Abstract
This initial evaluation of the My Family’s Accessibility and Community Engagement (MyFACE) tool highlights sound psychometric properties that clinicians and researchers can use with families of children with disabilities to measure parental perceptions of community inclusion. Further development of the tool is warranted.
Parents of children with a disability provide care and are the linchpin for access to services and community activities. Parents, particularly mothers, advocate and support their child’s inclusion and participation (Gagnon et al., 2020; Nachshen & Jamieson, 2000). Research indicates that mothers experience high stress (Da Paz & Wallander, 2017), higher rates of some mental health conditions (Marquis et al., 2020), chronic medical conditions (Chambers & Chambers, 2015), sleep disruption (Bourke-Taylor et al., 2013), and exhaustion (Green, 2007). These maternal factors may influence capacity to engage in activities outside of the home, including support for their child’s participation. Past research has demonstrated a strong relationship between maternal mental health and participation in healthy activity at home or in the community (Bourke-Taylor, Pallant et al., 2012). The extent to which these issues influence family community participation is understudied and unknown. Factors can only be explored if the concepts can be operationalized and measured. Hence, a tool that enables measurement of parental perceptions of community engagement and accessibility is needed.
The social model of disability recognizes the key role of the contextual factors that affect the social world, internal constructions related to disability, and interactions of a child or person with disability with their environment (Anastasiou & Kauffman, 2013). The International Classification of Functioning, Disability and Health (ICF; World Health Organization, 2001) provides a framework to consider environmental factors, such as accessibility and attitudes, that can affect the participation of children with disabilities. The interplay between environmental factors and participation for children with cerebral palsy was classically investigated in a multinational European study (Colver et al., 2011). In this project, researchers interviewed 743 parents of children with cerebral palsy and measured perceptions of environmental features that supported or prevented access for their child. The study revealed substantial differences among parent experiences across nine European cities with regard to local environmental features such as physical accessibility, social supports, and attitudes of family and friends. This study points to the need for improvements in community accessibility, family experience of inclusion, and opportunities for children with disabilities and their families (Chien, Branjerdporn, et al., 2017; Chien, Rodger, & Copley, 2017).
In this article, we describe a new measurement tool that was configured to enable further investigations of the differences in access to local community activities between families of children with disabilities and other local families unaffected by childhood disability. The My Family’s Accessibility and Community Engagement (MyFACE) tool was designed to measure the construct of parental perception of community accessibility and engagement for families raising a child with a disability. The MyFACE tool is used to investigate the “face” that families show the community based on their experiences of inclusion in the community to measure engagement. In this preliminary study, we focused on mothers’ perspectives.
In line with gold-standard guidelines in scale development, MyFACE was initially designed by identifying the concept or construct of interest, generating a pool of items and response items (Streiner et al., 2015). Specifically, the configuration of MyFACE followed DeVellis’s (2012) Stages 1 to 5 of scale development: (1) determine the construct to be measured, (2) generate item pool, (3) determine measurement format, (4) review item pool, and (5) consider inclusion of validation items. The construct was determined through the first author’s (Helen Bourke-Taylor’s) substantial research and clinical experience with mothers of children with a disability (Bourke-Taylor et al., 2010; Bourke-Taylor, Pallant et al., 2012). Research associated with a maternal health program (https://healthymothers-healthyfamilies.com/) was used to identify mental health, fatigue, and other health issues that are influential in family interactions with health providers and community activities (Bourke-Taylor et al., 2013, 2019 ; Bourke-Taylor & Jane, 2018).
Item pool generation occurred by considering common activities that families of preschool- or school-age children (ages 4–19 yr) are engaged in locally. The measurement format was informed by the Person–Environment–Occupation model, in which a good fit between a child or adult with a disability and their environment or an activity in the environment would involve physical accessibility, supportive attitudes and human assistance, appropriate technology, or adaptation of the activity to enable participation (Law et al., 1996). Moreover, good fit would mean that no changes were needed by the family supporting a child with a disability. Items were validated through author collaboration and consultation with three pediatric occupational therapy practitioners. This article adds three additional stages to DeVellis’s (2012) stages of scale development: (6) administer items to pilot sample, (7) evaluate items, and (8) produce final scale.
Internationally, the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) checklist is considered the gold standard in developing and evaluating the measurement properties of instruments (Mokkink et al., 2010). The COSMIN is a standardized framework used to both evaluate and select outcome measures that are considered reliable, valid, responsive, and clinically useful (Mokkink et al., 2010; Prinsen et al., 2018). The COSMIN represents international consensus regarding measurement properties (Mokkink et al., 2010, 2016). For this article, the COSMIN was used to guide the initial evaluation of MyFACE. Validity refers to how well the tool items represent the construct or dimension they purport to measure. Reliability refers to the extent to which the measurement is consistent and free from error (Mokkink et al., 2010).
This article is the first in a pair of studies to evaluate MyFACE. In this article, we investigate the content validity, construct validity, and internal reliability of the MyFACE tool with mothers of preschool- or school-age children ages 4 through 19 yr diagnosed with at least one disability. In the second study, we investigated factor structure and other psychometric properties (Bourke-Taylor et al., 2021). In this study, content validity is investigated because it is considered a precursor to other investigations. Content validity is the degree to which the instrument reflects the construct to be measured and specifically determines the relevance, comprehensiveness, and comprehensibility of the tool to a defined group (Terwee et al., 2018). Construct validity and internal reliability are important psychometric properties to explore when scales are initially developed. The following three research questions were used: What content (items) are most valid when the MyFACE tool is rated by mothers of preschool- or school-age children with disabilities? How adequate is the construct validity of the MyFACE tool? Hypotheses were that (1) child-related disability factors (behavior, health, and quality of life) will be associated with the total MyFACE scores; (2) mother-related health and well-being factors (health-promoting activity participation, mental health) will be associated with MyFACE total scores; (3) child- and mother-related factors together will explain a significant percentage of variation in MyFACE total scores; and (4) community factors that represent physical, attitude, and resource barriers affecting families will be scored by mothers with high frequency. Is the internal reliability of the MyFACE tool in the excellent range (Cronbach’s α > .80)?
Method
In this psychometric study, we used a cross-sectional anonymous online survey design to gather data at a single time point. The survey was hosted by Qualtrics Research Platform (Version December 2017). This project was approved by the Monash University Human Research Ethics Committee (Project 10958).
Participants and Recruitment
Inclusion criteria were met if participants were interested mothers (birth, adoptive, foster) of children with a diagnosed disability, able to complete the online survey in English, and residing in Australia at the time of survey completion. Participants were recruited over 6 mo via a flyer distributed through disability organizations, childhood advocacy organizations, social media, and other digital or paper-based channels; interested mothers self-selected to participate or shared the flyer with their own network.
Instrumentation
A specifically designed online questionnaire was configured, consisting of demographic questions, mother-related questions such as sleep quality (good or poor), child-related questions such as diagnosis, the MyFACE tool, and five psychometrically sound scales.
Initial Configuration of MyFACE
The construct measured by the MyFACE tool is parental perception of community accessibility and engagement for their family with a child who has a disability. Theoretical underpinnings included the ICF and rights-based constructions of equivalency, such that common local opportunities should be available to all, regardless of ability or disability, and the interface between biological and social dimensions of disability (Anastasiou & Kauffman, 2013; World Health Organization, 2001). The scale consisted of 21 local community items, such as attending local library activities; playgrounds, parks, beaches, or other open spaces; restaurants; religious places; shopping facilities; recreational facilities; movie theaters; museums; galleries; health, dental, and medical facilities; other people’s homes; and preschool or school gatherings.
Mothers rated their family’s level of inclusion and involvement in each activity by selecting the appropriate face on a 5-point Likert scale, on which 0 = not interested/not applicable to us (no face); 1 = excluded, maximal changes and/or support required (very sad face); 2 = somewhat included and involved, many changes and/or support required (sad face); 3 = moderately included and involved, minor changes and/or support required (happy face); 4 = included and involved, no changes and/or support required (very happy face). Higher scores indicated that mothers perceived the community to be more accessible and that the family was more engaged. Additionally, 16 options were provided for parents to identify barriers to family inclusion and engagement that may limit family involvement.
Mother-Related Factors
Depression Anxiety Stress Scales.
The Depression Anxiety Stress Scales (DASS; Lovibond & Lovibond, 1995) make up a 21-item tool that evaluates mental health symptomatology across three subscales: Depression, Anxiety, and Stress. Respondents self-rate their symptoms along a 4-point Likert scale ranging from 0 (did not apply to me) to 3 (applied to me very much or most of the time), indicating the presence of the issue over the past week. Each of the three DASS subscales has good internal consistency and reliability (Cronbach’s α = .91, .84, and .90 for Depression, Anxiety, and Stress, respectively; Lovibond & Lovibond, 1995). Higher scores indicate more symptoms of depression, anxiety, or stress.
Health Promoting Activities Scale.
The Health Promoting Activities Scale (HPAS; Bourke-Taylor, Law et al., 2012) is an eight-item scale that measures the frequency with which mothers self-report participating in a self-selected, health-promoting activity. Respondents estimate the frequency of their participation on a 7-point scale ranging from 1 (never) to 7 (once/more every day), indicating how often they participated over the week or in the past year. The HPAS demonstrates good internal consistency (Cronbach’s αs = .73–.78) and construct validity. Higher scores indicate more frequent participation in healthy activities.
Child-Related Factors
Pediatric Quality of Life Inventory.
The Pediatric Quality of Life Inventory (PedsQL; Varni et al., 2003) is a 23-item scale designed to evaluate the child’s health-related quality of life from the parent’s perspective. The PedsQL asks parents to rate the extent to which the child’s physical, emotional, social, and school functioning are a “problem.” The PedsQL has demonstrated excellent internal consistency and reliability for the Physical Health Summary Score (Cronbach’s α = .88) and Psychosocial Health Summary Score (Cronbach’s α = .86; Varni et al., 2001). Higher scores indicate higher health-related quality of life.
Child’s Challenging Behavior Scale, Version 2.
The Child’s Challenging Behavior Scale, Version 2 (CCBS–2; Bourke-Taylor et al., 2017), is a nine-item scale involving parent ratings of the prevalence of difficult behaviors. Statements ask respondents to rate their level of agreement on a 4-point ordinal scale ranging from 1 (strongly disagree) to 4 (strongly agree). The CCBS–2 demonstrates good internal reliability (Cronbach’s α = .77), construct validity, and criterion validity (Bourke-Taylor et al., 2018). Higher scores indicate that the child exhibits more challenging behaviors.
Data Management and Analysis
We analyzed data using IBM SPSS Statistics (Version 27). Descriptive statistics were used to explore demographic data as well as mothers’ ratings of inclusion and engagement within their local environments and need for change.
Content Validity
The first research question addressed the content validity of MyFACE. Content validity required determining relevance, comprehensiveness, and representativeness (Terwee et al., 2018). Hence, the scale was intended to be short (comprehensive), with different items representing common community destinations (representativeness). We determined the relevance of items by using a combination of (1) 100% agreement through expert opinion sought when scale items were developed, (2) frequency that mothers selected the option “not interested/not applicable” for the 21 items, and (3) checking high interitem correlations to identify redundant items. The percentage relevance was preset at 92%, requiring that the majority of mothers found the item relevant to their families and representative of community activities. Inferential statistics determined relationships among factors, including correlation analysis, between-group differences, and regression analysis.
Construct Validity
Child Factors and MyFACE Scores
Regarding Hypothesis 1, we expected that MyFACE total scores would correlate positively and significantly with child factors (child’s quality of life: PedsQL scores) and negatively and significantly with the child’s externalizing behaviors (CCBS–2 score). We investigated these hypotheses using parametric correlation statistics. Pearson’s product–moment correlation coefficients were computed to compare MyFACE with the PedsQL and CCBS–2. Discriminative, or known-groups validity, is another method used to determine construct validity (Mokkink et al., 2019). Hence, known-groups validity was selected to explore the validity of the MyFACE construct. It was expected that between-groups differences would be found in MyFACE scores among groups of children with more complex disabilities, such that groups of children with more diagnoses (higher complexity) were expected to have lower mean scores on the MyFACE tool. A single-factor analysis of variance (ANOVA) was run to compare mean MyFACE scores among groups.
Mother Factors and MyFACE Scores
Regarding Hypothesis 2, we expected that MyFACE total scores would correlate significantly and positively with mothers’ healthy activities and negatively and significantly with mothers’ symptoms of depression, anxiety, and stress. We investigated these hypotheses using correlation statistics among MyFACE, HPAS, and DASS. Known-groups validity was selected to further confirm construct validity. To further evaluate differences between MyFACE scores among mothers, we configured two groups: (1) mothers with and without mental health conditions and (2) mothers with and without good sleep quality.
Child and Mother Factors and MyFACE Scores
To address Hypothesis 3, we built a regression model using significant mother and child variables to predict MyFACE scores. It was envisaged that mothers of children with higher psychosocial functioning needs and those with higher levels of depression would score lower on MyFACE. It was expected that mothers who spend more time involved in health-promoting activities would score higher on MyFACE. A multiple regression was planned with MyFACE as the dependent variable.
Barriers to Community Inclusion
To address Hypothesis 4, we estimated that mothers would report environmental barriers related to MyFACE scores with high frequency. It was expected that mothers would rate with high frequency (>50%) community factors that represent physical, attitudinal, and resource barriers. Mothers were asked to select any of 16 community factors that limited their family’s inclusion and involvement in their local community.
Internal Consistency
Research Question 3, or internal consistency, was determined through measurement of Cronbach’s α on the final item scale. We examined the out-of-range values using frequency tables, and a missing-value analysis indicated that all cases had less than 30% missing values, suggesting that none of the participants had more than 30% of the questions unanswered. As a result, all cases were retained for the analyses. The overall missing values were less than 10% for the data as a whole. Missing-value analysis (Little’s missing completely at random [MCAR] test) was not significant (p = .79), indicating that the pattern of missing items was random, as indicated by a probability of >.05. As a result, missing data were treated as MCAR. After the data screening and before conducting the multiple regressions, the assumptions were evaluated. To obtain a result that can be generalized with other samples, Tabachnick and Fidell (2019) provided the following guideline for establishing the minimum sample size for a regression: N > 50 + 8m (where m is the number of independent variables). Therefore, 74 cases were required for the multiple regression analysis.
Results
All mothers had completed secondary school, and 58.5% had a bachelor’s degree or higher, although only 55.8% were in paid employment, reflecting caregiver responsibilities (Table 1). Most mothers had been diagnosed with a mental health condition (54.5%), were partnered (72.7%), and had only one child (49.3%). Moreover, DASS subscales revealed that mothers reported stress (n = 31; 40.3%), anxiety (n = 33; 42.9%), and depressive symptoms (n = 33; 42.9%) in the moderate to severe range with high frequency.
Characteristics of Participants
Note. N = 77. ASD = autism spectrum disorder; CP = cerebral palsy; DASS = Depression Anxiety Stress Scales.
Percentages are rounded to the first decimal place and may therefore add up to 101 or 99.
Some mothers were diagnosed with multiple mental health conditions, and no other data on maternal condition or disability were collected.
Participants checked as many as applied.
Content Validity and Initial Item Investigation
Seventy-seven mothers attempted the 21-item MyFACE, although 8 mothers missed at least one item. Content validity involved determining how well items represented community items from the mothers’ perspective. When items were rated “not interested/not applicable,” they were considered not representative of community items and not relevant to families of children with a disability. Nine items were discarded because more than 7% of mothers determined that the item was irrelevant. Twelve items remained. Two further steps were taken to reduce items: (1) Interitem correlations were investigated, and (2) total scale scores were examined by calculating the internal consistency of the MyFACE without the item. Three items were further discarded when high item–item correlations were determined, and the item with the lowest percentage of nonrelevant response selections was retained. The internal consistency of the scale was higher without each of these items. For example, “Using local playgrounds” and “Using local parks, beaches, or other open spaces” were similar, with 5.2% and 2.6% not applicable, respectively. The interitem correlation was .62. The Cronbach’s α was .86 when either item was deleted, having no impact on the internal reliability of the scale. Therefore, the latter item was retained. In total, nine items were retained for the evaluation of construct validity (Table 2).
Mothers’ Ratings of Community Inclusion and Engagement Items and Response Set for MyFACE
Note. N = 77. MyFACE = My Family’s Accessibility and Community Engagement.
Although 77 mothers completed most items on the MyFACE tool, 69 mothers completed every item in the final nine-item scale. Hence, the missing data rate was 10%, with 100% completion rate required for a participant to be included to align with relevance (under content validity). Items were summed to obtain the total score. The total MyFACE scores ranged from 9 to 36 (n = 69), with a mean total score of 23.5 (SD = 6) and a median total score of 25.0. MyFACE responses were normally distributed.
Investigations of Construct Validity
Child Factors and MyFACE Scores
To address Hypothesis 1, we associated child factors with MyFACE total scores. The PedsQL Physical Health Summary Score did not correlate significantly with MyFACE. As expected, the MyFACE total score correlated positively and significantly with the PedsQL Psychosocial Health Summary Score (r = .36, p = .004; n = 63). These findings support the hypothesis that mothers of children with psychosocial functioning challenges are more likely to rate their family’s inclusion or involvement lower. Moreover, as expected, children who exhibited more challenging behaviors (higher CCBS–2 scores) rated community engagement lower (r = −.27, p = .03; n = 65).
Three groups of children with different numbers of diagnoses were configured to investigate known-groups validity. A single-factor, between-subjects ANOVA showed a significant difference in the mean scores of MyFACE ratings of mothers of children with more than one diagnosis (representing more complex disability), F(2, 66) = 4.98, p = .01, η2 = .13. A Newman–Keuls post hoc test (α = .05) showed that mothers of children with 0 to 1 condition rated their inclusion and involvement in the community higher (M = 25.53, SD = 5.45; n = 34), on average, than the mothers of children in each of the other groups: either 2 diagnoses (M = 22.57, SD = 6.25; n = 21) or those with 3 to 5 diagnoses (M = 20.14, SD = 5.07; n = 14). No significant difference was found in scoring MyFACE between mothers with 2 diagnoses and those with 3 to 5 diagnoses. This finding indicates that mothers of children with higher care needs may have greater difficulty accessing the community.
Mother Factors and MyFACE Scores
To address Hypothesis 2, we related mother factors with MyFACE total scores. As expected, mothers’ total MyFACE scores correlated significantly and positively with mothers’ total HPAS scores (r = .35, p = .003; n = 69). Moreover, MyFACE scores correlated significantly and negatively with the DASS Stress (r = −.31, p = .009; n = 69), Anxiety (r = −.42, p = .001; n = 67), and Depression (r = −.43, p = .001; n = 69) subscale scores. Within the sample, a group of mothers reported being diagnosed with a mental health condition (n = 42), and another group reported having no mental health condition (n = 34). Total MyFACE scores of mothers with a mental health condition were compared with those of other mothers. A single-factor, between-groups ANOVA found a significant difference between mothers with a mental health diagnosis and those without, F(1, 66) = 5.10, p = .027, η2 = .07. On average, mothers with a mental health diagnosis (M = 22.03, SD = 5.88; n = 38) had lower MyFACE scores than mothers without a mental health diagnosis (M = 25.59, SD = 5.60; n = 30).
A high proportion of mothers reported poor quality sleep (51.9%). Two groups were configured to investigate differences in MyFACE scores on the basis of groups of mothers with good versus poor sleep. A single-factor, between-groups ANOVA found a significant difference between mothers who reported good versus poor sleep quality, F(1, 67) = 11.16, p = .001, η2 = .14. On average, mothers who reported poor quality sleep (M = 21.46, SD = 5.82; n = 37) had lower MyFACE scores than mother who reported good quality sleep (M = 25.94, SD = 5.22; n = 32).
Child and Mother Factors and MyFACE Scores
To address Hypothesis 3, we investigated which child and mother factors influenced MyFACE scores. Factors that correlated moderately and significantly were entered into a model to determine relative contribution to variance in MyFACE total scores. Post hoc analysis was done to investigate predictive factors. A multiple regression was performed on the data with MyFACE as the dependent variable. Three predictors were included in the model: DASS–Depression subscale, PedsQL Psychosocial Health Summary Score, and HPAS. One child factor (PedsQL Psychosocial Health Summary Score) and two mother factors (HPAS and DASS–Depression subscale) were among the predictors. The intercorrelations among the variables are provided in Table 3.
Intercorrelations Among the Variables and Results of Regression Analysis to Predict MyFACE Scale Scores With Child- and Mother-Related Factors
Note. N = 62. DASS–Depression = Depression Anxiety Stress Scales–Depression subscale; HPAS = Health Promoting Activities Scale; MyFACE = My Family’s Accessibility and Community Engagement; PedsQL = Pediatric Quality of Life Inventory.
p < .05.
p < .01.
p < .001.
As can be seen in Table 3, significant relationships were found between MyFACE and all the predictors in the model. Mothers with more depressive symptoms reported lower MyFACE scores. Mothers of children with fewer psychosocial problems had higher MyFACE scores. Moreover, mothers who reported more participation in health-promoting activities reported higher MyFACE scores. The three predictors together explained 27% (R 2 = .268; see Table 3) of the variation in MyFACE scores, which was significant, F(3, 61) = 7.09, p < .001. When all the predictors were considered, PedsQL Psychosocial Health Summary Score and HPAS were no longer significant (see Table 3). The most important predictor in this model was DASS–Depression subscale. Higher depression scores had a direct impact on MyFACE scores. Mothers of children with higher psychosocial functioning needs tended to score lower on MyFACE, and this finding is partly because they were more likely to have higher levels of depression. Similarly, mothers who spent more time involved in health-promoting activities scored higher on MyFACE but only because they had lower levels of depression.
Barriers to Community Inclusion
To address Hypothesis 4, we investigated barriers to MyFACE items. Mothers selected “the attitudes of other adults and children” (59.7%) with high frequency. Mothers also rated child-related barriers (e.g., “my child’s behavior”; 71.4%) and resource barriers (e.g., “my own and/or my partner’s energy levels” [68.8%] and “time” [51.9%]) with high frequency (Table 4).
Mothers’ Perspectives on Barriers to Community Inclusion for MyFACE Items
Note. N = 77. MyFACE = My Family’s Accessibility and Community Engagement.
Investigation of Internal Reliability
The Cronbach’s α for the nine-item MyFACE (Cronbach’s α = .85) was excellent.
Discussion
Initial psychometrical evaluations confirmed a nine-item, internally reliable MyFACE tool with an appropriate response set. Guided by Terwee et al. (2018), this study provides evidence of content validity, including relevance of items and response, feasibility of online and independent completion of the tool, and comprehensibility of the nine-item scale. Hypothesis testing supported construct validity of the MyFACE tool.
From mothers’ perspectives, poorer mental health and fewer healthy behaviors were associated with perceiving the community as less accessible and engaging for their families. Maternal depressive symptoms directly affected MyFACE scores. Poor sleep and low energy levels were both identified as barriers to engaging in the community. Past occupational therapy research has suggested that addressing maternal mental health as a primary outcome has a flow-on effect to improving participation in healthy activities in the community and improved quality of life for children with a disability (Bourke-Taylor et al., 2019).
Mothers of children with more complex needs scored their community as less accessible and engaging for their families. The psychosocial quality of life of children with a disability was associated with perception of family engagement in the community. These findings are in line with past research that indicated that families of children with behavioral challenges or higher care needs tend to prefer home-based activities and experience difficulty in the community (Bourke-Taylor et al., 2010; Resch et al., 2010). Past research has also highlighted that additional caregiving responsibilities can overwhelm parents and reduce involvement in everyday activities outside of the home (Mitchell et al., 2016).
When both child and maternal factors were compared to determine whether both child and maternal factors influenced the way mothers scored MyFACE, both factors were significant. Maternal depression was the strongest predictor and influenced both ratings of psychosocial issues experienced by the child and mothers’ participation in healthy activities. These findings were reinforced and congruent with the barriers to community accessibility and engagement that mothers identified. For example, the top barriers were “my child’s behavior” (71.4%) and “my own and/or my partner’s energy levels” (68.8%). In sum, the construct validity of the MyFACE tool suggests that preliminary investigations support the defined construct. Future studies should continue to examine the construct validity and other psychometrics with larger samples and with fathers of children with a disability.
Limitations
Limitations in this research include only seeking mothers’ perspectives, inadequate sample size for other investigations of validity such as structural validity, and limitations to investigate criterion validity because of a lack of similar tools to compare with MyFACE. Future studies might include an identifiable data set so that missing items can be retrieved from participants. Fathers must be included in future samples to confirm applicability of MyFACE to both parents. A larger sample size overall will enable further psychometric testing, such as structural validity, criterion validity, reliability, and properties such as the standard error of measurement, to determine suitability as an outcome measure. A second published study investigating structural validity confirmed unidimensionality of MyFACE and identified the importance of the relationship between healthy maternal behaviors and MyFACE scores (Bourke-Taylor et al., 2021).
Implications for Occupational Therapy Practice
The results of this study have the following implications for occupational therapy practice: Occupational therapists can use MyFACE, a nine-item measurement tool with a 5-point response scale, to measure parental perceptions of community accessibility and engagement as experienced by families of children with disabilities. Occupational therapists can apply the sound construct validity of the MyFACE tool. Lower scores may be associated with significantly lower maternal mental health as well as fewer healthy behaviors, poorer child-related quality of life, and more complex disability. The internal consistency of the MyFACE tool is strong (Cronbach’s α = .85) Future research is necessary to further explore the psychometric properties of the MyFACE tool for use in occupational therapy practice and research.
Conclusion
Within a family-focused paradigm, mothers’ perspectives are important in the provision of services for children with disabilities (King et al., 2017). In this preliminary study, the presence of barriers such as stigma, lack of skilled support, and a lack of time all challenged family perceptions of community accessibility and engagement. Clinically, it seems likely that family attitudes and perceptions are likely to influence interventions; therefore, addressing perceived barriers may enhance family engagement in healthy community-based activities. Strengths-based approaches such as coaching (King et al., 2019) may benefit from inclusion of the MyFACE tool when communicating with families to better understand maternal perceptions and to identify desired community destinations and barriers that can then be managed. Further family-focused practice might include resources to promote family participation and to build supports.
Measurement tools are an essential part of service provision for families raising a child with a disability. MyFACE is a brief, family-focused, self-report tool. Preliminary investigations suggest that MyFACE may be a sound tool for researchers or practitioners to use when community accessibility and engagement are a priority.
Footnotes
Acknowledgments
We acknowledge with gratitude the participation of all the mothers in this study.