Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Oct 1.
Published in final edited form as: Clin Gerontol. 2019 Jul 29;45(5):1214–1225. doi: 10.1080/07317115.2019.1645781

Programmed Activity Attendance in Assisted Living: An Application of the Theory of Planned Behavior with Additional Health Factors

Evan Plys a,b, Sara H Qualls a,c
PMCID: PMC6987002  NIHMSID: NIHMS1063933  PMID: 31354090

Abstract

Objectives:

Programmed activities contribute to the quality of life of assisted living (AL) residents, but attendance rates are often quite low. Using the theory of planned behavior (TPB), this study explores relationships among social cognitive variables, multidimensional health factors, and programmed activity attendance in AL.

Methods:

In a cross-sectional survey, 185 AL residents (Mage = 83.64, SD = 9.64) self-reported physical and mental health; TPB constructs (i.e., attitudes, subjective norms, behavioral control, and behavioral intention); and the percentage of programmed activities attended in the previous week.

Results:

Structural modeling showed that TPB was a good fit for the data (χ2/DF = 1.67; CFI = .97; TLI = .96; RMSEA = .06), explaining 82% of variance in behavioral intention and 44% of variance in activity attendance. After adding health factors to the model, only mental health yielded a significant indirect effect on activity attendance.

Conclusion:

Results provide preliminary support for the viability of TPB as a framework to explain resident activity attendance in AL. Thus, the decision to attend programmed activities in AL may represent a social cognitive process, influenced by mental health.

Clinical Implications:

Interventions may consider targeting TPB constructs and mental health to increase activity behaviors amongst AL residents.

Keywords: Theory of planned behavior, activity, recreation, leisure, assisted living, structural equation modeling

Assisted living (AL) communities advertise a person-centered model of care that emphasizes the importance of resident engagement with life after relocation (Center for Excellence in Assisted Living [CEAL], 2010; Kane & Wilson, 2007). Activity programming is a service that ALs offer to help promote active engagement within the community. In AL, programmed activities are an important component of everyday life and provide residents with opportunities for structured physical, cognitive, and social stimulation (Hanson et al., 2014). National data found that 100% of ALs offer programmed recreational activities (Khatutsky et al., 2016), which vary by type and difficulty (Plys & Qualls, 2016). Theoretically, offering a range of recreational activities within a close proximity should promote participation amongst residents (Miller & Buys, 2007); however, in practice, resident participation rates in recreational activities are quite low after relocation to AL (Mihalko & Wickley, 2003; Resnick, Galik, Gruber-Baldini, & Zimmerman, 2010). Poor attendance for programmed activities is an important clinical issue to consider in AL as increasing recreational engagement may help promote positive physical, cognitive, and psychosocial outcomes for residents, as well as positively contribute to the culture of the community (International Council on Active Aging [ICAA], 2009; O’Sullivan, 2005).

Few studies target resident factors that relate to programmed activity attendance in AL. A small body of literature suggests that attendance rates are influenced by personal preference, meaning, and interest (Decker, Cary, & Krautscheid, 2006; Lewinson, Robinson-Dooley, & Grant, 2012; Sefcik & Abbott, 2014; Williams & Warren, 2009); perceived behavioral outcomes (Crenshaw, Gillian, Kidd, Olivo, & Schell, 2001; Hanson et al., 2014); and social-relational factors (Cummings, 2002; Cummings & Cockerham, 2004; Mitchell & Kemp, 2000; Park, Zimmerman, Kinslow, Shin, & Roff, 2012). However, the bulk of activities research focuses on the relationship between resident multidimensional health factors (e.g., physical and mental health) and activity attendance, with better health relating to higher attendance rates (Crenshaw et al., 2001; Hanson et al., 2014; Park et al., 2012; Watson et al., 2006; Zimmerman et al., 2003). A recent review reported that the existing research on activities in AL lacks studies informed by theoretical frameworks (Plys, 2017). The current study will address this limitation in the literature by using cross-sectional survey data to test the viability of a social cognitive theoretical model for examining resident factors that influence programmed activity attendance rates in AL. Specifically, we will use the theory of planned behavior (TPB) as an explanatory framework for programmed activity attendance in AL, and will explore the role of physical and mental health as additional paths within this theoretical model. By using TPB as a guiding framework, this study offers a first look at an established behavioral theory that may offer insight into factors that influence resident attendance for programmed activities in AL.

Theoretical framework

TPB is a social cognitive theory that suggests behavioral intention, the motivational factor for engaging in a behavior based on the information and personal beliefs, is the underlying psychological determinant of human behavior (Ajzen, 1991). The model proposes that three inter-related variables (i.e., attitudes, subjective norms, and perceived behavioral control) influence behavioral intention. Attitudes refer to personal appraisals of a behavior, and are measured in two domains: instrumental (i.e., whether the behavior achieves a desired outcome) and affective (i.e., how it feels to engage in the behavior). Subjective norms refer to individual perceptions of social pressures to engage, or not engage, in a behavior. Perceived behavioral control refers to the subjective ease or difficulty of engaging in a behavior, which includes whether an individual believes they have the necessary internal and external resources to successfully participate in the targeted behavior (Francis et al., 2004). Meta-analyses suggest that attitudes, norms, and behavioral control have strong effects on behavioral intention, and that behavioral intention has a strong effect on activity behaviors (Hagger, Chatzisarantis, & Biddle, 2002).

Previous research also suggests that TPB provides a valid framework for explaining activity behaviors in older adulthood (see Motalebi, Iranagh, Abdollahi, & Lim, 2014). The majority of TPB studies with older adults target physical activity in community-living samples (e.g., Arbour-Nicitopoulos, Ginis, & Wilson, 2010; Dixon, Johnston, Rowley, & Pollard, 2008). To the knowledge of these authors, no study has applied TPB to resident activity behaviors in AL. It is important to evaluate the generalizability of behavioral theories to AL, as this setting has a unique socio-physical environment and population that may influence resident behaviors. In the context of unique settings and-or targeted behaviors, previous scholars tested additional setting-, population-, and-or behavior-specific variables within the original TPB model (Ajzen & Fishbein, 2005). Given the emphasis on multidimensional health in the activities literature, the current study will investigate the role of physical and mental health within the TPB model.

Health, TPB, and AL

Previous TPB research suggests that health factors can have direct effects on behavior, conceptualized as actual behavioral control (see Ajzen & Fishbein, 2005). Actual behavioral control refers to the ability for an individual to engage in a behavior, representing objective personal (e.g., physical ability) and-or environmental (e.g., opportunities for the behavior) barriers. In the current study, the direct path of physical and mental health on activity attendance may provide insight into person-environment fit. For example, a resident may intend to go to a programmed activity, but be unable to do so because of acute physical pain limiting their ability to ambulate to and-or participate in an activity offering, or a mood disturbance that increases forgetfulness and-or acute distress leading to social withdrawal. A study with community-living older adults tested a model of TPB with actual behavioral control and found that functional ability explained an additional 11% of variance in activity behaviors beyond TPB variables (Gretebeck et al., 2007). Therefore, in certain contexts, health may provide additional insight into activity behaviors beyond intention.

Another possible role for health in the TPB model is as background factors influencing attitudes and beliefs about activity attendance. In other words, residents’ physical and-or mental health may shape how they perceive activity participation, and, thus, influence intention and attendance for activities. Previous research suggests that engaging in functionally appropriate activities increases self-efficacy and positive attitudes toward the activity (Nakamura & Csikszentmihalyi, 2014). Physical and mental health may also relate to subjective norms, as previous studies suggest that health factors influence social opportunities, relationships, and norms in AL (Doyle, de Medeiros, & Saunders, 2012). Gretebeck et al. (2007) provide support for health as a background factor in the TPB model by reporting that functional ability had significant effects on attitudes, norms, and behavioral control for activity behaviors in a sample of community-living older adults. Taken together, previous research suggests that health may have a role in influencing social cognitive TPB variables; indirectly relating to activity behaviors.

Purpose

The purpose of this study is to investigate the viability of TPB as a theoretical framework to offer preliminary insight into residents’ decision-making process to attend programmed activities in AL. Using structural modeling, we will investigate relationships among social cognitive variables and activity attendance consistent with TPB. In addition, we will explore the role of physical and mental health within the TPB model. This study adds to the activities literature by being the first to test a social cognitive theory with AL programmed activity attendance as the dependent variable. This study also contributes to the TPB literature by offering insight regarding the applicability this theory to a previously unstudied setting, population, and behavior, as well as exploring the role of multidimensional health factors within the TPB model.

Method

Participant characteristics

The current sample included 185 AL residents. Participant characteristics are presented in Table 1. Most demographics were consistent with national data (see Khatutsky et al., 2016). The percentage of participants with more than a high school education (67%) was higher than national data (i.e., 41%; Khatutsky et al., 2016), as was the percentage of residents (29%) who used Medicaid to finance services (i.e., compared to 19%; Caffrey et al., 2012). Data were collected at a purposive sample of 21 ALs, ranging in size from 16 to 104 (M = 54.19, SD = 25.07). The majority of ALs were large (i.e., 25–100 residents; n= 15, 71%), owned by a chain company (n= 11, 53%), and cost $3,000 or more per month (n = 16, 76%). Nine (43%) ALs were all private-pay and four (19%) had mostly (i.e., 75%+) Medicaid beds. Almost all ALs had a separate space for programmed activities (n = 20, 95%) and 10 (48%) had separate exercise facilities.

Table 1.

Participant characteristics (n = 185).

N % Mean SD
Age 185 83.64   9.64
Time in AL (months) 185 20.74 19.88
Female 132 71%
White 168 91%
More than High school education 124 67%
Widowed 111 60%
Medicaid 54 29%
Previous residence
Senior independent living 35 19%
Private community home 128 69%
Different AL 22 12%
Open in a new tab

Measures

Health

Subscales from the PROMIS Short-Form Global Health Scale assessed physical and mental health (Hays, Bjorner, Revicki, Spritzer, & Cella, 2009). The physical health subscale contains four items related to global physical health, functional ability, pain, and fatigue. The mental health subscale consists of four items related to global mental health, impact on social activities, cognition, and mood. Scores on each subscale ranged from 4 to 20, with higher scores indicating greater health. Internal consistency with the sample was acceptable for both the physical (Cronbach’s α = .71) and mental (Cronbach’s α = .73) health scale.

TPB scales

The current study created TPB scales based on recommendations for health-related TPB questionnaires, a literature review of activity behaviors in AL, and previous TPB studies addressing activity behaviors with older adults (e.g., Francis et al., 2004; Gretebeck et al., 2007; Rhodes, Courneya, Blanchard, & Plotnikoff, 2007). Scales were piloted with a sample of 11 AL residents in one AL, and revised accordingly.

Attitudes.

Attitudes were measured with four affective items assessing the extent that activity attendance was enjoyable, fulfilling, interesting, or meaningful, and two instrumental items assessing the extent that attending activities would help stay busy or improve health. Participants responded to each item on a 7-point scale, 1 = Strongly disagree, 7 = Strongly agree. Scores ranged from 6 to 42, with higher scores indicating positive attitudes toward activity attendance. Internal consistency was good with the current sample (Cronbach’s α = .89).

Subjective norms.

Two items (i.e., “Most people who are important to me want me to attend activities,” and “I feel pressure to attend activities”), asked on a 7-point scale, 1 = Strongly disagree, 7 = Strongly agree, assessed subjective norms. Scores ranged from 2 to 14, with higher scores indicating greater perceived norms for activity attendance. Internal consistency was questionable with the current sample (Cronbach’s α = .64).

Perceived behavioral control.

A four-item scale measured perceived behavioral control (e.g., “I am confident that I could attend activities if I wanted to”). Participants responded to items on a 7-point scale, 1 = Strongly disagree, 7 = Strongly agree. Summed scores ranged from 4 to 28, with higher scores indicating greater perceived behavioral control for activity attendance. The current sample evidenced good internal consistency (Cronbach’s α = .81).

Behavioral intention.

Behavioral intention was measured with three items (e.g., “It is my intention to go to activities here” and “I am motivated to go to activities”) on a 7-point scale, 1 = Strongly disagree,7 = Strongly agree. Sum scores ranged from 3 to 21, with higher scores indicating greater intention for activity attendance. The current sample evidenced excellent internal consistency (Cronbach’s α = .92).

Activity attendance

Participants self-reported activity attendance by indicating each programmed activity (1 = Yes, 0 = No) they attended in the past week. The index of programmed activity offerings was taken directly from activity calendar at each participant’s AL.

Social desirability

The Five-item Socially Desirable Response Set assessed social desirability (Hays, Hayashi, & Stewart, 1989). Participants respond to five items on a 5-point scale, 1 = Definitely true,5= Definitely false. One item was omitted for poor contributions to internal consistency (i.e., “No matter who I’m listening to, I am always a good listener”), likely due to high rates of hearing difficulties observed with the current sample. Participants evidenced acceptable internal consistency for the revised four-item scale (Cronbach’s α = .70). Total scores ranged from 4 to 20; higher scores indicate more socially desirable responses.

Procedures

The institutional review board at the affiliated university approved all study procedures (IRB Protocol #16–237). Exclusion criteria for ALs were: (a) specialized care community for persons with serious mental illness or dementia; and (b) 10 or fewer residents. We imposed these criteria to maximize recruitment efficiency, and because activities may take on a different role in specialty care settings (Hyde, Perez, & Forester, 2007). The researchers generated a list of eligible ALs through existing relationships, public listings, and recommendations from AL staff or members of relevant long-term care organizations. We contacted executive directors or other relevant administrative staff (e.g., community relations) at 51 ALs across three metropolitan areas in the state of Colorado, yielding a response rate of 39%; 17 communities agreed to participate in the study. A regional administrator at an operating company referred four additional ALs. Overall, 21 ALs appeared in the study.

Frontline staff at each AL identified residents who were interested and appropriate for the current study. Recruitment efforts also involved announcements at meals and/or well-attended activities, and snowball sampling. The following inclusion criteria were used for residents: (a) age 50 or older; (b) able to speak and understand English; (c) own decision-maker; and (d) score a five or above on the Memory Impairment Screen (MIS; see Buschke et al., 1999). Three participants did not meet criterion a and 30 did not meet criterion d; an additional 17 participants were excluded due to the use of survey administration strategies that may have influenced results (e.g., take-home surveys). Overall, 185 resident participants were included in the sample.

A researcher met with each resident volunteer individually. After obtaining consent, the interviewer administered a brief cognitive screen to determine eligibility. Participants who met eligibility continued with a full interview. Interviews were conducted one-to-one in a quiet area in the AL, in which participants were read survey questions aloud and indicated answers on large-print laminated response sheets. On average, interviews lasted between 30 and 50 min. Interviews followed a uniform order with activity attendance assessed at the start of the interview, followed by health, and, then, TPB variables (i.e., attitudes, norms, behavioral control, and, then, intention). At each AL, staff also completed surveys describing the community and activity program.

Data preparation & statistical analyses

All statistical analyses were conducted using SPSS v.24 and AMOS v.24. The data met all statistical assumptions and confirmatory factor analyses supported paths and factor structures for health (Factor Loadings > .46) and TPB scales (Factor Loadings >.48). Structural equation modeling (SEM) with a maximum likelihood estimation-tested models. Given the small sample size, we used liberal cutoffs to test model fit: relative chi-square (χ2/DF) ≤2, comparative fit index (CFI) ≥ .90, Tucker–Lewis index (TLI) ≥ .90, and root mean square error of approximation (RMSEA) ≤ .08 (Garson, 2015; Kline, 2016; Weston & Gore, 2006). We also reported chi-square (χ2) and Akaike information criterion (AIC; lower scores indicate better model fit) to compare models.

Results

Descriptive analyses

The sampled ALs offered between 11 and 65 activities per week (M = 37.06, SD = 11.74; Median = 37), and typically offered a full day of programming on week-days (M = 7.71 h, SD = 1.68). We identified 50 different types of programmed activities offered across the 21 ALs; each community offered religious services and exercise, and the majority offered BINGO (91%) and socials (86%). The sampled ALs offered between six and 30 (M = 18.14, SD = 4.94; Median = 18) different types of activities each week. We conducted a median split on total activity offerings and the number of different activities offered per week. An independent samples t-test (t(129.26) = −2.14, p = .034) revealed that residents in ALs that offered more activities reported significantly higher attendance rates (M = 6.63, SD = 6.15) than residents in ALs that offered fewer activities (M = 7.91, SD = 9.08). Residents in ALs with high and low activity offerings did not significantly differ in the percentage of activities attended (t(183) = .73, p = .465). Residents in ALs with high and low variability in types of activity offerings evidenced no significant differences in the amount (t(183) = .31, p = .758) or percentage (t(183) = 1.95, p = .052) of activities attended. Therefore, to control for variability in the number of activity offerings, we used the percentage of activities attended, rather than a raw frequency count, in analyses.

Participants attended between 0 and 41 activities per week (M = 7.71, SD = 7.62) or 0 to 93% of the offered activities (M = 21.51%, SD = 16.22). Twenty-eight residents (15%) did not attend any activities, which is slightly lower than previous estimates with national data (i.e., 20%; Khatutsky et al., 2016). Table 2 presents descriptive statistics and correlations amongst study variables. No study variable significantly related to social desirability at a significance level of .05.

Table 2.

Descriptive statistics and bivariate correlations among study variables.

M SD 1 2 3 4 5 6
1. Physical Health 12.91   2.93
2. Mental Health 13.50   3.10  .47***
3. Attitudes 29.54   7.21  .15*  .22**
4. Subjective Norms   9.91   2.37  < .01  −.08 .37***
5. Behavioral Control 21.84   4.70  .38***    .35*** .32*** .13
6. Behavioral Intention 13.77   4.86  .17*    .17* .83*** .42*** .40***
7. Activity Attendance (%) 21.51 20.02  .17*    .18* .55*** .28*** .34*** .66***
Open in a new tab
*

p < .05,

**

p < .01,

***

p < .001.

Structural models

Table 3 presents fit indices for each structural model. The first model evidenced good fit (χ2(98, N = 185)= 163.70, p < .001; χ2/DF = 1.67; CFI = .97; TLI = .96; RMSEA = .06; AIC = 271.70), see Figure 1. Behavioral control did not yield a significant direct effect on activity attendance (β = .11, p= .130), and, thus, this path was omitted from the model. Intention (β = .67, p< .001) had a significant direct effect on activity attendance. Attitudes (β = .78, p< .001), norms (β = , p= .015), and behavioral control (β = .14, p= .006) all significantly related to behavioral intention, accounting for 82% of the variance. Attitudes (β = .52, p= .015) and control (β = .09, p= .044) had a significant indirect effect on activity attendance; norms (β = .08, p= .088) did not. Overall, the traditional TPB model accounted for 44% of the variance in activity attendance.

Table 3.

Structural model fit indices.

χ 2 χ2/DF CFI TLI RMSEA AIC
Model 1 163.70 1.67 .97 .96 .06 271.70
Model 2 464.10 1.93 .91 .89 .07 630.10
Model 3 414.27 1.71 .93 .92 .06 578.27
Open in a new tab

Bolded values represent the best fit of the three models.

Figure 1.

Figure 1.

Open in a new tab

Model 1: Traditional TPB model.

The second model evidenced borderline acceptable fit (χ2(241, N = 185) = 464.10, p < .001; χ2/DF = 1.93; CFI = .91; TLI = .89; RMSEA = .07; AIC = 630.10), see Figure 2. Neither mental health (β = .05, p= .553) nor physical health (β = .05, p= .579) had a direct effect on activity attendance beyond behavioral intention. After removing non-significant paths, the third model evidenced acceptable fit (χ2(242, N = 185) = 414.27, p < .001; χ2/DF = 1.71; CFI = .93; TLI = .92; RMSEA = .06; AIC = 578.27), see Figure 3. Mental health had a significant effect on attitudes (β = .34, p< .001) and behavioral control (β = .36, p= .002), but not norms. Physical health had a significant effect on behavioral control (β = .28, p= .011), but not attitudes or norms. Mental health accounted for 12% of the variance in attitudes, and physical and mental health accounted for 32% of the variance in control. Mental health had a significant indirect effect on intention (β = .31, p= .009) and activity attendance (β = .21, p= .012). Physical health did not have a significant indirect effect on either intention (β = .04, p= .054) or activity attendance (β = .02, p= .061).

Figure 2.

Figure 2.

Open in a new tab

Model 2: TPB with health directly relating to activity attendance.

Figure 3.

Figure 3.

Open in a new tab

Model 3: TPB with health indirectly relating to activity attendance.

Discussion

The traditional TPB model evidenced the best fit for the data, accounting for 82% of the variance in behavioral intention and 44% of the variance in programmed activity attendance. The observed effects on intention and activity attendance are comparable to previous studies investigating walking behaviors with a sample of pre-surgical orthopedic patients (i.e., Dixon et al., 2008), and slightly higher than previous research investigating physical activity in community-living older adults (i.e., Arbour-Nicitopoulos et al., 2010; Lee, 2016). These findings support TPB as a theoretical framework that is viable for further study of programmed activity attendance in AL. Distinct paths within the model provide additional insight into the applicability of TPB to AL activity behaviors.

In the current sample, behavioral intention had a strong direct effect on activity attendance, but behavioral control did not. When intention has an effect on a behavior and control does not, the behavior is thought to be under a high degree of volitional control (Ajzen, 1991; Ajzen & Fishbein, 2005). Applied to this study, attending programmed activities in AL may represent a planned and conscious behavioral decision; although, longitudinal data are required to investigate the predictive relationship between intention and behavior. Previous research also suggests that behavioral control is less likely to have a direct effect on behaviors that are readily available in the immediate environment (Ajzen & Driver, 1992). Therefore, it is possible that programmed activities were both available and accessible to the majority of residents in the current sample. Taken together, findings suggest that AL residents’ decision to attend programmed activities represents a volitional process, which supports the use of social cognitive frameworks in future research.

Consistent with previous TPB research (e.g., McEachan, Conner, Taylor, & Lawton, 2011), attitudes had a strong direct effect on behavioral intention and a significant indirect effect on activity attendance. Therefore, participants in this study reported stronger behavioral intention, and, thus, evidenced higher rates of activity attendance, if they also believed that attending programmed activities would make them feel good and-or produce a desirable outcome. This finding highlights the importance of preference assessments to ensure that activity offerings are enjoyable, and align with residents’ personal interests and recreational goals. Even though many ALs already assess resident activity preferences (Dobbs et al., 2005), residents still report high levels of dissatisfaction with activity offerings (Abrahamson, Bradley, Morgan, Fulton, & Ibrahimou, 2013; Gregory, Gesell, & Widmer, 2007). Future research may choose to investigate methods to increase the clinical utility of activity preference assessments, as results from this study suggest attitudes may be a key contributor to activity motivation and participation in AL.

Neither physical nor mental health had a significant direct effect on activity attendance beyond behavioral intention in the current sample. These results are inconsistent with previous research that found that health factors added to the TPB model explained additional variance in activity behaviors (Dixon et al., 2008; Gretebeck et al., 2007). It is possible that the current sample did not possess health limitations at the level necessary to prevent programmed activity attendance, which is discussed further in the limitations section. Alternatively, health may not directly relate to activity attendance after controlling for social cognitive factors. The AL environment is designed to help residents remain active and engaged despite health limitations, which may minimize the direct effect of health on resident activity behaviors. For example, health limitations may only impact activity behaviors if the resident also has low self-efficacy for attendance. Future research is needed to investigate relationships among health factors and activity behaviors in AL, as well as variables that may mediate this relationship.

The significant effect of both physical and mental health on behavioral control was expected as control assesses residents’ perception of their ability to attend activities. Given that physical and mental health had similar effects on control, residents in the current study may have evaluated their health in a multidimensional manner when conceptualizing behavioral control (i.e., physical, cognitive, and emotional limitations). Therefore, future research should note the importance of assessing multiple dimensions of health, not just physical ability, in studies targeting activity behaviors with frail older adults. In addition, in this study, only mental health significantly related to attitudes toward programmed activity attendance. This finding is likely related to the impact of mood on a belief system. TPB scholars suggest that individuals with higher mood states are more likely to perceive positive aspects of a behavior (Ajzen, 2011). Therefore, mentally healthy residents may be more likely to perceive attending activities as an enjoyable and productive behavior. Taken together, multidimensional health factors had a relatively small and targeted effect on the current sample of AL residents’ views toward programmed activity attendance.

Implications

The current study provides support for the applicability of TPB to AL programmed activity attendance, as evidenced by significant relationships among social cognitive variables and attendance rates consistent with the TPB model. These findings add to the AL activities literature by providing initial support for an established theoretical framework that may be able to explain residents’ decisional process, motivation, and attendance for programmed activities, an important behavior to support well-being after relocation. In this study, physical health evidenced minimal effects on both behavioral intention and activity attendance. Therefore, future models of activity behaviors in AL may choose to emphasize social cognitive and psychological variables that appreciate the individual decision-making process involved in human behavior, rather than relying solely on physical health and ability. It should be noted that while this study offers preliminary support for the utility of TPB as a framework for programmed activity attendance in AL, future research using larger longitudinal data is needed to provide evidence for the ability of TPB to predict activity attendance.

One of the benefits of TPB as a behavioral framework is that it can be used to develop clinical interventions targeting behavior change (French, Darker, Eves, & Sniehotta, 2013). Previous research suggests that AL residents spend the majority of their discretionary time in their rooms engaging in sedentary activities (Pruchno & Rose, 2002). Therefore, with additional empirical support, clinicians in AL may be interested in using the framework tested in the current study to plan interventions targeting resident activity levels. A meta-analysis supports that TPB-based interventions are efficacious for modifying beliefs, attitudes, and intention for various health-related behaviors (see Steinmetz, Knappstein, Ajzen, Schmidt, & Kabst, 2016). TPB-based interventions typically use multiple therapeutic techniques including psychoeducation, motivational interviewing, skills building, and behavioral strategies (e.g., goal setting and problem-solving; French et al., 2013). Our findings suggest that mental health may also influence activity participation through TPB variables, signifying the need for interventions to target mood (e.g., cognitive restructuring or pleasant activity scheduling) alongside activity beliefs, attitudes, and behavioral intention. Although, it should be noted that interventions targeting behavioral activation can also be an effective treatment for mental health concerns in older adults in residential care settings (Meeks, Van Haitsma, Schoenbachler, & Looney, 2014). Overall, future research may choose to develop and test TPB-based interventions targeting activity promotion in AL.

Limitations

Purposive sampling, small sample size, and exclusion of residents with significant cognitive impairment limit the generalizability of results from the current study. While we attempted to minimize bias by using a liberal cognitive screen for exclusion criteria (i.e., MIS; Buschke et al., 1999), findings are still limited to the approximately 80% of AL residents without severe cognitive impairment (Zimmerman, Sloane, & Reed, 2014). In addition, geographic homogeneity, purposive sampling, and exclusion of small ALs may also limit the generalizability of results; although, 90% of AL residents reside in communities consistent with the sizes included in this study (Park-Lee et al., 2011). Lastly, we controlled for variability in activity offerings by including percent of activities attended in analyses, but variability in the quality of activity programs, appeal and features of activity offerings, and environmental design likely had an impact on attendance rates in the current study. Given the breadth of environmental factors that may relate to activity attendance in AL (see Plys, 2017), it is difficult to control for potential confound without over-fitting statistical models. Therefore, developing a brief and comprehensive assessment of relevant organizational and environmental factors for activities and-or behavioral research is an important area of future investigation in the AL literature.

Cross-sectional design also limits the ability for the current study to test TPB as a predictive model. While a significant number of TPB studies employ cross-sectional design (e.g., Arbour-Nicitopoulos et al., 2010; Gretebeck et al., 2007; Lee, 2016), this theory is best utilized in a predictive model with behavior assessed at a follow-up time point (Ajzen & Fishbein, 2005). In AL, longitudinal research has unique challenges due to high risk of acute illness, hospitalization, or transfers, as well as the time necessary to collect data with AL residents. Investigating efficient and reliable longitudinal data collection methods is needed for the future of AL research.

Measurement issues may have also limited findings in this study. Specifically, the subjective norms scale evidenced questionable internal consistency, and, thus, results with this measure should be interpreted with caution. In addition, the current study utilized a retrospective self-report index to measure activity attendance. Even though retrospective self-report is often used in gold-standard activity measures (Godin, 2011; Lyons, Li, Tosteson, Meehan, & Ahles, 2010) and is the most common method of assessment for activity behaviors in AL (see Plys, 2017), it is also susceptible to bias. In our study, we took steps to minimize distortion bias (i.e., assessing social desirability), as well as recall bias (i.e., excluding participants with significantly impaired memory and cueing participants with a list of activities offered in the past week). Yet, retrospective self-report is still an imperfect estimate of human behavior and should be considered a limitation of this study.

Conclusion

Key findings provide preliminary support for the viability of TPB as a framework to explain programmed activity attendance in AL. However, additional longitudinal research is needed to test the predictive ability of this theoretical model with a larger, representative AL sample. Additional research may also choose to apply the TPB model to single activity outcomes (e.g., attending BINGO as the dependent variable) to provide information regarding attendance for specific activity offerings. In the current study, mental, but not physical health, had a significant indirect effect on programmed activity attendance, acting through social cognitive variables. Therefore, the decision to attend programmed activities may best represent a volitional process influenced by psychological factors, and may be minimally influenced by residents’ physical health and ability. As a result, future research investigating activity behaviors with AL residents may benefit from moving from prioritizing physical health and ability, toward appreciating person-centered psychological variables (e.g., preference, attitudes, and self-efficacy), for which TPB may offer an acceptable theoretical framework.

Clinical applications.

  • The decision to attend AL activities may represent a social cognitive process, strongly influenced by attitudes and preference.

  • The mental health of AL residents may shape attitudes toward activity programming, thus, affecting intention and attendance for programmed activities.

  • TPB-based clinical interventions with a secondary emphasis on elevating mood may help to increase activity engagement amongst AL residents.

Acknowledgments

Funding

This work was supported in part by a training grant from the National Institute on Aging, award number T32AG044296.

Footnotes

Disclosure statement

No potential conflict of interest was reported by the authors.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author, [EP], upon reasonable request.

Ethical Principles

The authors affirm having followed professional ethical guidelines in preparing this work. These guidelines include obtaining informed consent from human participants, maintaining ethical treatment and respect for the rights of human or animal participants, and ensuring the privacy of participants and their data, such as ensuring that individual participants cannot be identified in reported results or from publicly available original or archival data.

References

  1. Abrahamson K, Bradley DB, Morgan KH, Fulton BR, & Ibrahimou B. (2013). Influence of satisfaction with services on assisted living resident satisfaction. Journal of Housing for the Elderly, 27, 177–190. doi: 10.1080/02763893.2012.754821 [DOI] [Google Scholar]
  2. Ajzen I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. doi: 10.1016/0749-5978(91)90020-T [DOI] [Google Scholar]
  3. Ajzen I. (2011). The theory of planned behaviour: Reactions and reflections. Psychology & Health, 9, 1113–1127. doi: 10.1080/08870446.2011.613995 [DOI] [PubMed] [Google Scholar]
  4. Ajzen I, & Driver BL (1992). Application of the theory of planned behavior to leisure choice. Journal of Leisure Research, 24, 207–224. doi: 10.1080/00222216.1992.11969889 [DOI] [Google Scholar]
  5. Ajzen I, & Fishbein M. (2005). The influence of attitudes on behavior. In Albarracín D, Johnson BT, & Zanna MP (Eds.), The handbook of attitudes (pp. 173–221). Mahwah, NJ: Lawrence Erlbaum Associates. [Google Scholar]
  6. Arbour-Nicitopoulos KP, Ginis KAM, & Wilson PM (2010). Examining the individual and perceived neighborhood associations of leisure-time physical activity in persons with spinal cord injury. Annals of Behavioral Medicine, 39, 192–197. doi: 10.1007/s12160-009-9149-9 [DOI] [PubMed] [Google Scholar]
  7. Buschke H, Kuslansky G, Katz M, Stewart WF, Sliwinski MJ, Eckholdt HM, & Lipton RB (1999). Screening for dementia with the memory impairment screen. Neurology, 52, 231. doi: 10.1212/WNL.52.2.231 [DOI] [PubMed] [Google Scholar]
  8. Caffrey C, Sengupta M, Park-Lee E, Moss A, Rosenoff E, & Harris-Kojetin L. (2012). Residents living in residential care facilities: United States, 2010 (NCHS Data Brief No. 91). Hyattsville, MD: National Center for Health Statistics. doi: 10.1094/PDIS-11-11-0999-PDN [DOI] [PubMed] [Google Scholar]
  9. Center for Excellence in Assisted Living. (2010). Person-centered care in assisted living: An informational guide. Retrieved from www.theceal.org/…/k2/…/197_9dbab55605eb0a9d547b29d1fd84fb5e [Google Scholar]
  10. Crenshaw W, Gillian ML, Kidd N, Olivo J,& Schell BAB (2001). Assisted living residents’ perspectives of their occupational performance concerns. Activities, Adaptation, & Aging, 26, 41–55. doi: 10.1300/J016v26n01_03 [DOI] [Google Scholar]
  11. Cummings SM (2002). Predictors of psychological well-being among assisted-living residents. Health & Social Work, 27, 293–302. doi: 10.1093/hsw/27.4.293 [DOI] [PubMed] [Google Scholar]
  12. Cummings SM, & Cockerham C. (2004). Depression and life satisfaction in assisted living residents: Impact of health and social support. Clinical Gerontologist, 27, 25–42. doi: 10.1300/J018v27n01_04 [DOI] [Google Scholar]
  13. Decker S, Cary P, & Krautscheid L. (2006). From the streets to assisted living: Perceptions of a vulnerable population. Journal of Psychosocial Nursing and Mental Health Services, 44, 18–27. PMID:16789590. [DOI] [PubMed] [Google Scholar]
  14. Dixon D, Johnston M, Rowley D, & Pollard B. (2008). Using the ICF and psychological models of behavior to predict mobility limitations. Rehabilitation Psychology, 53, 191–200. doi: 10.1037/0090-5550.53.2.191 [DOI] [Google Scholar]
  15. Dobbs D, Munn J, Zimmerman S, Boustani M, Williams CS, Sloane PD, & Reed PS (2005). Characteristics associated with lower activity involvement in long-term care residents with dementia. The Gerontologist, 45, 81–86. doi: 10.1093/geront/45.suppl_1.81 [DOI] [PubMed] [Google Scholar]
  16. Doyle PJ, de Medeiros K, & Saunders PA (2012). Nested social groups within the social environment of a dementia care assisted living setting. Dementia, 11, 383–399. doi: 10.1177/1471301211421188 [DOI] [Google Scholar]
  17. Francis JJ, Eccles MP, Johnston M, Walker A, Grimshaw J, Foy R, … Bonetti D. (2004). Constructing questionnaires based on the theory of planned behaviour. Newcastle, UK: Centre for Health Services Research. [Google Scholar]
  18. French DP, Darker CD, Eves FF, & Sniehotta FF (2013). The systematic development of a brief intervention to increase walking in the general public using an “extended” theory of planned behavior. Journal of Physical Activity and Health, 10, 940–948. doi: 10.1123/jpah.10.7.940 [DOI] [PubMed] [Google Scholar]
  19. Garson GD (2015). Structural equation modeling. Asheboro, NC: Statistical Associates Publishing. Retrieved from http://www.statisticalassociates.com/sem_p.pdf [Google Scholar]
  20. Godin G. (2011). The Godin-Shephard leisure-time physical activity questionnaire. The Health & Fitness Journal of Canada, 4(1), 18–22. doi: 10.14288/hfjc.v4i1.82 [DOI] [Google Scholar]
  21. Gregory N, Gesell SB, & Widmer T. (2007). Improving assisted living care. Journal of Nursing Care Quality, 22, 50–58. PMID: 17149086. [DOI] [PubMed] [Google Scholar]
  22. Gretebeck KA, Black DR, Blue CL, Glickman LT, Huston SA, & Gretebeck RJ (2007). Physical activity and function in older adults: Theory of planned behavior. American Journal of Health Behavior, 31, 203–214. doi: 10.5993/AJHB.31.2.9 [DOI] [PubMed] [Google Scholar]
  23. Hagger MS, Chatzisarantis NL, & Biddle SJ (2002). A meta-analytic review of the theories of reasoned action and planned behavior in physical activity: Predictive validity and the contribution of additional variables. Journal of Sport & Exercise Psychology, 24, 3–32. doi: 10.1123/jsep.24.1.3 [DOI] [Google Scholar]
  24. Hanson HM, Hoppmann CA, Condon K, Davis J, Feldman F, Friesen M, … Ashe MC (2014). Characterizing social and recreational programming in assisted living. Canadian Journal on Aging, 33, 285–295. doi: 10.1017/S0714980814000178 [DOI] [PubMed] [Google Scholar]
  25. Hays RD, Bjorner JB, Revicki DA, Spritzer KL, & Cella D. (2009). Development of physical and mental health summary scores from the patient-reported outcomes measurement information system (PROMIS) global items. Quality of Life Research, 18, 873–880. doi: 10.1007/s11136-009-9496-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hays RD, Hayashi T, & Stewart AL (1989). A five-item measure of socially desirable response set. Educational and Psychological Measurement, 49, 629–636. doi: 10.1177/001316448904900315 [DOI] [Google Scholar]
  27. Hyde J, Perez R, & Forester B. (2007). Dementia and assisted living. The Gerontologist, 47, 51–67. doi: 10.1093/geront/47.Supplement_1.51 [DOI] [PubMed] [Google Scholar]
  28. International Council on Active Aging. (2009). The business care for wellness programs in retirement communities and seniors housing. Retrieved from https://www.icaa.cc/business/whitepapers/icaabusinesscase-wp.pdf [Google Scholar]
  29. Kane RA, & Wilson KB (2007). Improving practice through research in and about assisted living: Implications for a research agenda. The Gerontologist, 47, 4–7. doi: 10.1093/geront/47.Supplement_1.4 [DOI] [PubMed] [Google Scholar]
  30. Khatutsky G, Ormond C, Wiener JM, Greene AM, Johnson R, Jessup EA, & Harris- Kojetin L. (2016). Residential care communities and their residents in 2010: A national portrait (DHHS Publication No. 2016–1041). Hyattsville, MD: National Center for Health Statistics. [Google Scholar]
  31. Kline RB (2016). Principles and practice of structural equation modeling (4th ed.). New York, NY: The Guilford Press. [Google Scholar]
  32. Lee HS (2016). Examining neighborhood influences on leisure-time walking in older Korean adults using an extended theory of planned behavior. Landscape and Urban Planning, 148, 51–60. doi: 10.1016/j.landurbplan.2015.12.011 [DOI] [Google Scholar]
  33. Lewinson T, Robinson-Dooley V, & Grant KW (2012). Exploring “home” through residents’ lenses: Assisted living facility residents identify homelike characteristics using photovoice. Journal of Gerontological Social Work, 55, 745–756. doi: 10.1080/01634372.2012.684758 [DOI] [PubMed] [Google Scholar]
  34. Lyons KD, Li Z, Tosteson TD, Meehan K, & Ahles TA (2010). Consistency and construct validity of the activity card sort (modified) in measuring activity resumption after stem cell transplantation. American Journal of Occupational Therapy, 64, 562–569. doi: 10.5014/ajot.2010.09033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. McEachan RRC, Conner M, Taylor NJ, & Lawton RJ (2011). Prospective prediction of health-related behaviours with the theory of planned behaviour: A meta-analysis. Health Psychology Review, 5, 97–144. doi: 10.1080/17437199.2010.521684 [DOI] [Google Scholar]
  36. Meeks S, Van Haitsma K, Schoenbachler B, & Looney SW (2014). BE-ACTIV for depression in nursing homes: Primary outcomes of a randomized clinical trial. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 70, 13–23. doi: 10.1093/geronb/gbu026 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Mihalko SL, & Wickley KL (2003). Active living for assisted living: Promoting partnerships within a systems framework. American Journal of Preventive Medicine, 25, 193–203. doi: 10.1016/S0749-3797(03)00184-3 [DOI] [PubMed] [Google Scholar]
  38. Miller E, & Buys L. (2007). Predicting older Australians’ leisure-time physical activity: Impact of residence, retirement village versus community, on walking, swimming, dancing and lawn bowling. Activities, Adaptation, & Aging, 31, 13–30. doi: 10.1300/J016v31n03_02 [DOI] [Google Scholar]
  39. Mitchell JM, & Kemp BJ (2000). QoL in assisted living homes: A multidimensional analysis. The Journals of Gerontology Series B: Psychological Sciences & Social Sciences, 55, P117–P127. doi: 10.1093/geronb/55.2.P117 [DOI] [PubMed] [Google Scholar]
  40. Motalebi SA, Iranagh JA, Abdollahi A, & Lim WK (2014). Applying of theory of planned behavior to promote physical activity and exercise behavior among older adults. Journal of Physical Education and Sport, 4, 562–568. [Google Scholar]
  41. Nakamura J, & Csikszentmihalyi M. (2014). The concept of flow. In Csikszentmihalyi M. (Ed.), Flow and the foundations of positive psychology (pp. 239–263). Dordrecht, Netherlands: Springer. [Google Scholar]
  42. O’Sullivan G. (2005). Protocols for leisure activity programming. New Zealand Journal of Occupational Therapy, 52, 17–25. [Google Scholar]
  43. Park NS, Zimmerman S, Kinslow K, Shin HJ, & Roff LL (2012). Social engagement in assisted living and implications for practice. Journal of Applied Gerontology, 31, 215–238. doi: 10.1177/0733464810384480 [DOI] [Google Scholar]
  44. Park-Lee E, Caffrey C, Sengupta M, Moss A, Rosenoff E, & Harris-Kojetin L. (2011). Residents care facilities: A key sector in the spectrum of long-term care providers in the United States (NCHS Data Brief No. 78). Hyattsville, MD: National Center for Health Statistics. [PubMed] [Google Scholar]
  45. Plys E. (2017). Recreational activity in assisted living communities: A critical review and theoretical model. The Gerontologist, Gnx138. doi: 10.1093/geront/gnx138 [DOI] [PubMed] [Google Scholar]
  46. Plys EJ, & Qualls SH (2016). Multidimensional demand strength of recreational activity programming in assisted living: A conceptual model and prototypical rating scale. Journal of Housing for the Elderly, 30, 360–379. doi: 10.1080/02763893.2016.1224789 [DOI] [Google Scholar]
  47. Pruchno RA, & Rose MS (2002). Time use by frail older people in different care settings. Journal of Applied Gerontology, 21, 5–23. doi: 10.1177/0733464802021001001 [DOI] [Google Scholar]
  48. Resnick B, Galik E, Gruber-Baldini AL, & Zimmerman S. (2010). Perceptions and performance of function and physical activity in assisted living communities. Journal of the American Medical Directors Association, 11, 406–414. doi: 10.1016/j.jamda.2010.02.003 [DOI] [PubMed] [Google Scholar]
  49. Rhodes RE, Courneya KS, Blanchard CM, & Plotnikoff RC (2007). Prediction of leisure-time walking: An integration of social cognitive, perceived environmental, and personality factors. International Journal of Behavioral Nutrition and Physical Activity, 4, 51–61. doi: 10.1186/1479-5868-4-51 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Sefcik JS, & Abbott KM (2014). “Right back to square one again”: The experience of friendship among assisted living residents. Activities, Adaptation, & Aging, 38, 11–28. doi: 10.1080/01924788.2014.878872 [DOI] [Google Scholar]
  51. Steinmetz H, Knappstein M, Ajzen I, Schmidt P, & Kabst R. (2016). How effective are behavior change interventions based on the theory of planned behavior?: A three-level meta-analysis. Zeitschrift für Psychologie, 224, 216–233. doi: 10.1027/2151-2604/a000255 [DOI] [Google Scholar]
  52. Watson LC, Lehmann S, Mayer L, Samus Q, Baker A, Brandt J, … Lyketsos C. (2006). Depression in assisted living is common and related to physical burden. The American Journal of Geriatric Psychiatry, 14, 876–883. doi: 10.1097/01.JGP.0000218698.80152.79 [DOI] [PubMed] [Google Scholar]
  53. Weston R, & Gore PA Jr. (2006). A brief guide to structural equation modeling. The Counseling Psychologist, 34, 719–751. doi: 10.1177/0011000006286345 [DOI] [Google Scholar]
  54. Williams KN, & Warren CA (2009). Communication in assisted living. Journal of Aging Studies, 23, 24–36. doi: 10.1016/j.jaging.2007.09.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Zimmerman S, Scott AC, Park NS, Hall SA, Wetherby MM, Gruber-Baldini AL, & Morgan LA (2003). Social engagement and its relationship to service provision in residential care and assisted living. Social Work Research, 27, 6–18. doi: 10.1093/swr/27.1.6 [DOI] [Google Scholar]
  56. Zimmerman S, Sloane PD, & Reed D. (2014). Dementia prevalence and care in assisted living. Health Affairs, 33, 658–666. doi: 10.1377/hlthaff.2013.1255 [DOI] [PubMed] [Google Scholar]

RESOURCES