Disability and Rehabilitation
ISSN: 0963-8288 (Print) 1464-5165 (Online) Journal homepage: http://www.tandfonline.com/loi/idre20
Parent-mediated intervention: adherence and adverse effects David McConnell, Miriam Parakkal, Amber Savage & Gwendolyn Rempel To cite this article: David McConnell, Miriam Parakkal, Amber Savage & Gwendolyn Rempel (2015) Parent-mediated intervention: adherence and adverse effects, Disability and Rehabilitation, 37:10, 864-872, DOI: 10.3109/09638288.2014.946157 To link to this article: http://dx.doi.org/10.3109/09638288.2014.946157
Published online: 30 Jul 2014.
Submit your article to this journal
Article views: 104
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=idre20 Download by: [University of Florida]
Date: 12 November 2015, At: 06:55
http://informahealthcare.com/dre ISSN 0963-8288 print/ISSN 1464-5165 online Disabil Rehabil, 2015; 37(10): 864–872 ! 2014 Informa UK Ltd. DOI: 10.3109/09638288.2014.946157
RESEARCH PAPER
Parent-mediated intervention: adherence and adverse effects David McConnell1, Miriam Parakkal1, Amber Savage1, and Gwendolyn Rempel2 Family and Disability Studies Initiative, Rehabilitation Medicine, University of Alberta, Alberta, Canada and 2Faculty of Nursing, University of Alberta, Alberta, Canada
Downloaded by [University of Florida] at 06:55 12 November 2015
1
Abstract
Keywords
Purpose: The purpose of this study was to investigate parent implementation of home-based therapy regimens, and the relationship between implementation and family-level outcomes. Method: A stratified (by child age group) random sample of 538 families raising children with disabilities in Alberta, Canada took part. Participants completed the Family Life Survey, which incorporated child and family measures, and items measuring parent implementation of homebased therapy regimens. Results: Parents are more likely to implement therapeutic regimens when these are ‘‘enfolded’’ into other daily activities and routines. If parents have to ‘‘find a slot’’ in the daily routine to implement therapy, they will sacrifice personal leisure, participation in paid work, and time spent with other family members. Greater parent sacrifices/trade-offs was negatively associated with family well-being. Conclusion: As a general rule, children do well when their families do well, and families do well when they have the resources they need to juggle work and family and care demands. Recruiting parents as interventionists can tax family resources. Rehabilitation professionals must weigh up the pros and cons of parent-mediated intervention, and look to enfolding therapy into the everyday family routine.
Developmental disability, family-centered care, home programs, parent, therapy History Received 14 November 2013 Revised 8 July 2014 Accepted 15 July 2014 Published online 30 July 2014
ä Implications for Rehabilitation
Parents are more likely to implement prescribed therapeutic activities and exercises when these are ‘‘enfolded’’ into the everyday family routine. Higher levels of parental adherence and implementation of home-based therapy programs is associated with poorer family well-being. Rehabilitation professionals must weigh the pros and cons of parent-mediated intervention, and help families find ways to enfold intervention into their daily routine.
Background Rehabilitation professionals routinely prescribe home programs, comprising therapeutic activities or exercises, for parents to implement with their children. The potential benefits of such parent-mediated intervention are well documented. In experimental settings, benefits have been reported in areas such as promoting the social and communication skills of children with autism, and improving the upper limb functioning of children with cerebral palsy [1–7]. However, in practice settings, parental nonadherence with home programs is a seemingly intractable ‘‘problem’’ [8–11]. The limited available data suggest that parents typically implement some but not all prescribed activities, and/or they implement these some but not all of the time [8,9]. A small number of studies have investigated factors influencing parent adherence and implementation of home therapy regimens. To date, most studies have been exploratory and descriptive: Research has been guided by theory, but rarely have
Address for correspondence: David McConnell, PhD, Professor and Director, Family and Disability Studies Initiative, Rehabilitation Medicine, University of Alberta, Alberta, Canada. Tel: 1-780-492-7475. E-mail: [email protected]
explicit hypotheses been tested. One theory that has guided research on this topic is eco-cultural theory [12–14]. This theory offers a framework for understanding how parents allocate resources including, for example, whether and how much time and energy is allotted to implementing therapy at home. A core premise of eco-cultural theory is that all families face the same central adaptive challenge of sustaining a daily routine that is viable given available resources, and congruent with their beliefs, values and goals, and, with the needs, interests and competences of individual family members. Another core premise is that everyday family life is culturally patterned: families utilize cultural resources to make sense of the events and circumstances of their lives, and to create and adapt their everyday routine. A third core premise is that while families utilize cultural resources, they also develop new repertoires of deployable meanings and competences, and use these to construct new, perhaps even subversive strategies of action, such as cutting back on time spent implementing home programs. Exploratory research findings suggest that the extent to which parents implement home programs may be influenced by a multiplicity of inter-related eco-cultural factors. One potential influence is the level of concordance between parent and professional perspectives or goals. Higher levels of
Parents as interventionists
Downloaded by [University of Florida] at 06:55 12 November 2015
DOI: 10.3109/09638288.2014.946157
parent–professional concordance may be found when children are younger: Parents are themselves enculturated into the medicalrehabilitation model of disability and are therefore predisposed to viewing disability as abnormal and intrinsic to their child [15,16]. Consequently, parents may feel compelled to do whatever they can to ‘‘fix’’ their child. Leiter [15] describes this as the therapeutic imperative. However, as expectation meets experience, some parents begin to redefine themselves, their situation, and the very concept of normal itself [16,17]. Landsman [16] observed that ‘‘in one’s home, alone with the child and apart from the stresses of competitive mothering or medical diagnoses, a mother comes to understand her child as simply himself/herself, neither terribly unusual nor comparable to other children on any scale that makes sense’’ (p. 138). Subsequently, parent and professional perspectives may diverge, and the implementation of therapy at home may become a lesser priority for these parents. Several studies have documented a disconnect between parent and professional perspectives and goals, with professionals focusing more on impairment reduction and parents prioritizing their child’s subjective well-being [11,18,19]. In one recent study, Wiart et al. [11] conducted a series of focus groups exploring parent perspectives on goals and goal setting in relation to occupational and physical therapy for children with cerebral palsy. The study found that parents did not see any obvious connection between therapy goals and their own goals for their children. The parents prioritized their child’s happiness over therapy goals, and what the parents perceived as integral to their child’s happiness was being accepted for who they are, making friends, and the opportunity to make life choices without societal barriers. Notably, some parents reported forgoing therapy to focus on their child’s enjoyment and/or to simply enjoy family life. Another eco-cultural factor potentially influencing parent implementation is the degree of fit between family resources (e.g. time, energy, skills) and demands [10,11,20–22]. In the face of many changing and competing demands parents may have difficulty ‘‘finding a slot’’ in their daily routine to implement therapy [15,23–26]. Parents may, for example, have to contend with competing moral imperatives, such as the need to ensure that their other children are not always missing out (e.g. on quality time with their parents). Alternatively, or in addition, parents may have to attend to other more basic needs, such as earning a living, before allocating or re-allocating resources to implementing a home program. Notwithstanding, if parents believe that implementing the exercises and activities prescribed for their child is ‘‘the right thing to do’’, they may free-up resources by sacrificing personal leisure time, paid work and/or family activities and time together [15,20,23–25,27,28]. A third eco-cultural factor that may influence parent adherence and implementation is level of contextualization, that is, the extent to which intervention is enfolded into the everyday family routine. For example, interventions may be embedded within everyday activities such as meal and bath times, play dates, domestic chores and/or bed time rituals. When time is in short supply, higher levels of contextualization may translate into higher levels of parent implementation. Professionals may work with parents to contextualize interventions, or parents may exercise their own initiative [21,22]. Hinjosa and Anderson [21], for example, found that contextualizing is a common adaptive strategy used by parents to integrate multiple and potentially competing roles and responsibilities. The extent to which interventions are contextualized however may depend on the complexity, intensity and adaptability of the intervention or home program itself: simple, low intensity interventions are perhaps more likely to be contextualized and implemented than are complex or time-intensive interventions [9,10,19,29].
865
When the level of contextualization is low, the demands associated with carrying out prescribed home programs may have adverse effects on the parent and family. There is evidence that time pressure – and time spent providing health care – is associated with heightened parent stress and mental health problems [9,30–37]. For example, in Australia, Sawyer et al. [34] collected time-use and mental health data from 156 mothers of children with cerebral palsy and found a significant positive association between time devoted to care-work and maternal mental health problems, including depressive symptoms. However, to date, most studies have focused on the potential benefits of parent-mediated intervention with little attention given to the potential risks or adverse effects. Further research is needed to address this significant gap in knowledge. The purpose of the present study was to investigate parent implementation of prescribed home programs, including factors influencing adherence, and the relationship between implementation and family well-being. Our first hypothesis was that the degree to which parents implement home-based programs is associated with (a) the level of concordance between parent and professional goals, (b) difficulty in finding a slot in the daily routine to implement therapy, and (c) the extent to which the prescribed exercises or activities are contextualized, that is, enfolded into the everyday family routine. Our second hypothesis was that, in a cultural context where implementing therapy may be viewed as a moral imperative, (d) parents would make sacrifices/trade-offs – cutting back on personal leisure, family time and/or paid work – to free up the time required for implementation (i.e. there is a positive association between parent sacrifices/trade-offs and implementation); and, (e) that parent trade-offs/sacrifices would be negatively associated with family well-being.
Methods The study reported here was nested within a 3 year – survey plus interview – study of family life and out-of-home placement of children with disabilities in Alberta, Canada. The analysis reported in this paper is based on survey data collected in Year 1 (2009) of this study. The sampling frame for the study consisted of all English-speaking families registered with the Family Supports for Children with Disabilities (FSCD) program – a provincial government program for families raising children with disabilities – and in receipt of child-focused services. Following Ethics Board approval (File#B-060808), and with the assistance of FSCD personnel, a stratified (by child age group: early childhood, middle childhood, adolescence) random sample of families were invited to take part. The recruitment procedure was based on the Dillman [38] method. First, families were sent a letter informing them about the forthcoming survey. Within 2 weeks, each family received a copy of the survey package, including the ‘‘Family Life Survey’’, cover letter, consent form, pencil, and a return postage paid envelope. Approximately 2 weeks later, a follow-up reminder postcard was mailed out. Then approximately 2 weeks after this, another copy of the survey package was dispatched to families who had not yet responded. All respondents received an honorarium of CAN$30.00. Family life survey The Family Life Survey, which was completed by the primary parent–carer, incorporated previously validated items and scales, and new items created for the purpose of this study. Prior to administration, the Family Life Survey was reviewed (i.e. pilot tested) by a parent–carer advisory group. Members of the advisory group completed the survey individually and then
866
D. McConnell et al.
critiqued the survey together as a group. Based on the feedback received, changes were made to the layout of the survey, and some potentially confusing routing instructions were revised.
1 ¼ ‘‘never’’ to 4 ¼ ‘‘often’’. The internal consistency reliability of this scale was a ¼ 0.81. The mean frequency of integration item score was calculated and used in this study.
Demographic data
Trade-offs
The survey incorporated items used in the National Longitudinal Survey of Children and Youth (NLSCY) [39] to collect demographic data including but not limited to data on household composition; primary parent–carer sex, age, ethnicity, educational attainment and employment status; and, the sex, age, impairment type and activity limitations of the (nominated if more than one) child with disability.
A series of yes/no questions asked parents about trade-offs or sacrifices they had made. Parents were asked, ‘‘To keep your family life running in the last 12 months, have you or your partner (1) ‘‘reduced your participation in leisure or sports?’’, (2) ‘‘cutback on family activities or time together?’’, (3) ‘‘worked fewer hours?’’, (4) ‘‘taken a less demanding job?’’, (5) ‘‘turned down a promotion or a better job?’’ and/or (6) ‘‘quit working altogether?’’. To obtain a measure of trade-offs, we summed the number of reported trade-offs to a maximum score of 3. To obtain a score of 3, a parent would have to respond ‘‘Yes’’ to items (1) and (2), and yes to any one or more of the work related items (i.e. items 3–6).
Disability severity
Downloaded by [University of Florida] at 06:55 12 November 2015
Disabil Rehabil, 2015; 37(10): 864–872
Disability severity was measured by a series of items including, ‘‘To what extent does a long term physical condition, mental condition, learning or health problem reduce the amount or the kind of (age-appropriate) activities this child can do?’’, with five response categories ranging from 1 ¼ ‘‘not at all’’, through to 5 ¼ ‘‘a great deal’’. Four additional items asked respondents to indicate how much difficulty (i.e. ‘‘no difficulty’’, ‘‘some difficulty’’ or ‘‘a lot of difficulty’’) the child has with speaking, walking, learning and personal care, compared with other children his/her age. The internal consistency reliability of this scale was a ¼ 0.81. The mean item score was calculated and used in this study. Adherence Level of parent–carer implementation or adherence with the home therapy regimen was measured by the item ‘‘To what extent do you carry out the activities/exercises prescribed by a health professional for this child?’’, with five response categories ranging from 1 ¼ ‘‘we do not carry out any at all’’ to 5 ¼ ‘‘we routinely carry out all’’. Time implementing Time spent implementing home-based programs was measured by the item: ‘‘In a typical week, approximately how much time [i.e. in hours] do you spend implementing home (therapy) programs for your child with disabilities’’. Due to a number of extreme values (outliers) the raw data were re-coded with 0 ¼ 0, 1 ¼ 1–3 h, 2 ¼ 4–6 h, 3 ¼ 7–9 h, 4 ¼ 10–12 h, 5 ¼ 13–15 h, and 6 ¼ greater than 15 h. Difficulty ‘‘slotting’’ Difficulty finding time to implement therapeutic regimens or home programs was measured by the item, ‘‘How much difficulty have you had finding a slot in your daily routine to carry out the prescribed activities/exercises for this child?’’ with five response categories ranging from 1 ¼ ‘‘no difficulty at all’’ through to 5 ¼ ‘‘a great deal of difficulty’’. Contextualization To obtain a measure of contextualization, we adapted the approach used by Dunst et al. [40]. Parent–carers were asked ‘‘How often are the activities/exercises prescribed for this child integrated into (that is, just a routine part of) each of the following activities. . .’’. Six activities were listed including meal times; bath times (including dressing and undressing); play/leisure time at home; bed time; play/leisure at a local park, playground or recreation centre; and, grocery shopping and other routine community outings. For each activity, parent–carers reported frequency of integration using a four point scale ranging from
Concordance No direct measure of concordance between parent and professional need interpretations or goals was obtained. However, one item was used as a proxy indicator. Specifically, parent–carers were asked to respond, using a five-point scale (0 ¼ strongly disagree through to 5 ¼ strongly agree) to the statement ‘‘In general, professionals respond to our family’s needs and priorities’’. We reasoned that parents would be more likely to agree with this statement when there was a high level of concordance, and the home therapy regimen addressed their hopes or goals for their child. Family well-being Two aspects of Family well-being were measured: Family Cohesion and Family Life Congruence. The selection of items to measure each of these constructs was theory-driven rather than data-driven. Confirmatory factor analysis (CFA) was employed to test the theory (regarding relationships between items and between the items and the underlying latent constructs) against the data, and evaluate the convergent and discriminant validity of the scales (described further below). Factor scores were then derived from the CFA. Family Cohesion is a measure of the strength of emotional bonds between family members. It was measured with five items each having four response categories ranging from strongly disagree to strongly agree. Four of these items were derived from the General Family Functioning scale of the Family Assessment Device [41]: These are [FC1] ‘‘We confide in each other’’; [FC2] ‘‘In times of crisis we can turn to each other for support’’; [FC4] ‘‘There are lots of bad feelings in our family’’; and, [FC5] ‘‘We don’t get along well together’’. One item was researcher generated [FC3] ‘‘I often feel that our family stays together only out of necessity’’. The internal consistency reliability of this scale was a ¼ 0.90. Family Life Congruence is a construct derived from ecocultural theory [13], and is a measure of how well the family is doing with respect to the adaptive challenge of sustaining a daily routine that is congruent or ‘‘fitting’’ with their beliefs, values and goals, and with the needs, interests and competences of individual family members. It was measured with 10 items, each having four response categories ranging from strongly disagree to strongly agree. The items, which were sourced and/or adapted from the Family Life Interview [42], are [FLC1] ‘‘We are able to do things together that are important to us’’; [FLC2] ‘‘We are always on edge wondering ‘what next?’’’; [FLC3] ‘‘We exist: Any hopes or dreams we had now seem out of reach’’; [FLC4] ‘‘I am creating
Parents as interventionists
DOI: 10.3109/09638288.2014.946157
the life I want for my child/ren’’; [FLC5] ‘‘We are trapped by our daily routine’’; [FLC6] ‘‘I feel trapped by my duties as a parent and caregiver’’; [FLC7] ‘‘I often worry that I don’t spend enough quality time with my other child/ren’’; [FLC8] ‘‘I often worry that my other children are missing out’’; [FLC9] ‘‘Our family life routine revolves around our child with disabilities’’; and, [FLC10] ‘‘Overall, we are able to balance the needs and wants of each family member’’. The internal consistency reliability of this scale was a ¼ 0.88.
Downloaded by [University of Florida] at 06:55 12 November 2015
Perceived stress Stress was measured using the four-item Perceived Stress Scale (PSS-4) [43]. With five response categories ranging from ‘‘never’’ to ‘‘very often’’, parent-respondents were asked ‘‘In the last month, how often have you felt. . .’’ [PSS1] ‘‘that you were unable to control the important things in your life?’’, [PSS2] ‘‘confident about your ability to handle your personal problems?’’, [PSS3] ‘‘that things were going your way?’’, and [PSS4] ‘‘difficulties were piling up so high that you could not overcome them?’’. The internal consistency reliability of this scale was a ¼ 0.82. Data analysis The data were analyzed using PASW (SPSS) v.18 and AMOS v.20. Data were entered and then re-entered into SPSS. Any discrepancies were identified and errors corrected. After cleaning the data (e.g. reverse scoring items where necessary), a descriptive, demographic profile of study participants was generated. Bayesian estimation, utilizing the Markov chain Monte Carlo algorithm, was then employed to impute missing values and transform ordered-categorical data into numeric data. This approach has a number of advantages over traditional approaches to the treatment of missing data, such as listwise or pairwise deletion or mean substitution, which can result in a significant
.39
e
PSS1
.59
e
PSS2
.74
e
PSS3
e
PSS4
.30 .26
FLC1
.63
e
FLC2
.69
e
FLC3
.73
e
FLC4
e
FLC5
e
FLC6
e
FLC7
e
FLC8
.66
e
FLC9
.73
e
FLC10
e
FC1
.70 .76
.30
e
FC2
e
FC3
e
FC4
e
FC4
loss of data and/or biased estimates. In essence, the Bayesian approach solves the problem of incomplete data through an iterative and progressive process, utilizing what is known and what is learned, to replace missing values with highly likely values [44,45]. CFA was then employed to investigate the validity (convergent and discriminant) of the Family Cohesion and Family Life Congruence scales, and to derive factor scores. CFA is used when the researcher has an expectation of how items will factor: CFA is used to test this expectation against the data [46]. The adequacy of fit was evaluated using the v2 goodness-of-fit test and ‘‘practical’’ fit indices including the comparative fit index (CFI40.95 suggests good fit); and, the 90% confidence interval around the root mean square error of approximation (RMSEA90), which should contain a value less than 0.06 to indicate the possibility of close model fit [47]. In addition, the fit of the hypothesized three factor model (Figure 1) was compared with the fit of alternative two factor models (described further below). Data from parents who were not expected to implement any kind of therapeutic intervention or home program were excluded from further analyses. Path analysis (i.e. structural equation modeling with observed variables) – an extension of multiple regression analysis – was then used to test the study hypotheses, controlling for child age and disability severity. The same fit indices used to evaluate the measures of family cohesion and family life congruence were used to evaluate the path models. Based on Mardia’s coefficient of multivariate kurtosis, the assumption of multivariate normality was satisfied.
Results A total of 538 families (41%) accepted the invitation to take part in the study and completed the Family Life Survey. Of these, 390 (72%) indicated that they were expected to carry out therapeutic activities/exercises with their disabled child. On average, these
Perceived Stress
.75
e
-.37 .46
.75
.66 .72
-.73
Family Life Congruence
-.48
.70 .68 .73
.80
867
.63
Family Cohesion
.80 .79
Model fit indices: X 2= 369.39, DF=143, p
ISSN: 0963-8288 (Print) 1464-5165 (Online) Journal homepage: http://www.tandfonline.com/loi/idre20
Parent-mediated intervention: adherence and adverse effects David McConnell, Miriam Parakkal, Amber Savage & Gwendolyn Rempel To cite this article: David McConnell, Miriam Parakkal, Amber Savage & Gwendolyn Rempel (2015) Parent-mediated intervention: adherence and adverse effects, Disability and Rehabilitation, 37:10, 864-872, DOI: 10.3109/09638288.2014.946157 To link to this article: http://dx.doi.org/10.3109/09638288.2014.946157
Published online: 30 Jul 2014.
Submit your article to this journal
Article views: 104
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=idre20 Download by: [University of Florida]
Date: 12 November 2015, At: 06:55
http://informahealthcare.com/dre ISSN 0963-8288 print/ISSN 1464-5165 online Disabil Rehabil, 2015; 37(10): 864–872 ! 2014 Informa UK Ltd. DOI: 10.3109/09638288.2014.946157
RESEARCH PAPER
Parent-mediated intervention: adherence and adverse effects David McConnell1, Miriam Parakkal1, Amber Savage1, and Gwendolyn Rempel2 Family and Disability Studies Initiative, Rehabilitation Medicine, University of Alberta, Alberta, Canada and 2Faculty of Nursing, University of Alberta, Alberta, Canada
Downloaded by [University of Florida] at 06:55 12 November 2015
1
Abstract
Keywords
Purpose: The purpose of this study was to investigate parent implementation of home-based therapy regimens, and the relationship between implementation and family-level outcomes. Method: A stratified (by child age group) random sample of 538 families raising children with disabilities in Alberta, Canada took part. Participants completed the Family Life Survey, which incorporated child and family measures, and items measuring parent implementation of homebased therapy regimens. Results: Parents are more likely to implement therapeutic regimens when these are ‘‘enfolded’’ into other daily activities and routines. If parents have to ‘‘find a slot’’ in the daily routine to implement therapy, they will sacrifice personal leisure, participation in paid work, and time spent with other family members. Greater parent sacrifices/trade-offs was negatively associated with family well-being. Conclusion: As a general rule, children do well when their families do well, and families do well when they have the resources they need to juggle work and family and care demands. Recruiting parents as interventionists can tax family resources. Rehabilitation professionals must weigh up the pros and cons of parent-mediated intervention, and look to enfolding therapy into the everyday family routine.
Developmental disability, family-centered care, home programs, parent, therapy History Received 14 November 2013 Revised 8 July 2014 Accepted 15 July 2014 Published online 30 July 2014
ä Implications for Rehabilitation
Parents are more likely to implement prescribed therapeutic activities and exercises when these are ‘‘enfolded’’ into the everyday family routine. Higher levels of parental adherence and implementation of home-based therapy programs is associated with poorer family well-being. Rehabilitation professionals must weigh the pros and cons of parent-mediated intervention, and help families find ways to enfold intervention into their daily routine.
Background Rehabilitation professionals routinely prescribe home programs, comprising therapeutic activities or exercises, for parents to implement with their children. The potential benefits of such parent-mediated intervention are well documented. In experimental settings, benefits have been reported in areas such as promoting the social and communication skills of children with autism, and improving the upper limb functioning of children with cerebral palsy [1–7]. However, in practice settings, parental nonadherence with home programs is a seemingly intractable ‘‘problem’’ [8–11]. The limited available data suggest that parents typically implement some but not all prescribed activities, and/or they implement these some but not all of the time [8,9]. A small number of studies have investigated factors influencing parent adherence and implementation of home therapy regimens. To date, most studies have been exploratory and descriptive: Research has been guided by theory, but rarely have
Address for correspondence: David McConnell, PhD, Professor and Director, Family and Disability Studies Initiative, Rehabilitation Medicine, University of Alberta, Alberta, Canada. Tel: 1-780-492-7475. E-mail: [email protected]
explicit hypotheses been tested. One theory that has guided research on this topic is eco-cultural theory [12–14]. This theory offers a framework for understanding how parents allocate resources including, for example, whether and how much time and energy is allotted to implementing therapy at home. A core premise of eco-cultural theory is that all families face the same central adaptive challenge of sustaining a daily routine that is viable given available resources, and congruent with their beliefs, values and goals, and, with the needs, interests and competences of individual family members. Another core premise is that everyday family life is culturally patterned: families utilize cultural resources to make sense of the events and circumstances of their lives, and to create and adapt their everyday routine. A third core premise is that while families utilize cultural resources, they also develop new repertoires of deployable meanings and competences, and use these to construct new, perhaps even subversive strategies of action, such as cutting back on time spent implementing home programs. Exploratory research findings suggest that the extent to which parents implement home programs may be influenced by a multiplicity of inter-related eco-cultural factors. One potential influence is the level of concordance between parent and professional perspectives or goals. Higher levels of
Parents as interventionists
Downloaded by [University of Florida] at 06:55 12 November 2015
DOI: 10.3109/09638288.2014.946157
parent–professional concordance may be found when children are younger: Parents are themselves enculturated into the medicalrehabilitation model of disability and are therefore predisposed to viewing disability as abnormal and intrinsic to their child [15,16]. Consequently, parents may feel compelled to do whatever they can to ‘‘fix’’ their child. Leiter [15] describes this as the therapeutic imperative. However, as expectation meets experience, some parents begin to redefine themselves, their situation, and the very concept of normal itself [16,17]. Landsman [16] observed that ‘‘in one’s home, alone with the child and apart from the stresses of competitive mothering or medical diagnoses, a mother comes to understand her child as simply himself/herself, neither terribly unusual nor comparable to other children on any scale that makes sense’’ (p. 138). Subsequently, parent and professional perspectives may diverge, and the implementation of therapy at home may become a lesser priority for these parents. Several studies have documented a disconnect between parent and professional perspectives and goals, with professionals focusing more on impairment reduction and parents prioritizing their child’s subjective well-being [11,18,19]. In one recent study, Wiart et al. [11] conducted a series of focus groups exploring parent perspectives on goals and goal setting in relation to occupational and physical therapy for children with cerebral palsy. The study found that parents did not see any obvious connection between therapy goals and their own goals for their children. The parents prioritized their child’s happiness over therapy goals, and what the parents perceived as integral to their child’s happiness was being accepted for who they are, making friends, and the opportunity to make life choices without societal barriers. Notably, some parents reported forgoing therapy to focus on their child’s enjoyment and/or to simply enjoy family life. Another eco-cultural factor potentially influencing parent implementation is the degree of fit between family resources (e.g. time, energy, skills) and demands [10,11,20–22]. In the face of many changing and competing demands parents may have difficulty ‘‘finding a slot’’ in their daily routine to implement therapy [15,23–26]. Parents may, for example, have to contend with competing moral imperatives, such as the need to ensure that their other children are not always missing out (e.g. on quality time with their parents). Alternatively, or in addition, parents may have to attend to other more basic needs, such as earning a living, before allocating or re-allocating resources to implementing a home program. Notwithstanding, if parents believe that implementing the exercises and activities prescribed for their child is ‘‘the right thing to do’’, they may free-up resources by sacrificing personal leisure time, paid work and/or family activities and time together [15,20,23–25,27,28]. A third eco-cultural factor that may influence parent adherence and implementation is level of contextualization, that is, the extent to which intervention is enfolded into the everyday family routine. For example, interventions may be embedded within everyday activities such as meal and bath times, play dates, domestic chores and/or bed time rituals. When time is in short supply, higher levels of contextualization may translate into higher levels of parent implementation. Professionals may work with parents to contextualize interventions, or parents may exercise their own initiative [21,22]. Hinjosa and Anderson [21], for example, found that contextualizing is a common adaptive strategy used by parents to integrate multiple and potentially competing roles and responsibilities. The extent to which interventions are contextualized however may depend on the complexity, intensity and adaptability of the intervention or home program itself: simple, low intensity interventions are perhaps more likely to be contextualized and implemented than are complex or time-intensive interventions [9,10,19,29].
865
When the level of contextualization is low, the demands associated with carrying out prescribed home programs may have adverse effects on the parent and family. There is evidence that time pressure – and time spent providing health care – is associated with heightened parent stress and mental health problems [9,30–37]. For example, in Australia, Sawyer et al. [34] collected time-use and mental health data from 156 mothers of children with cerebral palsy and found a significant positive association between time devoted to care-work and maternal mental health problems, including depressive symptoms. However, to date, most studies have focused on the potential benefits of parent-mediated intervention with little attention given to the potential risks or adverse effects. Further research is needed to address this significant gap in knowledge. The purpose of the present study was to investigate parent implementation of prescribed home programs, including factors influencing adherence, and the relationship between implementation and family well-being. Our first hypothesis was that the degree to which parents implement home-based programs is associated with (a) the level of concordance between parent and professional goals, (b) difficulty in finding a slot in the daily routine to implement therapy, and (c) the extent to which the prescribed exercises or activities are contextualized, that is, enfolded into the everyday family routine. Our second hypothesis was that, in a cultural context where implementing therapy may be viewed as a moral imperative, (d) parents would make sacrifices/trade-offs – cutting back on personal leisure, family time and/or paid work – to free up the time required for implementation (i.e. there is a positive association between parent sacrifices/trade-offs and implementation); and, (e) that parent trade-offs/sacrifices would be negatively associated with family well-being.
Methods The study reported here was nested within a 3 year – survey plus interview – study of family life and out-of-home placement of children with disabilities in Alberta, Canada. The analysis reported in this paper is based on survey data collected in Year 1 (2009) of this study. The sampling frame for the study consisted of all English-speaking families registered with the Family Supports for Children with Disabilities (FSCD) program – a provincial government program for families raising children with disabilities – and in receipt of child-focused services. Following Ethics Board approval (File#B-060808), and with the assistance of FSCD personnel, a stratified (by child age group: early childhood, middle childhood, adolescence) random sample of families were invited to take part. The recruitment procedure was based on the Dillman [38] method. First, families were sent a letter informing them about the forthcoming survey. Within 2 weeks, each family received a copy of the survey package, including the ‘‘Family Life Survey’’, cover letter, consent form, pencil, and a return postage paid envelope. Approximately 2 weeks later, a follow-up reminder postcard was mailed out. Then approximately 2 weeks after this, another copy of the survey package was dispatched to families who had not yet responded. All respondents received an honorarium of CAN$30.00. Family life survey The Family Life Survey, which was completed by the primary parent–carer, incorporated previously validated items and scales, and new items created for the purpose of this study. Prior to administration, the Family Life Survey was reviewed (i.e. pilot tested) by a parent–carer advisory group. Members of the advisory group completed the survey individually and then
866
D. McConnell et al.
critiqued the survey together as a group. Based on the feedback received, changes were made to the layout of the survey, and some potentially confusing routing instructions were revised.
1 ¼ ‘‘never’’ to 4 ¼ ‘‘often’’. The internal consistency reliability of this scale was a ¼ 0.81. The mean frequency of integration item score was calculated and used in this study.
Demographic data
Trade-offs
The survey incorporated items used in the National Longitudinal Survey of Children and Youth (NLSCY) [39] to collect demographic data including but not limited to data on household composition; primary parent–carer sex, age, ethnicity, educational attainment and employment status; and, the sex, age, impairment type and activity limitations of the (nominated if more than one) child with disability.
A series of yes/no questions asked parents about trade-offs or sacrifices they had made. Parents were asked, ‘‘To keep your family life running in the last 12 months, have you or your partner (1) ‘‘reduced your participation in leisure or sports?’’, (2) ‘‘cutback on family activities or time together?’’, (3) ‘‘worked fewer hours?’’, (4) ‘‘taken a less demanding job?’’, (5) ‘‘turned down a promotion or a better job?’’ and/or (6) ‘‘quit working altogether?’’. To obtain a measure of trade-offs, we summed the number of reported trade-offs to a maximum score of 3. To obtain a score of 3, a parent would have to respond ‘‘Yes’’ to items (1) and (2), and yes to any one or more of the work related items (i.e. items 3–6).
Disability severity
Downloaded by [University of Florida] at 06:55 12 November 2015
Disabil Rehabil, 2015; 37(10): 864–872
Disability severity was measured by a series of items including, ‘‘To what extent does a long term physical condition, mental condition, learning or health problem reduce the amount or the kind of (age-appropriate) activities this child can do?’’, with five response categories ranging from 1 ¼ ‘‘not at all’’, through to 5 ¼ ‘‘a great deal’’. Four additional items asked respondents to indicate how much difficulty (i.e. ‘‘no difficulty’’, ‘‘some difficulty’’ or ‘‘a lot of difficulty’’) the child has with speaking, walking, learning and personal care, compared with other children his/her age. The internal consistency reliability of this scale was a ¼ 0.81. The mean item score was calculated and used in this study. Adherence Level of parent–carer implementation or adherence with the home therapy regimen was measured by the item ‘‘To what extent do you carry out the activities/exercises prescribed by a health professional for this child?’’, with five response categories ranging from 1 ¼ ‘‘we do not carry out any at all’’ to 5 ¼ ‘‘we routinely carry out all’’. Time implementing Time spent implementing home-based programs was measured by the item: ‘‘In a typical week, approximately how much time [i.e. in hours] do you spend implementing home (therapy) programs for your child with disabilities’’. Due to a number of extreme values (outliers) the raw data were re-coded with 0 ¼ 0, 1 ¼ 1–3 h, 2 ¼ 4–6 h, 3 ¼ 7–9 h, 4 ¼ 10–12 h, 5 ¼ 13–15 h, and 6 ¼ greater than 15 h. Difficulty ‘‘slotting’’ Difficulty finding time to implement therapeutic regimens or home programs was measured by the item, ‘‘How much difficulty have you had finding a slot in your daily routine to carry out the prescribed activities/exercises for this child?’’ with five response categories ranging from 1 ¼ ‘‘no difficulty at all’’ through to 5 ¼ ‘‘a great deal of difficulty’’. Contextualization To obtain a measure of contextualization, we adapted the approach used by Dunst et al. [40]. Parent–carers were asked ‘‘How often are the activities/exercises prescribed for this child integrated into (that is, just a routine part of) each of the following activities. . .’’. Six activities were listed including meal times; bath times (including dressing and undressing); play/leisure time at home; bed time; play/leisure at a local park, playground or recreation centre; and, grocery shopping and other routine community outings. For each activity, parent–carers reported frequency of integration using a four point scale ranging from
Concordance No direct measure of concordance between parent and professional need interpretations or goals was obtained. However, one item was used as a proxy indicator. Specifically, parent–carers were asked to respond, using a five-point scale (0 ¼ strongly disagree through to 5 ¼ strongly agree) to the statement ‘‘In general, professionals respond to our family’s needs and priorities’’. We reasoned that parents would be more likely to agree with this statement when there was a high level of concordance, and the home therapy regimen addressed their hopes or goals for their child. Family well-being Two aspects of Family well-being were measured: Family Cohesion and Family Life Congruence. The selection of items to measure each of these constructs was theory-driven rather than data-driven. Confirmatory factor analysis (CFA) was employed to test the theory (regarding relationships between items and between the items and the underlying latent constructs) against the data, and evaluate the convergent and discriminant validity of the scales (described further below). Factor scores were then derived from the CFA. Family Cohesion is a measure of the strength of emotional bonds between family members. It was measured with five items each having four response categories ranging from strongly disagree to strongly agree. Four of these items were derived from the General Family Functioning scale of the Family Assessment Device [41]: These are [FC1] ‘‘We confide in each other’’; [FC2] ‘‘In times of crisis we can turn to each other for support’’; [FC4] ‘‘There are lots of bad feelings in our family’’; and, [FC5] ‘‘We don’t get along well together’’. One item was researcher generated [FC3] ‘‘I often feel that our family stays together only out of necessity’’. The internal consistency reliability of this scale was a ¼ 0.90. Family Life Congruence is a construct derived from ecocultural theory [13], and is a measure of how well the family is doing with respect to the adaptive challenge of sustaining a daily routine that is congruent or ‘‘fitting’’ with their beliefs, values and goals, and with the needs, interests and competences of individual family members. It was measured with 10 items, each having four response categories ranging from strongly disagree to strongly agree. The items, which were sourced and/or adapted from the Family Life Interview [42], are [FLC1] ‘‘We are able to do things together that are important to us’’; [FLC2] ‘‘We are always on edge wondering ‘what next?’’’; [FLC3] ‘‘We exist: Any hopes or dreams we had now seem out of reach’’; [FLC4] ‘‘I am creating
Parents as interventionists
DOI: 10.3109/09638288.2014.946157
the life I want for my child/ren’’; [FLC5] ‘‘We are trapped by our daily routine’’; [FLC6] ‘‘I feel trapped by my duties as a parent and caregiver’’; [FLC7] ‘‘I often worry that I don’t spend enough quality time with my other child/ren’’; [FLC8] ‘‘I often worry that my other children are missing out’’; [FLC9] ‘‘Our family life routine revolves around our child with disabilities’’; and, [FLC10] ‘‘Overall, we are able to balance the needs and wants of each family member’’. The internal consistency reliability of this scale was a ¼ 0.88.
Downloaded by [University of Florida] at 06:55 12 November 2015
Perceived stress Stress was measured using the four-item Perceived Stress Scale (PSS-4) [43]. With five response categories ranging from ‘‘never’’ to ‘‘very often’’, parent-respondents were asked ‘‘In the last month, how often have you felt. . .’’ [PSS1] ‘‘that you were unable to control the important things in your life?’’, [PSS2] ‘‘confident about your ability to handle your personal problems?’’, [PSS3] ‘‘that things were going your way?’’, and [PSS4] ‘‘difficulties were piling up so high that you could not overcome them?’’. The internal consistency reliability of this scale was a ¼ 0.82. Data analysis The data were analyzed using PASW (SPSS) v.18 and AMOS v.20. Data were entered and then re-entered into SPSS. Any discrepancies were identified and errors corrected. After cleaning the data (e.g. reverse scoring items where necessary), a descriptive, demographic profile of study participants was generated. Bayesian estimation, utilizing the Markov chain Monte Carlo algorithm, was then employed to impute missing values and transform ordered-categorical data into numeric data. This approach has a number of advantages over traditional approaches to the treatment of missing data, such as listwise or pairwise deletion or mean substitution, which can result in a significant
.39
e
PSS1
.59
e
PSS2
.74
e
PSS3
e
PSS4
.30 .26
FLC1
.63
e
FLC2
.69
e
FLC3
.73
e
FLC4
e
FLC5
e
FLC6
e
FLC7
e
FLC8
.66
e
FLC9
.73
e
FLC10
e
FC1
.70 .76
.30
e
FC2
e
FC3
e
FC4
e
FC4
loss of data and/or biased estimates. In essence, the Bayesian approach solves the problem of incomplete data through an iterative and progressive process, utilizing what is known and what is learned, to replace missing values with highly likely values [44,45]. CFA was then employed to investigate the validity (convergent and discriminant) of the Family Cohesion and Family Life Congruence scales, and to derive factor scores. CFA is used when the researcher has an expectation of how items will factor: CFA is used to test this expectation against the data [46]. The adequacy of fit was evaluated using the v2 goodness-of-fit test and ‘‘practical’’ fit indices including the comparative fit index (CFI40.95 suggests good fit); and, the 90% confidence interval around the root mean square error of approximation (RMSEA90), which should contain a value less than 0.06 to indicate the possibility of close model fit [47]. In addition, the fit of the hypothesized three factor model (Figure 1) was compared with the fit of alternative two factor models (described further below). Data from parents who were not expected to implement any kind of therapeutic intervention or home program were excluded from further analyses. Path analysis (i.e. structural equation modeling with observed variables) – an extension of multiple regression analysis – was then used to test the study hypotheses, controlling for child age and disability severity. The same fit indices used to evaluate the measures of family cohesion and family life congruence were used to evaluate the path models. Based on Mardia’s coefficient of multivariate kurtosis, the assumption of multivariate normality was satisfied.
Results A total of 538 families (41%) accepted the invitation to take part in the study and completed the Family Life Survey. Of these, 390 (72%) indicated that they were expected to carry out therapeutic activities/exercises with their disabled child. On average, these
Perceived Stress
.75
e
-.37 .46
.75
.66 .72
-.73
Family Life Congruence
-.48
.70 .68 .73
.80
867
.63
Family Cohesion
.80 .79
Model fit indices: X 2= 369.39, DF=143, p
