Occupational therapists are concerned with people, the occupations in which they engage, and the relationships between occupation and health and well-being. Occupations are the everyday activities that structure time and meet performance requirements for life roles (American Occupational Therapy Association [AOTA], 2008). Many factors influence a person’s choice of activities, including health status, life roles, and individual interests and capabilities as well as what is available, meaningful, culturally relevant, and environmentally possible. People who have a health condition or disability may experience reduced capacity to participate in daily occupations in self-care, work or productivity, leisure, and social participation (AOTA, 2008; Law, 2002). Certain life situations, such as the role of family caregiver, may also prevent full participation in occupation, with consequences for health.

Mothers of children with developmental disabilities experience higher rates of stress and depression than other mothers (Singer, 2006). The well-being of children (both with and without disabilities) is inextricably tied to their parents’ health and capacity to meet the demands of the developing child (Schor, 2003). Although much occupational therapy research has examined the lived experience, goals, and time use of mothers of children with developmental disabilities, no research to date has investigated maternal participation in self-selected occupations that promote health and well-being. Mothers of children with disabilities have less free time (Crowe & Florez, 2006), may face barriers to participation in health-promoting occupations in the limited spare time they do have, and may experience health disparities as a consequence (Bourke-Taylor, Howie, & Law, 2010).

In this article, we describe a new scale that measures the frequency with which mothers participate in self-selected leisure activities that promote health and well-being. We review the role of occupational therapy in health promotion and leisure; the need for outcome measures to quantify participation in health-promoting occupations; and the link between caregiving and health for mothers of children with disabilities. We then describe the development and evaluation of the Health Promoting Activities Scale (HPAS).

Occupational therapy has a unique, occupation-based approach that promotes and supports health in people and populations (AOTA, 2008; Scaffa, Van Slyke, & Brownson, 2008). Health promotion is a multipronged prevention strategy in which agencies at all levels—government, organization, and community—and individual health practitioners work to keep or create healthy environments and opportunities for or with people (World Health Organization, 2009). Occupational therapy practitioners may provide services at any of these levels within a health promotion capacity, although they are most likely to work at the individual health practitioner level (AOTA, 2008; Gupta, Chandler, & Toto, 2009; Scaffa et al., 2008). Health promotion strategies may incorporate any type of human occupation, including work, social participation, and leisure (Scaffa et al., 2008).

Previous research in occupational therapy has described how leisure occupations maintain health, rejuvenate the spirit, build skills and talents, and ultimately are personally validating, meaningful, enjoyable, and rewarding (Passmore, 2003; Passmore & French, 2003; Sandqvist, Akesson, & Eklund, 2005; Specht, King, Brown, & Foris, 2002). Leisure occupations are self-selected and thus meaningful, enjoyable, and important to the person participating. Research that connects health status and engagement in leisure occupations, however, is needed. Such research is fundamental to occupational therapy and a health promotion perspective.

Substantiation of the relationship between participation in leisure occupations and health status provides evidence to third-party payers and health care organizations that occupational therapy interventions maintain and promote health. Clients who experience a medical condition or disability may participate in leisure occupations infrequently, although participation is likely to be instrumental in achieving a state of good health. A review of 23 studies (most with cross-sectional designs) that investigated the relationship between health and occupation among samples of people with and without a disability found strong positive relationships between leisure occupations and health status in all populations (Law, Steinwender, & LeClair, 1998). Several of the studies reviewed examined both occupational engagement in everyday life and the addition of meaningful occupation to enhance daily life. The review also examined 10 studies that investigated the effects of a withdrawal of meaningful occupation in the lives of people with medical conditions and disabilities; this review suggested that the effects may be detrimental, although more research was recommended. The profession is committed to substantiating the relationship between occupation and health, a goal that is difficult to achieve without outcome measures (Gutman, 2008).

Creating outcome measures that quantify participation is challenging. As Coster (2008)  eloquently described the issue, if occupational therapy research aims to examine participation, then as a profession

Instruments that both measure participation and promote efficacy research have the added advantage of promoting occupationally based interventions (Coster, 2008). Coster (2006)  emphasized the need for occupation-based outcome tools that explore the relationships between what people do and how they report their health and changes in this relationship over time; such tools would contribute to the theoretical foundations of the profession. Clients have reported that participation in leisure is a high priority compared with many other daily occupations (Specht et al., 2002).

Outcome tools must accurately measure a defined construct in a reliable and valid way and be sensitive to important changes within the construct being measured. They must be brief and easy to use, entail a low burden for participants, and be designed to collect data from the client about the client (DeVellis, 2003).

The occupational therapy profession has engaged in service and advocacy on behalf of caregivers and recognizes their crucial role in the lives of people with disabilities (AOTA, 2007). As caregivers, mothers are responsible for numerous and constant tasks as the child grows (Bourke-Taylor, Howie, & Law, 2010). Mothers of children with disabilities spend a disproportionate amount of time in child and care occupations (Crowe & Florez, 2006), have less time to care for themselves and sleep (VanLeit & Crowe, 2002), have fewer opportunities to participate in paid work (Brehaut et al., 2004; Montes & Halterman, 2008; Powers, 2003), and consistently report poorer mental health and subjective well-being than other mothers (Brehaut et al., 2004; Montes & Halterman, 2007; Singer, 2006).

The link between caring and maternal health has often been assumed to be the result of grief (Green, 2007), social isolation (Davis et al., 2010; Feldman et al., 2007; Skok, Harvey, & Reddihough, 2006), constant pressure to navigate a largely unresponsive health care system on their child’s behalf (Green, 2007; Llewellyn, Thompson, & Whybrow, 2004), and social stigma related to raising a child with a disability (Green, 2003). No research has explored the relationship between maternal health and the occupations that a mother participates in outside of caregiving. It is not possible to fully explore such relationships without instruments that measure participation in self-selected occupations. Mothers themselves have described their wish for more time, opportunities, and support to pursue leisure activities for their own health, well-being, and development (Crowe & Florez, 2006; Davis et al., 2010; Donovan, VanLeit, Crowe, & Keefe, 2005; VanLeit & Crowe, 2002).

This article describes the initial development and psychometric evaluation of the HPAS. The construct measured by the HPAS is the frequency with which mothers participate in self-selected leisure occupations that promote or maintain health. We used a mixed-methodology instrument design model (Creswell, Fetters, & Ivankova, 2004) and framed two studies—an initial qualitative study to generate items and scoring criteria (Bourke-Taylor, Howie, & Law, 2010) and a quantitative study to examine the psychometric properties of the instrument. The aims of this research were as follows:

• To create a brief scale to measure the frequency with which mothers caring for a school-age child with a disability participated in self-selected leisure occupations;

• To determine the internal consistency of the scale using a sample of mothers;

• To determine the construct validity of the scale through correlation with psychometrically sound instruments that measure subjective health status; and

• To determine the extent to which the scale is able to differentiate between mothers grouped by differences in reported mental health and sleep habits.

In addition to the HPAS, we developed two other instruments after the qualitative study (Bourke-Taylor, Law, Howie, & Pallant, 2009, 2010).

As noted in the previous section, the construct measured by the HPAS is the frequency with which mothers participate in self-selected leisure occupations that promote health and well-being. Items for the HPAS were generated in the initial qualitative study; in that study, mothers identified many daily occupations outside their caregiver life role that were instrumental to their own health, enjoyable, social, and rejuvenating (Bourke-Taylor, Howie, & Law, 2010). After an examination of these mothers’ lived experience, researcher expertise on leisure occupations, and a review of existing instruments, we developed eight items for the HPAS (see Table 1).

The scale items address the purpose and types of occupation, such as whether the occupation was active and was done with supportive people or alone; the specific details of each activity were not considered important to the construct of interest. The HPAS directions define leisure occupations as activities that provide access to and participation in social, emotional, physical, and spiritual experiences that are important and desirable to mothers. The directions instruct respondents to think about the sorts of activities they have participated in for health, social interaction, and leisure over the past year and to select the corresponding frequency response for each of the eight category items (full directions are available from Helen Bourke-Taylor). Some variation may exist in how a person categorizes the activity (e.g., she might categorize piano playing as an activity in Item 4, 6, or 7), but respondents are instructed to categorize the activity within only one item.

The 7-point Likert scale response format is 1 = never, 2 = 1–3 times per year, 3 = once a month, 4 = 2–3 times a month, 5 = once per week, 6 = 2–3 times per week, and 7 = once or more every day. The response format was configured at intervals similar to those in the scoring section of the Children’s Assessment of Participation and Enjoyment (King et al., 2004). The total HPAS score is calculated by adding the scores from the eight items. Lower scores indicate less frequent participation and higher scores more frequent participation.

The research project was approved by the La Trobe University Health Science Faculty Ethics Committee. Data for the qualitative and quantitative studies were collected over a 2-year period (2006–2007). This article presents the evaluation of the HPAS resulting from the study.

To be included in the research project, the mothers had to be the primary caregiver of a school-age child with a developmental disability; resident in the state of Victoria, Australia; and able to complete the survey in English. Disability was defined as any diagnosis accepted by the Australian government department that provided weekly subsidies to families of children with developmental disabilities. Table 2 describes the characteristics of the sample participating in the evaluation study (N = 152).

Participants were recruited through disability support networks. We posted a notice describing research “to investigate factors that affect the health of Victorian mothers of a child with a disability” on Web sites and printed the notice in the newsletters of 10 not-for-profit organizations. Interested mothers contacted Bourke-Taylor, who mailed a research packet to those who expressed willingness to participate.

Data were collected for the evaluation study by means of a mailed questionnaire, and a follow-up phone call was made to verify all responses. The questionnaire consisted of a specifically designed booklet with demographic questions, five psychometrically sound previously published scales, and three original scales (including the HPAS) designed for this study. Other publications describe the other data collected and results (Bourke-Taylor, Howie, & Law, 2011; Bourke-Taylor, Howie, Law, & Pallant, 2011; Bourke-Taylor, Law, et al., 2009, 2010). Over a 10-month period, 180 mothers contacted Bourke-Taylor, and 152 completed surveys (84% response rate).

Demographic data collected included mother’s age, family income, child’s age, and mother’s education and work status (see Table 2). Mothers reported whether they had been diagnosed with a medical condition and how often they were awoken at night to attend to their child with a disability. We expected to use the responses to these questions to identify groups of mothers who would vary in their capacity to participate in leisure occupations.

The Short Form 36, Version 2 (SF–36v2; Ware, Kosinski, & Dewey, 2002) is a health-related quality of life (HRQOL) tool that measures how healthy a person feels and the influence of perceived health on what he or she is able to do in daily life. The SF–36v2 has been the primary tool used in thousands of research studies in the United States, where it was developed to evaluate HRQOL among people with various health and medical conditions (Ware, 2000, 2003). We included the SF–36v2 in the research packet to enable us to explore the relationship between subjective health status and the HPAS. The SF–36v2 was validated for use in Australia (Hawthorne, Osborne, Taylor, & Sansoni, 2007) and has eight dimension scores: physical functioning (PF), role–physical (RP), bodily pain (BP), general health (GH), vitality (VT), social functioning (SF), role–emotional (RE), and mental health (MH). Norm-based summary mental and physical health scores were determined using published norms. Use of the SF–36v2 in this research has been described elsewhere (Bourke-Taylor, Howie, & Law, 2011; Bourke-Taylor, Howie, Law, & Pallant, 2011).

The SPSS–14 statistical package (SPSS, Inc., Chicago) was used for data entry and management. Descriptive statistics (means, standard deviations, and frequencies) were generated for all scale items. Exploratory factor analysis was performed after first confirming that the data were suitable for factor analysis. Principal components analysis (PCA) was used to extract the factors followed by oblique rotation of factors using oblimin rotation (δ = 0). The number of factors to be retained was guided by three decision rules: Kaiser’s criterion (eigenvalues >1), inspection of the scree plot, and use of Horn’s (1965)  parallel analysis. Parallel analysis is among the most accurate approaches to estimating the number of components (Hubbard & Allen, 1987; Zwick & Velicer, 1986). The eigenvalues obtained from PCA are compared with those obtained from a randomly generated data set of the same size (Watkins, 2000). Only factors with eigenvalues exceeding the values obtained from the corresponding random data set were retained for further investigation. Cronbach’s α values were calculated to assess the internal reliability of the scale.

Construct validity was investigated by correlating the HPAS with the SF–36v2 dimensions and norm-based summary scores using conventional nonparametric investigations of correlation and known-groups validity (Tabachnick & Fidell, 2007). We expected that the HPAS and the SF–36v2 summary and dimension scores would correlate positively, indicating that the frequency of participation in leisure activities that involved social, physical, emotional, and spiritual fulfillment was associated with better subjective health. We did not expect the HPAS to correlate with mother’s age, child’s age, or family income.

Known-groups validity was tested by assessing the ability of the HPAS to discriminate between groups that were based on self-reported mental health condition and healthy sleep habits. We hypothesized that mothers with a mental health condition and poor sleep habits would report lower participation in leisure occupations (i.e., lower HPAS scores).

Table 3 describes the frequency of responses to the eight HPAS items. The mean total score on the HPAS was 21.67 (standard deviation = 8.57, range = 8–51). Standard tests of normality indicated that the HPAS was not normally distributed. These tests revealed a skewed distribution (0.723) with kurtosis (0.408) and a significant Kolmogorov-Smirnov statistic (p = .001; Pallant, 2010), indicating violation of the assumption of normality.

Factor analysis of the HPAS followed standard protocols (Tabachnick & Fidell, 2007). To explore the underlying factor structure of the HPAS, the eight items were subjected to PCA. The suitability of the data for factor analysis was confirmed with a highly significant Bartlett’s test of sphericity (p < .001) and a high Kaiser–Meyer–Olkin (KMO) measure of sample adequacy value (KMO = 0.79; Pallant, 2010). PCA revealed two components with eigenvalues exceeding 1 (3.21 and 1.18) explaining 54.2% of the variance; however, all eight components loaded >0.56 on the first component (see Table 3). Oblimin rotation showed that the two components moderately correlated (−.39), thus indicating a single underlying dimension that accounted for 40.1% of the variance. The results of the scree test and parallel analysis also supported a single-factor solution. The single scale had good internal consistency (Cronbach’s α = .78). The total HPAS score was calculated by adding the scores from the eight items.

Correlations between the HPAS and the SF–36v2 dimensions and component summary scores revealed moderate positive correlations with overall mental health, GH, VT, SF, RE, and MH that support the construct validity of the scale (see Table 4). Between-groups analysis confirmed that mothers who reported that they did not have a mental health condition and those who had healthy sleep habits (i.e., whose sleep was not interrupted regularly) reported more frequent participation in health-promoting leisure activities (i.e., higher HPAS scores; see Table 5). These relationships confirmed that more frequent participation in activities reflected in the items in the HPAS was associated with better subjective overall general and mental health and more healthy sleep habits.

The HPAS was developed to measure the frequency with which mothers of school-age children with developmental disabilities participate in self-selected leisure occupations that promote health and well-being. Initial evaluations presented in this study support the internal consistency, factor structure, and construct validity of the HPAS. Construct validity was supported in several ways. The HPAS correlated moderately with most health dimensions, indicating the association between participation in healthy leisure pursuits and well-being. The HPAS also discriminated between mothers with and without a diagnosed mental health condition and between those with healthy and unhealthy sleep habits.

Examination of the reported frequency with which mothers in this study participated in activities related to their subjective health status provides cause for reflection. More than half of the mothers did not participate in active recreational pursuits at all. In general, these mothers preferred inactive, quiet leisure pursuits, although more than one-third never participated in quiet leisure pursuits, either. About three-quarters of the mothers found time for themselves at least once per year and socialized with supportive others about once per month. The results of this study suggest that healthy sleep habits are somehow related to mothers’ leisure choices. Support programs for mothers with mental health conditions who are not making health-promoting decisions about their leisure time may be warranted. Findings from this research about the mental health status of mothers (Bourke-Taylor, Howie, Law, & Pallant, 2011) and participation in paid employment (Bourke-Taylor, Howie, & Law, 2011) have been described elsewhere.

The association between mothers’ physical health and participation in leisure pursuits was weak. Possible reasons may be that mothers in this sample were relatively young (mean age = 41 yr), 54% were working for pay, all were involved in busy family lives with many domestic duties, and all were involved in physical caregiver duties. Because mothers participated in many physically active occupations in their life roles, it seems reasonable to suggest that maternal physical health may be attributable to participation in occupations other than leisure (i.e., caregiving may necessitate moving equipment for child, lifting child, and assisting child in play and activities of daily living).

This study indicates that mothers’ participation in health-promoting leisure occupations is associated with their subjective health, particularly mental health. Applied to occupational therapy practice, these results suggest that mothers should be educated about this association so that lifestyle redesign that supports mental health and well-being may occur. The World Health Organization (WHO; 2009)  described several strategies to promote health among populations. One is to address individual health:

The HPAS may be used in educating mothers about ways to support better health and in enabling them to make choices conducive to better health. Occupational therapists can facilitate the improved participation and health of mothers by disseminating the findings of this study and contributing to future data collection. Recent research demonstrated the relationship between health-promoting behaviors and some aspects of health-related quality of life among mothers with multiple sclerosis; Tyszka and Farber (2010)  used a lengthy tool (52 items) to measure health-promoting behaviors and demonstrated the need for occupational therapists to use health promotion strategies to support the life role participation of this group of mothers. The HPAS could be used similarly in future research and clinical work.

Some limitations in this study may be addressed in future research. Different sampling methods that provide population-based data and additional verification of medical history are required. Future research might also investigate the relationships between satisfaction and both health status and participation in leisure occupations as measured by the HPAS. Future research should also investigate the relative contribution of participation in health-promoting occupations and other life factors to mothers’ mental health status. Evaluation of the sensitivity of the HPAS to detect changes in health-promoting behaviors and subjective health over time will indicate whether the HPAS is a valid outcome measure. Application to other types of occupational therapy clients requires further validation studies.

The HPAS provides practicing occupational therapists with a brief tool to

• Address the mother’s capacity to participate in health-promoting leisure pursuits,

• Educate mothers about their own health needs,

• Assist mothers with healthy lifestyle redesign of leisure pursuits, and

• Include client-centered goal setting in relation to HPAS items.

Use of the HPAS reminds occupational therapists that caregiver health and well-being are essential parts of family-centered practice in all work with children with disabilities.

The HPAS provides clinical occupational therapists and researchers with a tool that is brief, psychometrically sound, user friendly, and clinically relevant to measure the frequency with which mothers participate in self-selected leisure occupations that are associated with health. Initial evaluation of the scale presented in this article indicates that further research and development of the tool are warranted.