Programmed Activity Offerings and Attendance Patterns in Assisted Living

Abstract

The scope of programmed activity offerings and attendance rates for specific activities are not thoroughly captured in the assisted living (AL) literature. The purpose of this study is to report activity offerings and associated attendance rates, as well as relationships between individual factors and activity attendance patterns in a sample of 202 residents of 21 ALs. Sampled communities offered 50 different types of programmed activities. Each AL offered exercise and religious services, most offered BINGO (n = 19, 91%) and socials (n = 18, 86%). BINGO was the most frequently attended activity (n = 83; 47%), followed by religious services (n = 75; 38%), socials (n = 67; 40%), and musical performances (n = 62, 37%). Additional findings provide insight into the features of frequently attended activities, and the relationships between attendance and resident characteristics. The authors conclude with a discussion of implications for service delivery and future research.

Introduction

Population trends and industry growth suggest that an increasing number of older adults will relocate to assisted livings (ALs) in the near future (Grabowski et al., 2012; Houser et al., 2012). ALs market a homelike environment designed to support a vibrant, active lifestyle for individuals that require assistance with activities of daily living (CEAL, 2010). Activity programming is an important service that promotes social, recreational, and physical engagement, and may lead to various positive outcomes for AL residents (Chao & Chen, 2019; ICAA, 2009; Ouyang et al., 2015). Recent national data suggests that 100% of ALs offer programmed activities (Khatutsky et al., 2016); yet, in general, residents evidence low rates of recreational activity participation (for review see Plys, 2019). Multiple studies suggest that individual (e.g., mood, vitality, and cognition), socio-environmental (e.g., social support and activity-related norms), and activity (e.g., recreational preferences and satisfaction with programming) variables relate with activity participation in AL (Holmes et al., 2017; Park et al., 2017; Plys & Qualls, 2019). Even though the literature has increasingly captured the variability in predictors of resident activity behaviors, rarely does this body of research appreciate variability in the targeted behavior itself (i.e., activity participation). Failure to capture behavioral variability is particularly true of research targeting programmed activities in AL. A deeper look at which activities are offered, as well as the attendance rates and characteristics of resident attendees for each activity could be the first step toward a more sophisticated understanding of resident attendance patterns for programmed activities in the AL setting.

Activity programs offer a variety of recreational opportunities in AL (Hanson et al., 2014). Yet, inconsistent activity categorizations and measurement protocols limit the scientific investigation of attendance rates across the spectrum of activity offerings. A recent review of recreational behaviors in AL concluded that most studies combine programmed and self-initiated activities to assess overall recreational participation (see Plys, 2019). Therefore, by investigating global activity rates, the existing literature cannot comment on issues important to activity programming as a service, like: are certain activity offerings better attended? Some studies group activities into categories, like within-outside the AL (Jang et al., 2014) or group-individual (Gaugler & Kane, 2005); but these categories are often too broad to yield results that could inform practice. Other studies assess attendance for a subset of specific activities, but then combine these activities into an overall index (Ouyang et al., 2015; Park, 2009). Taken together, the activities literature has failed to identify specific recreational opportunities available in AL as a basis for categorization and measurement. This lack of specificity in measurement limits the practical significance of empirical findings for activity professionals. In fact, no known study has reported a full range of specific programmed activities offered in AL.

As mentioned, previous studies, as well as national datasets, fail to report detailed information on specific programmed activity offerings and attendance rates in AL, leaving the following questions unanswered: What activities do ALs offer? What activities do residents attend? What are the characteristics of the residents who attend specific activities? The purpose of this study is to provide initial insight into the aforementioned questions by describing activity offerings, attendance rates, and attendee characteristics in a sample of 202 residents across 21 ALs. First, we will identify a comprehensive index of programmed activity offerings across the sampled ALs. Next, we will report attendance rates for each identified activity offering in a sample of residents from the targeted communities. Lastly, we will describe the relationships between attendance for specific activities and relevant resident characteristics (i.e., health, social, and personality variables). Results from this study are a first step toward detailed measurement and scientific inquiry into variability in programmed activity offerings in AL.

Materials and Methods

AL Characteristics

Twenty-one ALs from the state of Colorado (USA) ranging in size from 16 to 104 residents (M = 54.19, SD = 25.07) participated in this study. The majority of sampled ALs were owned by a chain operating company (n = 11, 52%); one (5%) was owned by a religious organization. Ten (48%) were free-standing communities and 11 (52%) were part of a continuing care community. Most of the ALs (n = 16, 76%) cost more than $3,000 per month and targeted private-pay residents (i.e., 75–100% private-pay beds; n = 14, 67%). In five (24%) ALs, more than half of the beds were for residents using Medicaid. Almost the entire sample of ALs had a room designated for activities (n = 20; 95%), most had outdoor walking areas (n = 17; 81%), ten (48%) communities had an exercise facility, and five (24%) had a pool on site. Most of the communities were in a suburban area (n = 13, 62%); only two (10%) identified their surrounding community as rural.

Resident Characteristics

Resident participants ranged in age from 51 to 100 (M = 83.03, SD = 10.27), the majority were aged 85 or older (n = 112, 55%). Most participants were female (n = 144, 71%), white (n = 182, 90%), and widowed (n = 119, 59%). The majority had higher than a high school education (n = 135, 67%), and used private forms of payment (n = 141, 70%). Length of stay ranged from less than a month to over 13 years (M = 21.17 months, SD = 20.44); most residents relocated to their AL from a private community home (n =132, 65%).

Activity Program Characteristics

The majority of ALs employed one full-time activity professional (n = 11, 52%); three (14%) had no full-time activity staff members. Ten (48%) activity programs had more part-time staff or volunteers than full-time staff, and three (14%) had the same number of full-time and part-time or volunteer staff. Table 1 presents a summary of activity program characteristics.

Table 1:

Activity Program Characteristics

	N	%	Range	M	SD	Correlation with Offeringsa (n = 21)
Activity Offerings	21		11–65	38.01	12.64
Activity Director Length of Job (months)	20		2–240	55.40	60.46
Activity Director Education
High School Diploma	8	40
Some College/Associate’s Degree	4	20
Bachelor’s Degree	7	35
Master’s Degree	1	5
Activity Director Certification	2	10
Full-Time Activities Staff	21		0–8	1.71	1.79	.29
Part-Time Activities Staff	21		0–12	3.00	3.49	.26
Daily Length of Programming (hours)	21		2–10	7.71	1.68	.36
Weekend Activity Offerings (hours)
1–5	10	48
6–8	9	43
9–16	2	9
Activities after 5pm	21		0–5	1.48	1.21	.29

Note.

^aCorrelation coefficients presented are between continuous activity programming variables and total activity offerings in the community. Items describing activity directors have a missing data point (n = 20), reflecting one community that did not have an activity director.

Procedures

To identify data collection sites, the authors contacted AL administrators via email, phone call, or site visit (49 ALs contacted, 15 participated). After this initial recruitment effort, we partnered with a regional senior housing operator, state activity directors association, and local non-profit organization to identify additional communities (6 additional ALs contacted, 6 participated). Staff at each community identified residents who were interested and appropriate for the current study; other recruitment strategies included in-person announcements and snowball sampling. Inclusion criteria were: (a) age 50 or older; (b) able to speak and understand English; (c) own decision maker; and (d) score of five or above on the Memory Impairment Screen (see Buschke et al., 1999). Of the 235 residents who volunteered, 202 met criteria and were included in the current sample.

All resident interviews began with informed consent and a brief cognitive screen to determine eligibility. The majority of surveys were administered one-to-one. Rarely (< 10%), alternative strategies were used to complete surveys, including a group interview or a take-home copy of the interview. There were no difference in study variables or social desirability (F(2, 199) = .24, p = .789), as measured by the Five-item Socially Desirable Response Set (Hays et al., 1989), across survey administration method.

At each AL, staff (i.e., executive directors, administrative assistants, marketing specialists, or activity directors) completed surveys describing the community and surrounding area. The researchers also administered an activity programs survey describing activity staffing and programming. Activity directors, or staff with similar titles (e.g., resident programs coordinator), completed most of the activity program survey (n = 19). When activity directors could not be reached (n = 1) or the community did not have an acting activity director (n = 1), administrative staff completed the activity programs survey. Programmed activity calendars for the month of data collection were collected from each community.

Measures

AL Characteristics

Staff reported community ownership type, size, costs, resident financing strategies, and physical characteristics, as well as an activity program survey targeting department staffing (e.g., job length, training and education, certification, and full-time and part-time staff) and programming characteristics (e.g., length of programming, evening activities, and weekend offerings).

Resident Characteristics

Residents self-reported demographic information including gender, age, education, relationship status, ethnicity, Medicaid status (i.e., a proxy for socioeconomic status), and length of stay in months.

Clock Drawing Task-Command (CDT)

The CDT measured cognitive ability (Shulman, 2000), using the Cahn et al. (1996) scoring method: integrity of the clock face (0–2); presence and sequencing of numbers (0–4); and presence and placement of the hands (0–4). Higher scores indicate greater cognitive ability; scores below seven are indicative of cognitive impairment (Cahn et al., 1996). Inter-rater reliability between two independent raters using a subsample of 28 drawings from the current sample was acceptable (ICC = .71, p < .001).

PROMIS Short-Form Global Health Scale

Subscales from the PROMIS measured physical and mental health (Hays et al., 2009). Subscale scores range from 4–20 for each subscale, with higher scores indicating better health and function. Internal consistency with the sample was acceptable for both the physical health (Cronbach’s α = .71) and mental health (Cronbach’s α = .73) subscale.

Vision and Hearing

Participants self-reported visual (i.e., “How would you rate your eyesight, even when wearing glasses?”) and hearing (i.e., “How would you rate your hearing, even when wearing hearing aids?”) ability on a five-point scale (1 = Poor, 5 = Excellent). Items were summed, higher scores indicate greater sensory ability.

Lubben Social Network Scale (LSNS-6)

A modified version of the LSNS-6 (Lubben et al., 2006) measured social relationships with residents, staff, and outside visitors (i.e., family and friends). Items assessing network size, perceived tangible support, and perceived emotional support were summed for each social network; the tangible support item was omitted from the staff subscale. The staff subscale ranged from 0–10, while the resident and outside visitors subscales ranged from 0–15, with higher scores indicating greater social support. Internal consistency with the sample was acceptable for the co-resident (Cronbach’s α = .83), staff-resident (Cronbach’s α = .74), and outside visitors (Cronbach’s α = .87) subscale.

Big Five Inventory (BFI-10)-Extraversion Subscale

Two items from the BFI-10 measured extraversion (Rammstedt & John, 2007). Participants responded to the prompt, “I see myself as someone who…,” on a five-point scale (1= disagree strongly, 5 = agree strongly). Scores ranged from 2–10, with greater scores indicating higher extraversion. Internal consistency with the sample was acceptable (Cronbach’s α = .73).

Activity Attendance

Participants self-reported activity attendance by indicating whether they attended each programmed activity listed on their community’s activity calendar (1 = Yes, 0 = No) in the seven days prior to the date of interview.

Data Analyses

Missing data was observed only for the CDT (n = 17; 8%). To minimize bias, we did not impute scores for missing data; therefore, participants with missing data points were not included in correlations with the CDT (n = 185). Otherwise, the data met all statistical assumptions. Descriptive and correlational analyses were performed using SPSS v.24.

Results

The sampled ALs offered between 11 and 65 (M = 38.01, SD = 12.64) programmed activities per week. Fifty different types of activities were offered across the 21 calendars. Each AL offered religious services and exercise. Most ALs offered BINGO (91%), socials (86%), musical performances (81%), and movies (81%). Table 2 presents programmed activity offerings and associated attendance rates.

Table 2:

Types of Programmed Activities Offered by ALs and Attended by Residents

Activity	Offered by AL		Residents Attended		Activity	Offered by AL		Residents Attended
Religious Service	21	100%	75	38%	Reminiscence	6	29%	19	28%
Exercise	21	100%	59	29%	Chair Tai Chi/ Chair Yoga	6	29%	13	29%
BINGO	19	91%	83	47%	Cooking	6	29%	8	14%
Social Hour	18	86%	67	40%	Book Club	6	29%	7	16%
Music	17	81%	62	37%	Holiday/ Special Event	5	24%	18	41%
Movie	17	81%	25	16%	Story Time	5	24%	13	30%
Outing	15	71%	23	16%	News/ Current Events	5	24%	11	23%
Board Game	15	71%	19	14%	Gambling	5	24%	7	19%
Manicure or Pedicure	14	67%	23	20%	Gardening	5	24%	6	12%
Cards	13	62%	16	14%	Visits	5	24%	3	10%
Crafts/ Sewing	12	57%	27	25%	Game Show	4	19%	12	23%
Trivia	11	52%	29	35%	Coffee or Tea Social	4	19%	9	24%
Presentation/ Lecture	11	52%	28	27%	Balance	4	19%	7	18%
Stretch	11	52%	18	20%	Tech Lessons	4	19%	6	13%
Painting or Coloring	11	52%	13	12%	Watching Sports	4	19%	6	11%
Resident Council	10	48%	28	29%	Table Games (e.g., dice)	4	19%	5	13%
Brain Fitness	9	43%	29	34%	Men’s Group	4	19%	2	9%
Walking	9	43%	3	4%	Water Aerobics	3	14%	8	20%
Birthdays	8	38%	28	37%	Wii Games	3	14%	5	13%
Word Games (e.g., crosswords)	8	38%	21	28%	Pet Visits	3	14%	3	20%
Van Ride/ Scenic Drive	8	38%	15	23%	Meditation/ Aromatherapy	3	14%	3	16%
Chair Exercises	8	38%	14	23%	Volunteering	3	14%	1	6%
Singing	8	38%	13	20%	Humor-Based	2	10%	5	42%
Conversations	7	33%	6	11%	Nature/ Bird Watching	1	5%	1	33%
Puzzles	7	33%	4	7%	Caregiver Support Group	1	5%	0	0%

Note. Percentages of attendance are based on the number of residents that attended the activity only if offered by their community.

Participants attended between 0 and 41 (M = 7.49, SD = 7.48) programmed activities per week, or 0 to 93% (M = 20.64%, SD = 19.58) of the activities offered by their community. In the current sample, 30 participants (15%) did not attend any activity. Because of variability in the activities that communities offered, we present both the raw number and percentage of activities attended. By number, BINGO was the most frequently attended activity (n = 83; 47%), followed by religious services (n = 75; 38%) and socials (n = 67; 40%). By percentage, BINGO was also the most attended activity (47%; n = 83), followed by humor-based activities (42%; n = 5) and holiday celebrations or special events (41%; n = 18). Relationships between resident characteristics and attendance rates for the most frequently attended activities are presented in Table 3.

Table 3:

Relationships Between Resident Characteristics and Attendance for Popular Activities

	BINGO	Religious Service	Socials	Music	Exercise	Trivia	Brain Fitness	Birthdays	Resident Council	Presentation/ Lecture	Crafts/ Sewing	Movie	Manicure/ Pedicure	Outing
Cognitive Ability	.12	.05	.04	.04	−.11	.03	.22	−.14	.06	.14	−.01	−.10	−.07	.09
Physical Health	.03	.04	.16 *	.12	.12	.17	.31 **	.27 *	.08	.22 *	.13	.25 **	.11	.22 **
Mental Health	.11	.06	.20 *	.13	.07	−.05	.08	.14	.13	.25 *	.11	.20 *	.14	.12
Sensory Ability	−.01	−.04	.15 *	.02	−.06	−.10	.07	.22	.23 *	.08	.05	.09	−.11	.13
Outside Relationships	.13	.15 *	.18 *	.13	.15*	.02	.28 *	.07	.20	.31 **	.06	.20 *	−.06	−.11
Resident Relationships	.21 **	.19 **	.24 **	.27 ***	.12	.17	.22 *	.24 *	.15	.19	.25 **	.13	.12	.17 *
Staff Relationships	.05	.14	.13	.15*	.07	.11	.19	.16	−.01	.14	.18	.18 *	−.01	.13
Extraversion	.10	.13	.25 **	.18 *	.07	.08	.11	.12	.13	.11	.04	.09	.07	.16

Note.

^*p < .05.

^**p < .01.

^***p < .001. Residents were excluded from analyses if their community did not offer the targeted activity.

Discussion

The purpose of this study was to identify a comprehensive set of programmed activity offerings in AL and associated attendance rates, as well as to report the relationships between attendance for specific activities and select resident characteristics. Key findings suggest that the sampled ALs offered a core set of programmed activities typical of a residential long-term care setting, and that resident attendance rates varied by activity type and behavioral features. The current results have implications for practice, including planning activity calendars, as well as for future research, particularly for studies addressing activity measurement.

What Activities Do ALs Offer?

The most commonly offered programmed activities, that appeared on the majority of sampled calendars, were conventional for a residential long-term care setting (i.e., religious services, exercise, BINGO, and socials). ALs appearing in this study were less likely to offer activities that required substantial community resources or specialized staff training, such as caregiver support groups, water aerobics, balance class, or pet visits. This finding, coupled with low rates of full-time staffing for activity professionals, suggests that the sampled communities may have invested modest resources into activity programs, despite the centrality of this service to the AL philosophy and utility for marketing to potential residents (Castle & Standtlander, 2009; CEAL, 2010). Future research is needed to investigate the resources ALs invest into activity programs in a national sample, as well as the impact of various resources on service delivery.

What Activities Do Residents Attend?

Despite variability in attendance rates, a few activities emerged as yielding high attendance rates, offering insight into the features of popular programmed activities. Consistent with previous findings (Hanson et al., 2014; Thomas et al., 2013), four of the ten most frequently attended activities were primarily sources of entertainment or social interaction (i.e., BINGO, socials, music, and birthday parties). Three of the most frequently attended activities related to learning or cognitive stimulation (i.e., brain fitness, presentations or lectures, and trivia). Other top attended activities related to personal roles (e.g., community member), goals (e.g., improve health), or values (e.g., spirituality), such as religious services, exercise classes, and resident council meetings.

The features of popular activities in this study offer insight into possible behavioral goals or perceived benefits. For example, attending resident council may be viewed as a way to assert choice, control, and previous roles (e.g., civic engagement) after relocation to AL. Similarly, cognitive decline is a salient fear among AL residents (Dobbs et al., 2008), and, thus, residents may have attended certain cognitive-based activities, like brain fitness, with the goal of preserving cognitive abilities. Future research is needed to identify additional predictors of attendance for specific activities among AL residents. Overall, programmed activities that featured entertainment, socialization, cognitive stimulation, and opportunities for role-, goal-, or value-related behaviors were particularly popular amongst the current sample of AL residents.

Some of the lowest attended activities identified in this study were walking, volunteering, puzzles, conversations, visits, watching sports, meditation, and book clubs. Most of the participants in this study likely had the physical and cognitive abilities to participate in these types of activities without assistance from staff. Therefore, low attendance rates may reflect residents’ preference to engage in accessible activities on their own time. Relatedly, most of the well-attended activities identified in this study are best delivered in a group context and also benefit from a lead facilitator (e.g., BINGO and trivia). These findings suggest that traditionally solitary activity offerings (e.g., puzzles and watching sports) may attract fewer residents when offered in a group context in the AL setting. It should be noted that residents with significant health limitations, who may benefit more from staff assistance to promote engagement, may be more likely to attend the activities that were unpopular among the current sample. Future research is needed to investigate activity needs and attendance rates in a representative sample of AL residents that includes individuals with physical and cognitive limitations greater than what was captured in the current study.

What Are the Characteristics of the Residents Who Attend Specific Activities?

Relationships between attendance for specific activities and health, social, and personality variables yielded targeted results, suggesting attendee profiles differed by activity. Many of the significant relationships identified in this study linked activity features with related resident characteristics. For example, participants high in extraversion were more likely to attend activities that involve social behaviors, like socials. Similarly, residents with greater co-resident relationships were more likely to attend activities that involved inter-personal interactions or bonding based on mutual interests, like birthday parties or crafts. These findings suggest that specific programmed activities attract residents with characteristics congruent with the features of that activity (e.g., outgoing residents are more likely to attend socials). Alternatively, specific programmed activities may produce different outcomes based on behavioral features (e.g., attending BINGO increases co-resident friendships for persons of all levels of extraversion). Because the current study was cross-sectional, we cannot interpret the direction of these relationships. To address this, future research may consider testing person-centered predictive models of activity attendance in AL.

Limitations and Conclusion

As is the case with many studies conducted in residential care settings (Wrights et al., 2015), a non-representative sample due to recruitment strategies and inclusion criteria limit the generalizability of our results to comparable ALs. Further, precise assessments of activity behaviors are difficult in AL as typical methods of measurement (e.g., likert scales, open-ended questions, or checklists) are reliant on self-report, and, thus, are susceptible to recall bias. The current study is no different, as participants retrospectively self-reported activity participation. To address bias, we prompted recall by using an activities calendar and excluded residents with significant memory impairment; however, self-report is still an imperfect estimate of human behavior and is a limitation of the current study. Because staff-reported activity data is rarely documented in a consistent manner and behavioral observations are difficult to implement in multi-site studies, there is a need for future research to address the issue of bias in the measurement of AL activity behaviors. Lastly, as mentioned, our ability to comment on causal relationships in the current study is limited by cross-sectional design.

Despite limitations, this study provides a unique and detailed description of programmed activity offerings and attendance profiles in a sample of ALs. The current results may have implications for activities professionals and researchers. First, findings related to popular programmed activities and the characteristics of resident attendees may help activity professionals plan monthly calendars. Specifically, entertainment and social-oriented activities, as well as activities that might promote cognitive and health outcomes were popular in the current sample. Further, consistent with previous research (Dobbs et al., 2005; Plys & Qualls, 2016), findings suggest that preference assessments, coupled with information on health and personality, may be useful for identifying activities that align with resident interests and needs. Second, the methods and measures used in this study may help guide future research on programmed activities in AL. Specifically, the index of activity offerings identified in this study may help improve activity measurement by informing the inclusion of micro-level (i.e., specific activities) items in future studies in AL (Nimrod & Shrira, 2016). Additional research is needed to investigate programmed activity offerings and attendance rates with a larger representative sample, for which the current study may serve as a guide.