Original Article

© 2003 Blackwell Publishing Ltd

357

Blackwell Science, LtdOxford, UKCCHDChild: Care, Health and Development1365-2214Blackwell Publishing Ltd, 2003

2003

29

5357366

Original Article

Factors

affecting family-centred serviceM. Law

et al.

Correspondence:Mary Law, CanChild Centre for Childhood Disability Research, McMaster University, Institute for Applied Health Sciences, 1400 Main Street West, Room 403, Hamilton, Ontario L8S 1C7, Canada E-mail: lawm@mcmaster.ca

Original Article

Factors affecting family-centred service delivery for children with disabilities

M. Law,* S. Hanna,* G. King,*† P. Hurley,* S. King,* M. Kertoy*‡ and P. Rosenbaum*

*CanChild Centre for Childhood Disability Research, McMaster University, Institute for Applied Health Sciences, Hamilton, †Thames Valley Children’s Centre, London, ‡Department of Communicative Disorders, University of Western Ontario, Elborn College, London, Ontario, Canada

Accepted for publication 7 April 2003

Abstract

Background

The provision of family-centred services (FCS) emphasizes a partnership between

parents and service providers so that families are involved in every aspect of services for their child.

There is evidence that providing FCS is associated with improvements in parents’ satisfaction with

services, decreased parental stress, and positive child outcomes.

Methods

The purpose of this study was to examine factors that are most important in determining

parent perceptions of the family centredness of care and parent satisfaction with service. A cross-

sectional survey was completed by 494 parents, 324 service providers, and 15 CEOs from 16

organizations delivering children’s rehabilitation services. Analyses were completed using a structural

equation modelling approach.

Results

Survey return rates ranged from 77 to 94%. Findings indicate that the principal determinants

of parent satisfaction with services are the family-centred culture at the organization and parent

perceptions of FCS. Parent satisfaction with services was also influenced by the number of places

where services were received and the number of health and development problems experienced by

their child.

Conclusion

Parent satisfaction with services is strongly influenced by the perception that services

are more family centred, fewer places where services were received and fewer health and

development problems for their child. Ways in which organizations can improve satisfaction through

carrying out family-centred behaviours are discussed.

Keywords

childhood disability, family-centred service, survey, structural equation modelling

Introduction

Services to children with disabilities and their fam-

ilies have changed to become more family centred

over the past 20 years. Family-centred service (FCS)

is a philosophy and method of service delivery for

children and parents which emphasizes a partner-

ship between parents and service providers, focuses

on the family’s role in decision-making about their

child, and recognizes parents as the experts on their

child’s status and needs (Rosenbaum

et al.

1998).

The guiding principles of FCS are:

• Each family should have the opportunity to

decide the level of involvement they wish in

decision-making for their child.

358

M. Law

et al

.

© 2003 Blackwell Publishing Ltd,

Child: Care, Health & Development

,

29

, 5, 357–366

• Parents should have ultimate responsibility for

the care of their children.

• Each family member should be treated with

respect (as individuals).

• The needs of all family members should be

considered.

• The involvement of all family members should

be encouraged.

(Rosenbaum

et al

. 1998).

To deliver services in a manner consistent with

the guiding principles of FCS, service providers

exhibit particular behaviours that respect and sup-

port families and enhance their partnership with

families. These behaviours, as well as other factors

such as number of services required, are thought

to influence parents’ perceptions of care and satis-

faction with service. The purpose of this study was

to examine factors that are most important in

determining parent perceptions of the family cen-

tredness of care and parent satisfaction with ser-

vice. Within the context of children’s rehabilitation

services, knowledge about factors influencing ser-

vice delivery can be used to improve the satisfac-

tion of a family’s experiences with service delivery.

Background information

FCS is an approach to service delivery that empha-

sizes working in partnership with families.

Research indicates that providing FCS is associated

with improvements in parents’ satisfaction with

services, decreased parental stress, and positive

child outcomes (Rosenbaum

et al

. 1998). The two

characteristics of service delivery that appear to

influence a family’s perceptions of FCS and their

satisfaction are the process and structure of ser-

vices (Dunst & Trivette 1996; Hasnat & Graves

2000; King

et al.

2001). The process of service

delivery centres on the way in which services are

provided, including components such as interper-

sonal relationships, providing respectful care, and

service continuity. Aspects of service delivery struc-

ture that influence perceptions of care and satisfac-

tion consist of environmental characteristics of the

service organization such as physical comfort,

waiting lists, and ease of access to services

(King

et al.

2001).

Research indicates that the service delivery pro-

cess, in particular, the interpersonal aspects of ser-

vice delivery, is important to families. For example,

Garwick

et al

. (1998), in a survey of 63 families,

indicated that families placed importance on indi-

vidualized care, provision of information, and

inclusion of family in care planning and delivery.

A study of parents of children with diabetes melli-

tus (

n

=

80) and cystic fibrosis (

n

=

45) indicated

that parents ranked continuity and consistency of

service providers over time, parent involvement,

provision of information, and availability of care as

the most important aspects of caregiving (Baine

et al

. 1995). Dunst & Trivette (1996) posited that

effective help giving has three components: techni-

cal quality, help giver traits, and family or service

provider involvement. They state that service pro-

viders often place too much emphasis on technical

quality as opposed to enhancing participation of

families.

In examining interpersonal aspects of care that

increase satisfaction, King

et al

. (1996a) showed

that three factors, information exchange, respectful

and supportive care, and partnership/enabling, are

all associated with improved levels of satisfaction,

with information exchange having the strongest

association. Families’ beliefs that service provider

relationships were more family centred are associ-

ated with greater satisfaction and greater intention

to seek help from health care providers (Van Riper

1999). In this study of 89 families of children with

Down syndrome, Van Riper (1999) demonstrated

that satisfaction with care is associated with less

discrepancy between mothers’ ideal family service

provider relationships and the actual nature of

these relationships. Providing information about

the child’s disability and service availability, and

interpersonal communication between parents and

service providers have been found to have a signif-

icant effect on parents’ satisfaction (Knafl

et al

.

1992; Stallard & Lenton 1993; Hasnat & Graves

2000). King

et al

. (2001) found that respectful and

supportive care was associated with higher parent

satisfaction with children’s rehabilitation services.

Research indicates that the provision of FCS is

associated with higher levels of parent satisfaction

with service (King

et al

. 1999, 1999; Van Riper

1999).

Factors affecting family-centred service

359

© 2003 Blackwell Publishing Ltd,

Child: Care, Health & Development

,

29

, 5, 357–366

There is very little information about the effect

of the structural characteristics or culture of an

organization on parent perceptions of service and

satisfaction. King

et al

. (2001) found that parents

who were less satisfied with children’s rehabilita-

tion services focused on identified aspects of both

the process and the structure of service delivery.

Related to structure, access to care was most com-

monly cited as a problem for those parents who

were less satisfied with service.

In a study of emergency room and emotional

support for families, the importance of the organi-

zational culture and climate was examined

(Hemmelgarn

et al

. 2001). Through open-ended

interviews and unstructured surveys of 131 paedi-

atric emergency service providers, this study mea-

sured the level of emphasis on the provision of

emotional support provided to families of children

receiving emergency care. Findings indicated that

the emphasis on provision of emotional support of

the service providers was directly related to the

organizational culture of the emergency room.

Purpose

The findings reported in this article come from a

large, multicentred survey of FCS completed in the

province of Ontario, Canada, by CanChild Centre

for Childhood Disability Research at McMaster

University, Hamilton, Canada. One purpose of the

survey was to determine factors associated with

parent perceptions of the family centredness of ser-

vice delivery and their satisfaction with service.

Methods

This cross-sectional survey involved parents of

children with disabilities, service providers, and

executive directors or managers of children’s reha-

bilitation services. Twenty-two centres/organiza-

tions who were major providers of rehabilitation

services for children with disabilities and their fam-

ilies in Ontario were invited to participate in this

survey, and 16 consented. These organizations

were selected to represent both urban and rural

centres and all regions across the province of

Ontario to ensure representativeness of the sample.

All organizations served a broad range of disabili-

ties, including children with cerebral palsy,

spina

bifida

, various syndromes, developmental delay,

communication disorder, acquired brain injury,

autism, pervasive developmental disorder, and

muscular disease.

The sample size was large enough to provide

meaningful data across the province and individu-

alized feedback to each participating organization.

Our goal was to have 30–40 parents and a mini-

mum of 20 service providers from each centre/

agency. For those centres with a staff of 60 or less,

20 randomly selected service providers were sent a

questionnaire package. Larger centres (i.e.

>

60

staff) randomly selected 33% of their staff to

receive the package. CEO/managers of each of 16

centres/organizations also received a questionnaire

package.

Parents were randomly selected from those cur-

rently receiving services from each participating

organization. The only exclusion criterion was the

inability of parents to respond to English language

questionnaires. Parents were first contacted

through a mailing from their organization and

received the FCS questionnaire package if they gave

consent. The package included materials request-

ing information about their child with a disability,

the nature of services received, the family’s beliefs

about participating in FCS, their perceptions of the

family centredness of service delivery, and their

judgements about satisfaction with services

(Table 1).

All service providers involved in the provision of

rehabilitation services to children with disabilities

were eligible to participate. Since the centre/agency

had agreed to participate on behalf of its staff, a

package prepared by the research team at

CanChild

was sent to randomly selected service providers via

each centre’s mail system. This package included a

covering letter explaining the study and the ques-

tionnaires, and was returned directly to the

research team. Service providers completed a pack-

age that included the same questionnaire on beliefs

about FCS that parents completed, a measure of

their perceptions of their own family-centred

behaviours, and a demographic form.

CEOs/managers completed a questionnaire

requesting information about key structural fea-

tures of their organizations, clients and service

360

M. Law

et al

.

© 2003 Blackwell Publishing Ltd,

Child: Care, Health & Development

,

29

, 5, 357–366

Table 1.

Indicators of family-centred service (FCS) in the model

Service conceptStandardizedloadings

% Variance Measurements

Parent beliefs about FCS

0.7180.4800.825

51.523.068.0

Parent’s response to three subscales of

Measure of

Beliefs about Participation in FCS

(King

et al

. 2002)• beliefs about benefits of FCS• beliefs about the absence of negative outcomes from FCS• beliefs about self-efficacy to implement FCS

Parent satisfaction with service Parent responses to the eight items of

Client

Satisfaction Questionnaire

(Larsen

et al

. 1979)0.906 82.1 • quality of service?0.8140.832

66.369.2

• kind of service you wanted?• organization met your needs?

0.876 76.8 • would you recommend this organization?0.862 74.4 • satisfied with amount of help?0.7830.918

61.384.3

• services help you deal with child’s problems?• how satisfied overall?

0.860 74.7 • would you go back?Parent perceptions of receiving

FCSParent’s response to five subscales of the

Measure of

Processes of Care

(King

et al

. 1996a):0.9020.712

81.350.6

• enabling and partnership• providing general information

0.7100.927

50.486.0

• providing specific information about the child• co-ordinated and comprehensive care for the child and

family0.935 87.4 • respectful and supportive care

FCS culture at the organization N/A N/A From Chief Executive Officer:

Number of five FCS-relevant systems and services

present in the organization

• person who co-ordinates services• resource centre for families• formal adoption of FCS approach• provide information about FCS to families• FCS educational activities

Count of recent changes designed to make organization

more family centred

• changes in the way information is provided to families• changes in the way psychosocial support is provided• changes in physical setting of centre

Count of changes to procedures in last 5 years that made

organization more or less family centred

• intake procedures• waiting lists• goal setting• meetings• information to parents about services• communication with parents• user-friendly reports to parents

CEO’s beliefs about FCS

• barriers to FCS• positive outcomes of FCS• negative outcomes of FCSFrom service providers:

Centre-average responses to four subscales of Measure of

Processes of Care for Service Providers

(Woodside

et al.

2001):• showing interpersonal sensitivity• providing general information• communicating specific information about the child• treating people respectfully

Loadings and percentage variance are not applicable (N/A) for these variables because latent variables were not estimated.

Factors affecting family-centred service

361

© 2003 Blackwell Publishing Ltd,

Child: Care, Health & Development

,

29

, 5, 357–366

providers, which was used to operationalize fam-

ily-centred culture in each organization. This form

was developed by the research team and asked

questions such as the number of clients served, the

nature of services provided, amount of informa-

tion and/or services provided, global budgets,

whether FCS had been formally adopted by the

organization and changes in any of these features

in the past five years. CEOs/managers also com-

pleted the questionnaire indicating their beliefs

about FCS (see Table 1).

Analysis

Structural equation modelling (Klein 1998) was

used to model the determinants of two indepen-

dent variables, parental perceptions of care and sat-

isfaction with service. The structural equation

approach involves the sequential development of

measurement and structural models to describe

the data. The measurement model specifies how

the relevant health service concepts are measured.

In this case, the health service concepts of interest

included the outcomes of client satisfaction and

perceptions of service, as well as explanatory con-

cepts such as the degree of family centredness of

the organization and beliefs about FCS. For each

of these concepts, relevant measurements were

available from among the questionnaire responses

provided by CEOs, service providers, and parents.

When more than one measure of a concept is avail-

able, they can be used together to estimate the con-

cept without measurement error (Kline 1998). This

is conceptually similar to the older statistical prac-

tice of factor analysis, and some applications of the

measurement model are referred to as confirma-

tory factor analysis in the measurement literature.

When only one measurement of a concept is avail-

able, it can be included in the analysis, as it is, just

as in conventional regression analysis.

The structural model involves specifying the

hypothesized causal relationships among the

health concepts developed in the measurement

model. Specified causal pathways are estimated

as regression coefficients that provide evidence

regarding the direction, size, and statistical signifi-

cance of causal relationships. Summary measures

of overall model fit are also available, and can be

used to compare models.

The structural equation modelling approach is

particularly well suited to the development and

testing of multivariable health service models.

The use of latent health concepts developed in

the measurement model improves the precision

of the analysis by reducing measurement error. It

can also simplify the model by reducing a large

number of measured variables to fewer essential

concepts. The relationships in the structural

model are tested simultaneously in a way that

expresses theories about how they function to

determine outcomes. Estimates of the effects of

each factor take into account the effects observed

for the other factors, given the way they are

arranged together in the model. This helps to sort

out the relative influence of all the factors we

have measured, and provides a coherent test of an

entire theory, rather than disconnected tests of

individual relationships.

For the structural equation modelling, missing

data were handled using a standard and well-

established method involving maximum likelihood

estimation of the required means, variances, and

covariances, using the EM-algorithm (Little &

Rubin 1987). These estimates allow all subjects to

be retained in the analysis, and are more efficient

and less prone to bias than exclusion of subjects

Number of service sources N/A N/A Parent-reported total number of sources/places child goes to receive services

Complexity of child’s health/development problem

N/A N/A Parent-reported total number of health and development problems that each child has (e.g. communication, mobility, behaviour)

Service conceptStandardizedloadings

% Variance Measurements

Loadings and percentage variance are not applicable (N/A) for these variables because latent variables were not estimated.

Table 1.

Continued

362

M. Law

et al

.

© 2003 Blackwell Publishing Ltd,

Child: Care, Health & Development

,

29

, 5, 357–366

with partially missing data, mean substitution, or

other common

ad hoc

methods.

Results

Participants

From the 641 consenting parents, 494 question-

naires were returned and analysed (return rate

77%). From the 411 service providers who were

sent survey packages, 324 questionnaires were

returned and analysed (return rate 79%).

The families responding to this survey lived pri-

marily in urban communities (80.0%), were from

two-parent families (85.4%), and spoke primarily

English in the home (94.7%). For 86.4% of the

surveys, the respondent was the child’s natural

mother, and the majority of the respondents had

completed education beyond high school (72.3%).

Sixty-one percent of the respondents were

employed at the time of the survey. Family income

ranged from

<

$15 000 (6.1%) to

>

$75 000 (23.1%),

with most families (53.4%) having an income

between $30 000 and $75 000. Twenty percent of

the families belonged to a parent support group.

The majority of the children in the survey

(50.0%) were between the ages of 3 and 8 years,

although children from infancy to late adolescence

were included. Sixty-three percent were male. The

primary health or development problem of the

children was acquired brain injury (3.0%), autism

or pervasive developmental disorder (5.1%), cere-

bral palsy (25.3%), communication disorder

(12.6%), developmental co-ordination disorder

(2.2%), developmental delay (13.6%), muscle dis-

ease (2.2%),

spina bifida

(6.5%), syndrome

(12.3%), or other (17.2%). The category of other

included less prevalent diseases or disorders such

as cancer, learning and motor difficulties. The

majority of children (77.7%) had two or more

health or development problems in addition to

their primary diagnosis, such as behaviour, com-

munication, learning disorder, epilepsy, or hearing

or visual impairment. Thirty-five percent had three

to four problems and 17.5% had five or more

health or development problems.

Service provider respondents came from 19 dis-

ciplines, with the most frequent from occupational

therapy (24.4%), speech language pathology

(19.1%), and physiotherapy (18.5%). Eighty-two

percent of the service providers described their pri-

mary role as direct service provision. The service

provider group had considerable clinical experi-

ence with 61.9% having practised for more than

5 years. The majority of service providers (61%)

had received formal training in FCS.

The organizations surveyed were classified as

small, medium, or large based on population of the

catchment area and annual budget. Four organiza-

tions in the sample were large, four were medium,

and seven were small. Of the 16 participating orga-

nizations, 15 completed questionnaires from

CEOs/managers were received (response rate

94%). Sixty-seven percent of the organizations had

formally adopted FCS as their service delivery

method of choice.

Factors influencing perceptions of service delivery

The hypothesized model of factors affecting service

delivery involved six health service concepts,

including the outcomes of parent perceptions of

service and parents’ satisfaction. These are listed in

Table 1, along with the measured indicators used

to estimate them. Parents’ satisfaction, parents’

perceptions of FCS, and parents’ beliefs about FCS

were each assessed by multiple measurements.

They have been estimated as error-free latent con-

structs. The relationships between the measure-

ments and their underlying health service concept

are reported as standardized loadings, indicating

the correlation between the underlying concept

and the measured indicator. The square of these

loadings is also reported, interpretable as the per-

centage of variance in the measurement accounted

for by the factor. This is an indication of how

strongly each measurement is related to its under-

lying concept. Complexity of the child’s health/

development problem and number of sources of

service were each estimated by single questionnaire

responses. No latent health concept can be esti-

mated for these measurements, and so they are

used to represent the concepts in the analysis, as is.

The FCS culture at the organization construct

focuses on the organization’s shared values about

Factors affecting family-centred service

363

© 2003 Blackwell Publishing Ltd,

Child: Care, Health & Development

,

29

, 5, 357–366

FCS and the strategies used within the organization

to implement these values. FCS culture at each

organization was assessed by multiple measure-

ments, with these measurements taken from the

questionnaire responses of CEOs and an aggregate

of service providers’ self-perceptions of the delivery

of FCS. With only one set of measurements per

centre, it was not possible to use these to estimate

latent concepts. Instead, a composite index of FCS

culture was constructed by combining the

observed measurements (Table 1).

The final structural model, specifying the

hypothesized causal relationships among the

health service concepts, is illustrated in Fig. 1.

The direction of the arrows indicates the hypothe-

sized causal direction of the relationship. To facili-

tate interpretation by putting all effects on the

same scale, the effects are reported adjacent to their

arrows as standardized regression coefficients.

These coefficients indicate the predicted increase

(

+

) or decrease (–) in the outcome, measured in

standard deviations, for every one standard devia-

tion change in the predictor. This provides some

basis for comparing the size of effects. All pathways

remaining in the model are statistically significant

(

P

<

0.05). Non-significant pathways tested and

deleted from the model included a path from

FCS

Culture

to

Parent’s Perception of Receiving FCS

(

P

<

0.11), and a direct path from

Parent’s Beliefs

about FCS

to

Satisfaction with Service

(

P

<

0.52).

The addition of the latter path did not alter the

significance of the path from

Parent’s Perception of

Receiving FCS

to

Satisfaction with Service

, indicat-

ing that parents’ beliefs about FCS cannot account

for this relationship.

Because these effects are estimated simulta-

neously, they are adjusted for each other. As evi-

dent from Fig. 1, a health concept such as parent

perceptions of service can function both as a pre-

dictor and as an outcome in the model. The num-

bers of sources of service and parents’ beliefs about

FCS account for a total of 6% of the variance in

parents’ perceptions of receiving FCS. For the out-

come variable of parents’ satisfaction with service,

59% of the variance in this outcome is accounted

for by predictors in the model.

The factor ‘parent perceptions of FCS’ depends

directly on two factors – parent beliefs about FCS

(pathway F) and the number of sources of service

their child and family are receiving (pathway E) –

and indirectly on one factor – complexity of child’s

health/development problem (number of diag-

noses) through its relationship with the number of

sources of service (pathway D). These findings

indicate that children with an increased number of

health and development problems (number of

diagnoses) are likely to receive services at more

locations. Receiving an increased number of ser-

vices is likely to lead to parents perceiving services

as less family centred (that is why the number on

the arrow is negative). The results also indicate that

parent beliefs about FCS influence their percep-

tions of FCS (

+

0.21).

The other outcome of interest was ‘parent satis-

faction with service’. In this model, parent satisfac-

tion depends directly on FCS culture (pathway A),

the number of sources of service (pathway B), and

parent perception of FCS (pathway C). The two

factors with the strongest relationship to parent

satisfaction were FCS culture at the centre (

+

0.42)

Figure 1.

Factors influencing perceptions of and satisfaction with service.

–0.13 –0.07 +0.42

+0.63+0.21

+0.44

Complexityof child’s health/

developmentproblem

Number of sourcesof service

Family-centered service culture at

center/organization

Parent beliefsabout family-centered

services

Parent perceptions of family-centered

service

Parent satisfactionwith service

B A

C

D

F

E

364

M. Law

et al

.

© 2003 Blackwell Publishing Ltd,

Child: Care, Health & Development

,

29

, 5, 357–366

and parent perception of FCS (

+

0.63). These results

indicate that a more FCS culture and higher parent

perception of FCS are likely to lead to greater parent

satisfaction. The relative influence of the number

of sources of service is weak (

-

0.07) once the influ-

ence of receiving FCS has been taken into account.

Some indexes that evaluate the overall fit of the

model to these data are reported in Table 2, along

with conventional guidelines often used to assess

models using these indexes. These indexes indicate

that the model fits moderately well, with two of

four values exceeding the standard criteria, and the

remaining two indexes having moderate values. Fit

indexes offer rough guidelines to the empirical ade-

quacy of the model that must be balanced against

the complexity and theoretical fit of the model.

Discussion

The primary findings from this survey indicate that

several factors contributed significantly to explain

the relationships between factors measured in this

survey and the outcomes of parent perceptions of

FCS and parents’ satisfaction with service. The

principal determinants of parent satisfaction with

service are the family-centred culture at the orga-

nization and parent perceptions of FCS. Parent sat-

isfaction with services is strongly influenced by the

perception that services are more family centred,

with this perception of family centredness being

more positive when there are fewer places where

services were received and fewer health and devel-

opment problems for their child. The findings con-

firm that parents are more satisfied when services

are provided in a family-centred manner. Satisfac-

tion with service is important because it has been

shown to increase adherence to treatment recom-

mendations and to lead to fewer parental feelings

of distress and depression and improved well-being

(King

et al

. 1996b,1999; Rosenbaum

et al

. 1998;

Van Riper 1999).

Parents’ perceptions that they are receiving FCS

are influenced by their beliefs about FCS. Because

parents’ beliefs about FCS influences their percep-

tions of FCS within a centre or organization, it is

useful for organizations to provide information to

families about FCS, what it means, and what they

should expect.

Children in this survey who have more health/

development problems receive services from a

greater number of organizations at different loca-

tions. When children receive services from more

sources, parents report that services are less family

centred. These findings also indicate that children

who have more complex problems are likely to

receive services at a greater number of sources, and

this influences perceptions of services and therefore

satisfaction. These findings confirm what many

families have indicated, that services at fewer loca-

tions will increase their satisfaction with services.

After the influence of perceptions of FCS and fam-

ily-centred culture on satisfaction with service are

accounted for, the remaining influence of the num-

ber of sources of service is weak. This suggests that

the provision of FCS across multiple locations

decreased the negative effects of a greater number

of sources of service. Breslau & Mortimer (1981)

also found that continuity of care positively influ-

enced the relationship between source of service

and parent perception and satisfaction with service.

One of the most interesting findings of this anal-

ysis is that the family-centred culture of a centre or

organization strongly influenced satisfaction with

service. There has been little research to date exam-

ining the relationship between organizational cul-

ture and service delivery outcomes such as

satisfaction (Aiken

et al

. 1997). For these analyses,

family-centred culture was determined by infor-

Table 2.

Indexes of fit for the model

CFI TFI Discrepancy/d.f. RMSEA (95% CI)

Model fit statistic 0.987 0.982 3.52 0.071(0.065; 0.078)

Conventional criteria for well-fitting model

>

0.90

>

0.90

<

2.0

=

close fit

<

5.0

=

moderate fit

<

0.05

=

close fit

<

0.10

=

moderate fit

CFI, comparative fit index; TLI, Tucker-Lewis Index; Discrepancy/d.f., discrepancy per degree of freedom; RMSEA, root mean squared error of approximation.

Factors affecting family-centred service

365

© 2003 Blackwell Publishing Ltd,

Child: Care, Health & Development

,

29

, 5, 357–366

mation collected about the physical, procedural,

and strategic characteristics of an organization

used to operationalize FCS as well as service pro-

viders’ and CEOs beliefs about FCS. Parents,

service providers, and organizations should be

encouraged to work together to ensure that a fam-

ily-centred culture is present within an organiza-

tion. The support and active involvement of the

CEO in the development of a family-centred cul-

ture is very important to an organization. Organi-

zations that provided training in FCS to staff

members (and had a more family-centred culture

overall) received higher satisfaction ratings from

parents.

Organizations can improve satisfaction through

carrying out the following family-centred behav-

iours: (1) formally adopting a family-centred

approach to service delivery; (2) having a specific

person/team lead the development of FCS; (3) pro-

viding information about FCS to families starting

with the initial visit; (4) training staff in FCS deliv-

ery; (5) changing procedures to be more family

centred (e.g. decreasing wait lists, instituting family

friendly intake procedures, involving parents in

goal setting, user-friendly reports to parents); and

(6) providing a welcoming environment (e.g.

attractive space, resource centre). Findings from

other FCS research supports the implementation of

these strategies as a method to improve the family-

centred nature of an organization (Baine

et al

.

1995; Garwick

et al

. 1998; Hemmelgarn

et al

.

2001).

Importantly, the effects of FCS culture and par-

ent perceptions of FCS on parent satisfaction are

not directly and significantly linked to one another,

indicating that FCS culture has an effect on satis-

faction that does not involve indirect path through

parent perceptions of FCS. Another important fac-

tor, in addition to beliefs, sources of service, and

number of health/development problems likely to

contribute to satisfaction, is the behaviour of the

service provider from whom they actually receive

care. In our model, we used only average service

provider behaviour and this is included in the FCS

culture factor. The addition of information about

the behaviour of a family’s own service providers

would be important to include in future surveys of

this nature.

The results of this survey have implications for

parents, service providers, administrative decision-

makers, and legislative policy makers. For parents,

the findings confirm that parents’ beliefs about

FCS are an important factor in determining their

perceptions of service. It is important for parents

to gain information about FCS and what it means

for them, and to feel comfortable with its imple-

mentation. Their active involvement in the design

and implementation of services will likely increase

their feelings of ease with FCS (Hill

et al

. 1999).

For parents whose children have complex devel-

opmental disabilities, it is particularly important

for their services to be co-ordinated and available

from fewer locations. The findings suggest to orga-

nizations within a community to work together to

decrease the number of sources of service for fam-

ilies. This is particularly important for families

who have children with several health or develop-

ment problems and who are more likely to receive

services from more sources and a greater number

of service providers. If several locations of service

are required, co-ordination of these services to

ensure that they provide consistent FCS will likely

have a significant influence on parent perceptions

of FCS and, thereby, their satisfaction.

The findings strongly suggest that FCS should be

considered a ‘best practice approach’ to meeting

the needs of children with disabilities and their

families (King

et al

. 2002). Organizations can eval-

uate the family centredness of their services using

assessment tools such as the Measure of Processes

of Care (parent and service provider versions)

(King

et al

. 1996a; Woodside

et al

. 2001) and the

Measure of Beliefs about Participation in FCS

(King

et al

. 2003).

For legislative policy makers, the findings indi-

cate the importance of FCS in leading to parents’

satisfaction with services. Family-centred princi-

ples can serve as the underpinning for policies con-

cerning services for children with disabilities and

their families. The findings indicate that decreasing

the number of sources of service within a commu-

nity is an important policy goal because it leads to

improved positive perceptions about and satisfac-

tion with service. While co-ordination is the goal,

providing FCS can buffer the effect of parents

receiving services from several different organiza-

366

M. Law

et al

.

© 2003 Blackwell Publishing Ltd,

Child: Care, Health & Development

,

29

, 5, 357–366

tions. The findings indicate that organizations with

family-centred cultures have an important influ-

ence on outcomes: parents experienced the services

as more family centred and were more satisfied

with these services. Because family centredness is

also linked to better outcomes for children and

better parent well-being, policies endorsing a fam-

ily-centred approach will have important payoffs

for children and families.

Acknowledgements

Our sincere thanks to the families, service provid-

ers and their organizations who participated in this

study and who provided CanChild Centre with

feedback about the survey and findings. This

research was supported by the Ministry of Health

and Long-term Care, Ontario, Canada.

References

Aiken, L. H., Sochalski, J. & Lake, E. T. (1997) Studying

outcomes of organizational change in health service.

Medical Care

,

35

, NS6–NS18.

Baine, S., Rosenbaum, P. & King, S. (1995) Chronic

childhood illnesses: what aspects of caregiving do par-

ents value?

Child: Care, Health and Development

, 21,

291–304.

Breslau, N. & Mortimer, E. A. (1981) Seeing the same

doctor: determines the satisfaction with specialty care

for disabled children. Medical Care, 19, 741–758.

Dunst, C. & Trivette, C. (1996) Empowerment, effective

helpgiving practices and family-centered care. Pediat-

ric Nursing, 22, 334–343.

Garwick, A., Kohrman, C., Wolman, C. & Blum, R.

(1998) Families’ recommendations for improving ser-

vices for children with chronic conditions. Archives of

Pediatric and Adolescent Medicine, 15, 440–448.

Hasnat, J. & Graves, P. (2000) Disclosure of developmen-

tal disability: a study of parent satisfaction and the

determinants of satisfaction. Journal of Pediatric and

Child Health, 36, 32–35.

Hemmelgarn, A. L., Glisson, C. & Dukes, D. (2001)

Emergency room culture and the emotional support

component of family-centered care. Children’s Health

Care, 30, 93–110.

Hill, I., Schwalberg, R., Zimmerman, B. & Tilson, B.

(1999) Achieving service integration for children with

special health care needs: an assessment of alternative

medicaid managed care models. Volume 1: Synthesis of

study results. United States Department of Health and

Human Services, Washington, DC.

King, G., King, S. & Rosenbaum, P. (1996a) Interper-

sonal aspects of care-giving and client outcomes: a

review of the literature. Ambulatory Child Health, 2,

151–160.

King, S., Rosenbaum, P. & King, G. (1996b) Parents’

perceptions of caregiving: development and validation

of a measure of processes. Developmental Medicine and

Child Neurology, 38, 757–772.

King, G., King, S., Rosenbaum, P. & Goffin, R. (1999)

Family-centred caregiving and well-being of parents of

children with disabilities: Linking process with out-

come. Journal of Pediatric Psychology, 24, 41–53.

King, G., Cathers, T., King, S. & Rosenbaum, P. (2001)

Major elements of parents’ satisfaction and dissatisfac-

tion with pediatric rehabilitation services. Children’s

Health Care, 30, 111–134.

King, G., King, S., Law, M., Kertoy, M., Rosenbaum, P.

& Hurley, P. (2002) Family-Centred Service in Ontario:

A ‘Best Practice’ Approach for Children with Disabilities

and Their Families. CanChild Centre for Childhood

Disability Research, McMaster University, Ontario,

Canada.

King, G., Kertoy, M., King, S., Law, M., Rosenbaum, P.

& Hurley, P. (2003) A measure of parents’ and service

providers’ beliefs about participation in family-

centered services. Children’s Health Care, 32, 91–124.

Klein, R. (1998) Principles and Practice of Structural

Equation Modeling. Guilford, New York.

Knafl, K., Breitmayer, B., Gallo, A. & Zoeller, L. (1992)

Parents’ views of health care providers: an exploration

of the components of a positive working relationship.

Children’s Health Care, 21, 90–95.

Larsen, D., Attkisson, C., Hargreaves, W. & Nguyen, T.

(1979) Assessment of client/patient satisfaction: devel-

opment of a general scale. In: Evaluation and Program

Planning, Vol. 2, pp. 197–207.

Little, R. J. A. & Rubin, D. B. (1987) Statistical analysis

with missing data. John Wiley and Sons, New York.

Rosenbaum, P., King, S., Law, M., King, G. & Evans, J.

(1998) Family-centred service: a conceptual frame-

work and research review. Physical and Occupational

Therapy in Pediatrics, 18, 1–20.

Stallard, P. & Lenton, L. (1993) How satisfied are parents

of pre-school children who have special needs with the

services they have received? A customer survey. Child:

Care Health and Development, 18, 197–205.

Van Riper, M. (1999) Maternal perceptions of family-

provider relationships and well-being in families of

children with Down syndrome. Research in Nursing

and Health, 22, 357–368.

Woodside, J., Rosenbaum, P., King, S. & King, G. (2001)

Family-centred service: developing and validating a

self-assessment tool for pediatric service providers.

Children’s Health Care, 30, 237–252.