographic and situa-

tional factors alone explain 20% or less of the variability of

patient preferences in decision making [23].

Appendix A. Final SEAPS survey

How often do you do the following tasks? (behaviors) I never

do this

1. How often do you teach yourself about

your own health problems and medicines?

&

2. How often do you ask your doctors questions

about your health problems, lab tests and medicines?

&

3. How often do you keep an updated a list of all your

medicines, including those from the drug store and

health food store?

&

4. How often do you call the doctor’s office if you

haven’t received the results of laboratory

or X-ray tests?

&

5. How often do you get a second opinion from

another doctor when needed?

&

There are limitations to our survey development. Our

population consisted only of urban and suburban white and

African-American participants in a Midwest City. Other races

or ethnic groups, or those in other regions of the country may

have differences that would have affected our psychometrics or

item selection. As with all surveys asking about self-reported

behaviors, we have no objective corroboration that our subscale

BEH accurately reports what patients really do. However, it was

positively correlated to our other subscales, and we encourage

the use of objective measures along with the SEAPS in future

work. The survey is written at the 8th–9th grade level, which

may be limiting for those of low literacy; however the option to

read it aloud and explain questions is possible. A limiting factor

to the detection of question and subject differences is the small

range of responses of the four-point Likert scale. Continuously

measured responses would have allowed more sensitive

statistical analyses to have been performed.

4.2. Conclusions

Elderly patients are being asked be active participants in

patient safety in the ambulatory setting [14,34,35]. Early

research and anecdotal reports note that there are patients of

all ages who now incorporate patient safety self-advocacy

activities, and more research is ongoing [5,34,35,42]. We have

developed a short, 21-item self-administered survey, the Seniors

Empowerment and Advocacy in Patient Safety (SEAPS) survey

to assess elderly patients’views about their participation in safety

tasks. We used both published recommendations and direct

patient experience in its development, and applied rigorous

psychometric analyses to question selection.

4.3. Practice implications

We believe the SEAPS survey, along with objective measures

of participation and outcome can be used by researchers and

educators to assess interventions and training programs for

patients about being active in their own ambulatory medical care

safety. Effective interventions could increase patients’ safety

advocacy skills and abilities and, ultimately, lead to their

increased use by patients in the practice setting.

I rarely

do this

I often

do this

I always

do this

& & &

& & &

& & &

& & &

& & &

Appendix A (Continued )

6. How often do you ask a friend or family member

to come with you to doctors’ visits?

& & & &

How confident are you that you could actually do

the following tasks? (self-efficacy)

I don’t think

I could

I doubt

I could

I probably

could

I definitely

could

1. How confident are you that you could teach

yourself about your own health problems

and medicines?

& & & &

2. How confident are you that you could call

the doctor’s office if you haven’t received

the results of laboratory or X-ray tests?

& & & &

3. How confident are you that you could get

a second opinion from another doctor if

you think it is needed?

& & & &

4. How confident are you that you could give your

doctors a complete and thorough story of your

health problems?

& & & &

5. How confident are you that you could ask a friend

or family member to come with you to doctors’ visits?

& & & &

How much do you agree or disagree with each

statement? (attitudes)

Strongly

disagree

Disagree Agree Strongly

agree

1. I feel comfortable changing doctors if I think my

health concerns are not being met

& & & &

2. I feel it is difficult to ask a friend or family member

to come with me to doctors’ visits

& & & &

3. It would be hard to ask my doctors a lot of

questions about my health problems, lab tests

and medicines

& & & &

4. I would feel comfortable calling the doctor’s office

if I haven’t received the results of laboratory

or X-ray tests

& & & &

5. It would be easy to keep an updated list of all my

medicines, including those from the drug store

and health food store

& & & &

How strongly do you believe that doing the following

will improve your health? (outcome efficacy)

Strongly

do not believe

Do not

believe

Believe Strongly

believe

1. How strongly do you believe that giving your doctors a complete and thorough story of your

health problems and concerns will improve your health?

& & & &

2. How strongly do you believe that complaining in writing if you have problems with office

staff or doctor will improve your overall health?

& & & &

3. How strongly do you believe that getting a second opinion when needed will improve your health? & & & &4. How strongly do you believe that asking a friend or family member to come with you to

doctors’ visits will improve your overall health?

& & & &

5. How strongly do you believe that teaching yourself about your own health problems and medicines

will improve your overall health?

& & & &

N.C. Elder et al. / Patient Education and Counseling 69 (2007) 100–107106

References

[1] Roter DL, Stashefsky-Margalit R, Rudd R. Current perspectives on patient

education in the US. Patient Educ Couns 2001;44:79–86.

[2] Cegala DJ, Marinelli T, Post D. The effects of patient communication

skills training on compliance. Arch Fam Med 2000;9:57–64.

[3] Greenfield S, Kaplan S, Ware Jr JE. Expanding patient involvement in

care. Effects on patient outcomes. Ann Intern Med 1985;102:

520–8.

[4] Kaplan SH, Greenfield S, Ware Jr JE. Assessing the effects of physician–

patient interactions on the outcomes of chronic disease. Med Care

1989;27:S110–27.

[5] Elder NC, Jacobson C, Zink T, Hasse L. How experiencing medical error

has changed patients’ interactions with the primary healthcare system.

Ann Fam Med 2005;3:537–44.

[6] Waterman AD, Gallagher TH, Garbutt J, Waterman BM, Fraser V,

Burroughs TE. Brief report: hospitalized patients’ attitudes about

and participation in error prevention. J Gen Intern Med 2006;21:

367–70.

[7] Hibbard JH, Peters E, Slovic P, Tusler M. Can patients be part of the

solution? Views on their role in preventing medical errors. Med Care Res

Rev 2005;62:601–16.

[8] Patient fact sheet: 20 tips to help prevent medical errors. In: Agency for

healthcare research and quality; 2000. www.ahrq.gov/consumer/20tips.htm

[accessed July 18, 2007].

[9] Speak up: help prevent errors in your care. In: JCAHO; 2002. http://

www.jointcommission.org/GeneralPublic/Speak+Up/gp_speakup_bro.htm

[accessed July 18, 2007].

[10] Ask Me 3. In: Partnership for clear health communication; 2003.

www.Askme3.org [accessed July 18, 2007].

N.C. Elder et al. / Patient Education and Counseling 69 (2007) 100–107 107

[11] National Patient Safety Foundation. Public opinion of patient safety

issues. Chicago, IL: National Patient Safety Foundation; 1997.

[12] Green L, Fryer G, Yawn B, Lanier D, Dovey S. The ecology of medical

care revisited. N Engl J Med 2001;344:2021–5.

[13] McGreevey M, editor. Patients as partners. How to involve patients and

families in their own care. Chicago: Joint Commission on Accreditation of

Healthcare Organizations; 2006.

[14] Entwistle VA, Mello MM, Brennan TA. Advising patients about patient

safety: current initiatives risk shifting responsibility. Jt Comm J Qual

Patient Saf 2005;31:483–94.

[15] Salazar MK. Comparison of four behavioral theories. A literature review.

Aaohn J 1991;39:128–35.

[16] Cummings KM, Jette AM, Rosenstock IM. Construct validation of the

health belief model. Health Educ Monogr 1978;6:394–405.

[17] Rosenstock IM, Strecher VJ, Becker MH. Social learning theory and the

Health Belief Model. Health Educ Q 1988;15:175–83.

[18] Champion VL. Instrument development for health belief model con-

structs. Adv Nurs Sci 1984;6:73–85.

[19] Aalto AM, Uutela A. Glycemic control, self-care behaviors, and psycho-

social factors among insulin treated diabetics: a test of an extended health

belief model. Int J Behav Med 1997;4:191–214.

[20] Mahoney CA, Thombs DL, Ford OJ. Health belief and self-efficacy

models: their utility in explaining college student condom use. AIDS

Educ Prev 1995;7:32–49.

[21] Belcher VN, Fried TR, Agostini JV, Tinetti ME. Views of older adults on

patient participation in medication-related decision making. J Gen Intern

Med 2006;21:298–303.

[22] Guse CE, Yang H, Layde PM. Identifying risk factors for medical injury.

Int J Qual Health Care 2006;18:203–10.

[23] Benbassat J, Pilpel D, Tidhar M. Patients’ preferences for participation in

clinical decision making: a review of published surveys. Behav Med

1998;24:81–8.

[24] Levinson W, Kao A, Kuby A, Thisted RA. Not all patients want to

participate in decision making. A national study of public preferences. J

Gen Intern Med 2005;20:531–5.

[25] Swift EK, Koepke CP, Ferrer JA, Miranda D. Preventing medical errors:

communicating a role for Medicare beneficiaries. Health Care Financ Rev

2001;23:77–85.

[26] Elder N, Dovey S. A classification of medical errors and preventable

adverse events in primary care: a synthesis of the literature. J Fam Pract

2002;51:927–32.

[27] You can help improve patient safety. In: National Patient Safety Foundation;

2002. http://npsf.org/pdf/paf/WhatYouCanDo.pdf [accessed July 18, 2007].

[28] Linzer M, Konrad TR, Douglas J, McMurray JE, Pathman DE, Williams

ES, Schwartz MD, Gerrity M, Scheckler W, Bigby JA, Rhodes E.

Managed care, time pressure, and physician job satisfaction: results from

the physician worklife study. J Gen Intern Med 2000;15:441–50.

[29] Epstein RM. Time, autonomy, and satisfaction. J Gen Intern Med

2000;15:517–8.

[30] Dovey SM, Meyers DS, Phillips Jr RL, Green LA, Fryer GE, Galliher J,

Kappus J. A preliminary taxonomy of medical errors in family practice.

Qual Saf Health Care 2002;11:233–8.

[31] Kuzel AJ, Woolf SH, Gilchrist VJ, Engel JD, LaVeist TA, Vincent C,

Frankel RM. Patient reports of preventable problems and harms in primary

health care. Ann Fam Med 2004;2:333–40.

[32] West D, Westfall JM, Araya-Guerra R, Hansen L, Quintela J, Van Vorst R,

Staton EW, Matthews B, Pace WD. Using reported primary care errors to

develop and implement patient safety interventions: a report from the

ASIPS Collaborative. In: Advances in patient safety: from research to

implementation. Rockville, MD: Agency for Healthcare Research and

Quality Publication number 050021; 2005.

[33] Arora NK, Ayanian JZ, Guadagnoli E. Examining the relationship of

patients’ attitudes and beliefs with their self-reported level of participation

in medical decision-making. Med Care 2005;43:865–72.

[34] Gibson R, Singh JP. Wall of silence: the untold story of the medical

mistakes that kill and injure millions of Americans. Washington, DC:

Lifeline Press; 2003.

[35] Wachter RM, Shojania KG. Internal bleeding. The truth behind America’s

terrifying epidemic of medical mistakes. New York: Rugged Land; 2004.

[36] DiGiovanni CW, Kang L, Manuel J. Patient compliance in avoiding

wrong-site surgery. J Bone Joint Surg Am 2003;85-A:815–9.

[37] Gattellari M, Ward JE. Measuring men’s preferences for involvement in

medical care: getting the question right. J Eval Clin Pract 2005;11:237–46.

[38] Street Jr RL, Gordon HS, Ward MM, Krupat E, Kravitz RL. Patient

participation in medical consultations: why some patients are more

involved than others. Med Care 2005;43:960–9.

[39] Say R, Murtagh M, Thomson R. Patients’ preference for involvement in

medical decision making: a narrative review. Patient Educ Couns

2006;60:102–14.

[40] Aspden P, Corrigan J, Wolcott J, Erickson S. Patient safety: achieving a

new standard of care. Washington, DC: National Academies Press; 2004.

[41] Cegala DJ, Post DM. On addressing racial and ethnic health disparities.

The potential role of patient communication skills interventions. Am

Behav Scientist 2006;46:853–67.

[42] Goeltz R, Hatlie M. Trial and error in my quest to be a partner in my health

care: a patient’s story. Crit Care Nurs Clin N Am 2002;14:391–9.