The Effectiveness of Adult Day Services for Older Adults

ADS and Caregiver Well-Being

In terms of outcomes related to caregiver well-being, the literature can be divided into two categories—studies that examined the impact of ADS attendance (respite), in general, and studies that examined the impact of specific interventions in ADS settings. Given the diversity present in ADS programming, it is difficult to determine whether the respite benefits of ADS impacted caregiver well-being or whether the unspecified services offered by ADS impacted the burden and stress of caregivers. Although most ADS centers provide caregiver-specific services such as support groups and individual counseling, there is a critical need for standardized definitions of ADS in order to more effectively and rigorously evaluate ADS outcomes (Tarrant, 2010). Despite these limitations, the recent literature does provide a general understanding of the effectiveness of ADS for caregivers and the potential that ADS has as a platform for caregiver interventions.

The majority of the studies since 2000 have focused on general ADS attendance and caregiver well-being, rather than specific programs and interventions offered through ADS. As indicated in Table 1, the studies were diverse in terms of design, sample size, and analysis. In the one study to use a randomized controlled trial design (N = 103), researchers compared “non-pharmacological treatments” for individuals with dementia. ADS was considered a “treatment” along with several others, including cognitive stimulation, counseling, and education. The researchers found that caregivers in ADS experienced significantly lower levels of hostility from their care recipient compared with caregivers in the other treatment groups (Quayhagen et al., 2000). Hostility was considered to be a component of emotional well-being in this study. A number of studies used quasi-experimental designs, which provide some valuable insight, despite threats to internal validity and limitations regarding causation. Using a relatively large dataset of dementia caregivers (N = 400), Gaugler et al. (2003a, 2003b) found that caregivers who used ADS spent less time addressing behavior problems in care recipients than caregivers who did not use ADS. This was related to reduced feelings of role overload, an important indicator of caregiver burden. In another study using a quasi-experimental design and a fairly large European sample (N = 300), researchers found that employed caregivers and those caring for persons with moderate levels of disability experienced significantly lower levels of caregiver burden when ADS was used rather than traditional home care (Gramain & Malavolti, 2004). Two smaller studies using quasi-experimental designs also found that ADS attendance was associated with significantly less caregiver burden (Mossello et al., 2008) and stress related to family conflict and employment (Schacke & Zank, 2006). Several studies that used cross-sectional approaches further supported the findings that ADS attendance is associated with higher levels of caregiver well-being (e.g., isolation, worry, guilt) and lower levels of caregiver burden and stress (Colvez, Joel, Ponton-Sanchez, & Royer, 2002; Måvall & Thorslund, 2007; Sussman & Regehr, 2009; Valadez, Lumadue, Gutierrez, & de Vries-Kell, 2005). In a more recent study, Zarit et al. (2011) used a within-subject withdrawal design to address the limitations of quasi-experimental and cross-sectional approaches. These researchers found that ADS attendance was associated with decreased levels of stress for dementia caregivers and fewer behavior and sleep problems in care recipients. Although certain studies failed to find significant effects for ADS on caregiver well-being (Baumgarten, Lebel, Laprise, Leclerc, & Quinn, 2002; Warren, Kerr, Smith, Godkin, & Schlam, 2003; Zank & Schacke, 2002), these studies tended to be small in scale and had methodological issues, such as the use of untested measures. While additional research is warranted to understand why ADS might be effective, the studies presented here demonstrate that ADS attendance and the general benefits provided by ADS (e.g., respite, basic programming) can lower caregiver burden and stress and contribute to the overall well-being of caregivers, particularly for those caring for family members with dementia.

Only a small group of studies over the past 10 years have examined the effects of specific interventions and programs on the well-being of caregivers in ADS. While such studies may speak more toward the effectiveness of the intervention rather than ADS, per se, the studies do provide valuable insight into the ability of ADS to serve as a platform for innovative approaches to care. In the most rigorous and informative of these studies, Gitlin, Reever, Dennis, Mathieu, and Hauck (2006) used a quasi-experimental approach (N = 129) to evaluate the effects of ADS Plus, an intervention that combines “typical ADS” with care management, supplemental education and training, and enhanced support for caregivers. Compared with caregivers receiving only ADS services, ADS Plus was associated with significantly lower levels of depressive symptoms in caregivers and an enhanced capacity to manage challenging behaviors. Danner and McGuire (2010) used a pre-post-test design (N = 35) to evaluate the impact of the Best Friends approach to dementia care in ADS. This approach uses a person-centered philosophy and special training for staff and volunteers. The researchers found that the Best Friends approach in ADS was associated with slight gains in overall quality of life for caregivers. Droes, Breebaart et al. (2004) examined the effects of an integrated support program (e.g., supplemental support groups, education, and counseling) on stress and coping for dementia caregivers and their care recipients. The results indicated that the support program had few significant effects; however, caregivers did report significantly higher feelings of competence compared to caregivers in ADS “as usual.” The final study used a cross-sectional approach (N = 62) to compare the effects of support groups in ADS with support groups in the community (Bartfay & Bartfay, 2011). The researchers failed to find significant differences in the effects of support groups between the settings. Caregivers in both groups experienced similar benefits. Despite the small number of studies that have tested specific interventions in ADS and limitations in sample sizes and design, it does appear that the congregate nature of ADS could make it an appropriate and effective platform for the delivery of interventions and programs. Moreover, the provision of interventions in ADS may potentially be more efficient than in other community-based settings (e.g., senior centers) because of the accessibility and availability of programs, services, and staff in ADS. For example, most ADS centers already employ professionals such as nurses, social workers, and direct care staff that could assist with and/or provide interventions to participants attending ADS. Additionally, there are opportunities for group interventions in ADS as most centers currently provide programs such as caregiver education and support groups. However, more research is needed to support the rationale for ADS as a unique milieu for participant and caregiver interventions.

ADS and Participant Well-Being

The literature as it relates to participant outcomes can be categorized by interventions designed to address activity engagement/participation, cognition, behaviors, physical functioning, and overall well-being. As indicated in Table 2, the diversity of purpose, design, and type of interventions makes it difficult to ascertain the effectiveness of ADS on participant outcomes. However, the literature does provide a starting place for understanding interventions in these settings and future directions for ADS research.

Many studies over the past decade have examined the effectiveness of interventions related to activity engagement/participation in ADS. Among the most recent is a quasi-experimental study of the effectiveness of a civic engagement intervention with ADS participants (N = 43) conducted by Dabelko-Schoeny, Anderson, and Spinks (2010). Participants receiving the intervention reported higher, yet nonsignificant, levels of purpose in life, self-esteem, and perceived physical health. However, when the intervention was removed significant declines were reported in self-esteem and perceived physical health. Cohen-Mansfield, Parpura-Gill, and Golander (2006) used a randomized controlled study to examine the impact of interventions that incorporate a participant’s self-identity (e.g., lifetime role, hobbies, important family members) with activities in ADS and nursing homes (N = 93). Researchers found that ADS participants reported higher self-identity awareness and demonstrated higher level of interest in these activities than nursing home participants. A number of studies have also explored the effect of horticulture therapy in ADS and have found positive effects on engagement for participants in these activities (Gigliotti & Jarrott, 2005; Gigliotti, Jarrott, & Yorgason, 2004; Jarrott, Kwack, & Relf, 2002). In the largest of these studies (N = 129), Jarrott and Gigliotti (2010) found that participants in the horticultural therapy treatment group demonstrated higher levels of active and passive engagement than the comparison group. A small-scale analysis (N = 30) conducted by Bruin et al. (2009) found that “green farm” programming (e.g., feeding animals, cultivating vegetables) resulted in more intense, frequent activities for ADS participants in these programs compared to regular day care facilities. However, the results of these studies should be interpreted with caution given a number of methodological issues, including small sample sizes.

In addition to the use of ADS as an effective “intervention” for dementia (Cherry & Simmons-D’Gerolamo, 2005; Ishizaki et al., 2002), interventions such as art-based programs (Rentz, 2000) and music therapy (Jennings & Vance, 2002) have shown positive effects for cognitively impaired ADS participants. A number of small-scale studies have also focused on the effectiveness of Montessori-based programs in ADS. Judge, Camp, and Orsulic-Jeras (2000) used a pre-post test design and found that compared with traditional ADS activities, participants in the Montessori-based programs exhibited greater amounts of engagement. Similarly, Vance and Porter (2001) and Vance and Johns (2002) found a moderate benefit for ADS participants in Montessori programming, including a positive influence on cognition. Skrajner and Camp (2007) also found engagement and affect to be more positive among participants in Montessori-programming than in standard ADS activities. Furthermore, several small studies have examined the effectiveness of interventions designed to address various problematic behaviors and have shown benefits for participants engaged in dementia-specific programs (Danner & McGuire, 2010). A small, quasi-experimental pre-post test control group study conducted by Droes, Meiland et al. (2004) found that dementia support groups showed moderate effects on behavior problems and large effect on depressive behaviors (N = 112). These findings support earlier studies that found a positive effect on the total number of behavior problems among participants in the dementia support group compared to participants in the regular ADS programming (Dröes, Breebaart, Ettema, van Tilburg, & Mellenbergh, 2000). In a more recent quasi-experimental study, Mossello et al. (2008) found that behavioral and psychological symptoms were significantly decreased in the ADS participants compared to participants only using home health care services. Finally, Femia, Zarit, Stephens, and Greene (2007) explored whether ADS use was associated with reductions in behavior and psychological symptoms of dementia in a larger quasi-experimental study (N = 201). Findings suggested that within the ADS group, behaviors such as agitation and symptoms of depression were lower on days when participants attended ADS; however, there were no significant differences between the experimental and control group (non-ADS users).

Several non-randomized quasi-intervention studies have examined the effectiveness of interventions on the functional status of ADS participants including falls (Diener & Mitchell, 2005), gait (Hageman & Thomas, 2005), and motor skills (Davis, Hoppes, & Chesbro, 2005). Study findings suggest an overall positive impact of these interventions on ADS participants (e.g., less falls); however, it is difficult to determine what fully accounts for these effects due to small sample sizes. Other studies, such as Baumgarten et al. (2002), did not find demonstrable effects of ADS on functional status compared to the waiting list comparison group (N = 292). Similarly, in a small longitudinal study conducted by Burns, McCarten, Adler, Bauer, & Kuskowski (2004), participants transitioning to ADS from a veteran’s work program experienced functional declines. One of the few qualitative studies of participant outcomes in ADS related to physical well-being was conducted by Weintraub and Killian (2009). Through interviews with participants in an intergenerational ADS center (N = 13), content analysis suggested that participants perceived that the impact of ADS on their physical well-being depended in part on the degree to which they were involved with children at the center and participated in the intergenerational programming.

Studies that investigate how ADS affects the overall wellness of participants suggest that ADS attendance is associated with improvements in physical and emotional problems (Schmitt, Sands, Weiss, Dowling, & Covinsky, 2010), perceived psychosocial well-being (Dabelko-Schoeny & King, 2010), and positive changes in social support and quality of life (Washburn, Luxenberg, Brod, Steinhauer, & Katsap, 2001). Zank and Schacke (2002) conducted a quasi-experimental study investigating the effects of ADS on well-being, health indicators, and ADLs. Findings suggested that when compared to the untreated control group, ADS participants experienced overall positive effects of day care on their well-being and dementia symptoms. A qualitative study exploring participant outcomes in ADS conducted by Valadez et al. (2005) found that socialization and social support from other ADS clients had a positive impact on the physical and mental well-being of participants. Similarly, ADS participants in a qualitative study of intergenerational programs reported overall positive experiences and feelings of “being needed” (Weintraub & Killian, 2007).

ADS and Health Care Utilization

Given the escalating financial costs of institutional care and the toll that institutionalization can have on individuals and families, researchers have long explored whether ADS could play a key role in delaying or preventing NHP. Findings from research conducted prior to 2000 suggested that ADS does not appear to delay or prevent NHP. Questions remain, however, as to whether institutionalization is a valid measure of the effectiveness of ADS (Gaugler & Zarit, 2001). ADS may serve as a valuable preparatory stage for moving into an institutional setting. While the nursing home is generally not the preferred residence for most individuals and families, it may be the most appropriate setting for care when acuity levels surpass the capacities of HCBS. In the current review, we move beyond looking at the relationship between ADS and NHP and examine whether ADS use can impact other types of service usage, most notably hospitalization rates. In doing so, we begin to expand our understanding of the impact of ADS on health care utilization, a topic of keen interest to health care policymakers.

As indicated in Table 3, 10 articles have been published that focused on ADS and health care utilization since 2000, 9 of which used NHP as the outcome variable. In two of these articles, Gaugler et al. (Gaugler., Kane, Kane, Clay, & Newcomer, 2003; Gaugler, Kane, Kane, & Newcomer, 2005) used data from the Medicare Alzheimer’s Disease Demonstration Evaluation (MADDE), a large, multi-site study of the effectiveness of expanded community-based services (including ADS) for individuals with dementia and their caregivers. In the earlier of these two studies, the objective was to identify predictors of NHP over a 3-year period for 3,944 older adults diagnosed with dementia. Interestingly, the findings indicated that ADS usage had a curvilinear effect on NHP. Caregivers who used either low or high amounts of ADS were 1.30 times more likely to institutionalize sooner. It was concluded that there may be a dosage effect and caregivers who used low amounts of ADS may not have used sufficient amounts of ADS to benefit. Caregivers who used high amounts of ADS may have used these services as a “stepping stone” to institutionalization. In the second study using the MADDE dataset (N = 4,761 in this case), Gaugler et al. (2005) tried to determine whether utilization of community-based services (including ADS) early in the dementia caregiving process could delay NHP. Results revealed that there were no demonstrable benefits from ADS attendance in terms of delaying NHP. The authors noted, however, that ADS can be beneficial to caregivers and may delay NHP if used in certain patterns.

Two sets of researchers drew upon data from the Adult Day Collaborative Study (ADCCS), a quasi-experimental study that evaluated the effectiveness of ADS on a number of outcomes, including time until NHP. In the first of these analyses, the researchers sought to determine if the pattern with which dementia caregivers (N = 305) interface with informal help (e.g., family, friends) and formal help (including ADS) impacted the decision to institutionalize care recipients (Lyons, Zarit, & Townsend, 2000). Results indicated that caregivers who used ADS more frequently were more likely to institutionalize compared with caregivers who relied on themselves only or on other family members. In the second analysis, the researchers examined whether there were differences between daughter and wife dementia caregivers (N = 371) who used ADS in terms of their decision to institutionalize care recipients (Cho, Zarit, & Chiriboga, 2009). Findings indicated that daughters who used ADS were more likely to postpone NHP than daughters who did not use ADS. The researchers also found that wives using ADS placed care recipients into nursing homes significantly sooner than daughters using ADS. These findings point toward the complexity of the decision to place a family member into a nursing home and suggest that ADS use can play a role in delaying NHP, particularly for adult children.

Three independent studies used primary longitudinal data to examine the relationship between ADS and NHP. The earliest of these studies focused on a sample of older adults with dementia (N = 298), some of which were using ADS (McCann et al., 2005). The researchers found that risk of NHP significantly increased with increases in the number of days that care recipients attended ADS. These findings remained significant after controlling for disease severity and caregiver burden. A more recent study conducted with a Swedish sample of older adults with dementia (N = 880) found similar results; increase in ADS use was a significant predictor of shorter time to NHP (Wattmo, Wallin, Londos, & Minthon, 2011). Both sets of researchers acknowledged that the rationales and factors related to NHP are complex and that increased ADS use may be a last gasp attempt at keeping care recipients in the home. Running contrary to these findings, Wilson et al. (2007) found that higher levels of ADS use substantially reduced the risk of NHP for older adults experiencing accelerated cognitive decline (N = 432). These researchers concluded that ADS may be used as a transitional stage between community and institutional living.

The final two studies used quasi-experimental approaches to evaluate interventions in ADS compared with ADS services “as usual.” Gitlin et al. (2006) evaluated the effects of ADS Plus, an approach that combines usual ADS services with supplemental care management, support, and education. ADS Plus caregivers were found to use ADS services significantly more frequently and had a significantly lower rate of NHP. Droes, Breebaart et al. (2004) evaluated a similar integrated support program, the Meeting Centres Support Program, for individuals with dementia and their caregivers. The researchers found that participation in the intervention was related to significant delays in NHP. While both of these interventions appear to be promising, replication and larger studies are needed regarding both the effectiveness and feasibility of such interventions in ADS.

Physical and cognitive limitations as well as chronic disease are high among the ADS population. The most prevalent medical needs among ADS participants are dementia (47%), hypertension/high blood pressure (46%), and physical disability (42%) (MetLife, 2010b). Reflective of advances in the capacity of ADS to serve complex medical needs (for detailed information, see the recent MetLife, 2010a), researchers have begun to look at the effectiveness of ADS to reduce health care utilization, most notable emergency room visits and rehospitalization. In a 2011 study, Jones et al. used a quasi-experimental, retrospective approach to evaluate the impact of a pilot program for the provision of post-acute care in ADS. Participants requiring post-acute services (N = 78) were discharged from one hospital into either ADS or alternative destinations, including skilled nursing facilities, home health, or home with outpatient services or self-care. Those who participated in ADS services were significantly less likely to be readmitted to the hospital or to visit the emergency department within 30 days following discharge. The same effect was also evident within 90 days following discharge; however, the differences did not reach levels of significance. There were limitations in this study, especially in terms of sample size and the self-selection process. Nonetheless, these findings point toward the potential effectiveness of ADS to serve as an effective step down following hospitalization.

In looking at the effects of ADS on health care utilization, the picture is far from clear. Reviews of research prior to 2000 concluded that ADS programs do not consistently delay NHP and may, in actuality, “facilitate family members’ decisions to place” through a step-down process of relinquishing care (Gaugler & Zarit, 2001). Research conducted since 2000 on “usual” services offered through ADS appears to confirm this conclusion and further confirms that the decision to place a family member into a nursing home is typically a complex process.