NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Wysocki A, Butler M, Kane RL, et al. Long-Term Care for Older Adults: A Review of Home and Community-Based Services Versus Institutional Care [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2012 Nov. (Comparative Effectiveness Reviews, No. 81.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Cover of Long-Term Care for Older Adults

Long-Term Care for Older Adults: A Review of Home and Community-Based Services Versus Institutional Care [Internet].

Show details

Results

Introduction

Our findings are organized by Key Question (KQ), following an overview of the literature search results and a description of the included studies. The section includes tables with characteristics and outcomes for each KQ. Articles that used the same population sample are grouped together within these tables.

Due to heterogeneity in study populations, settings, and measurement of outcomes, we could not synthesize results quantitatively. Therefore, we based all conclusions on qualitative analysis. We used cross-sectional studies only to examine case mix addressed in KQ 1a. We assessed risk of bias only for the longitudinal peer-reviewed studies from the United States for KQs 1b, 1c, and 2. The risk of bias assessments and overall strength of evidence ratings appear in Appendix E and Appendix F. Only outcomes that were assessed for strength of evidence are reported in the strength of evidence table in Appendix F. We excluded no studies from analysis based on risk of bias.

Results of Literature Searches

Our search yielded 2,043 unique references. Figure 2 describes the results of the literature search and screening process. A total of 73 full-text articles were screened for final inclusion. Of the 37 articles excluded at the full-text screening stage, 3 were nonapplicable countries, 10 had no relevant comparison, 8 had no relevant settings or sample, 15 had no relevant characteristics or outcomes, and 1 was a review. Excluded references and reasons for exclusion are listed in Appendix C. This step in database searching produced a total of 36 articles for final inclusion. Hand searching produced an additional 6 articles, 5 of which were grey literature reports, for a total of 42 articles that addressed 32 unique studies.

Figure 2 is the literature flow diagram. It describes the results of the literature search and the screening process to identify eligible articles. These results are also described in the text of the report.

Figure 2

Literature flow diagram for review of long-term care for older adults.

Description of Included Studies

No randomized controlled trials (RCTs) were identified. Of the 42 observational studies included in the review, 37 were peer-reviewed journal articles,32-68 and five were grey literature reports.69-73 The peer-reviewed journal articles included 22 cross-sectional studies33-38,40,41,43,45-47,49,53,55,57,59,61,63,65-67 and 15 longitudinal studies.32,39,42,44,48,50-52,54,56,58,60,62,64,68 Of the15 peer-reviewed longitudinal studies, 14 were used as the analytic set for outcomes for KQs 1b, 1c, and 2.32,39,42,44,48,50,52,54,56,58,60,62,64,68 Three grey literature reports that provided longitudinal data are also included as part of the analytic set for outcomes for KQs 1b, 1c, and 2, but we did not assess risk of bias for these reports. Eight studies included in the review are international.32,35,37,38,49,57,58,66 Twelve studies focused on populations with dementia.34,35,40,45,47,49,52,53,62,65,68 One study64 focused on PACE (Program for All-Inclusive Care for the Elderly) enrollees. Although there have been numerous studies on PACE, no others met the inclusion criteria for this review by explicitly comparing Home and Community-Based Services (HCBS) recipients and nursing home (NH) residents. Most PACE studies compared PACE enrollees to non-PACE enrollees but did not specify the residence of the individuals (i.e., in home and community settings or in NHs). Thus these studies were not eligible for inclusion in this review.

KQ 1. Benefits and Harms of HCBS Compared With NHs for Older Adults Using LTC

KQ 1a. Similarity of Populations Served by HCBS and NHs

Key Points
  • NH samples were consistently more impaired than HCBS (including assisted living [AL] samples in terms of both physical function and cognition.
  • The picture with mental health and clinical status was mixed.
Detailed Synthesis

Tables 4–15 present cross-sectional data that compare samples of individuals receiving care in NHs, through HCBS, and/or in AL settings at baseline. Tables 4 and 5 contrast the physical function (measured in various ways) in peer-reviewed studies for those receiving HCBS and/or in AL settings with those in NHs. In all cases, NH residents were more disabled than their counterparts.

Table 4. Description of participant physical function: U.S. cross-sectional and longitudinal peer-reviewed studies.

Table 4

Description of participant physical function: U.S. cross-sectional and longitudinal peer-reviewed studies.

Table 5. Description of participant physical function: international cross-sectional and longitudinal peer-reviewed studies.

Table 5

Description of participant physical function: international cross-sectional and longitudinal peer-reviewed studies.

Tables 6 and 7 contrast levels of cognitive functioning of study participants from the peer-reviewed studies. Once again, NH samples were consistently more impaired than HCBS and AL samples.

Table 6. Description of participant cognitive function: U.S. cross-sectional and longitudinal peer-reviewed studies.

Table 6

Description of participant cognitive function: U.S. cross-sectional and longitudinal peer-reviewed studies.

Table 7. Description of participant cognitive function: international cross-sectional and longitudinal peer-reviewed studies.

Table 7

Description of participant cognitive function: international cross-sectional and longitudinal peer-reviewed studies.

The picture with mental health is more mixed. As shown in Tables 8 and 9, three studies35,48,60 found more depression in NH residents than in those receiving HCBS, while one study found more depression in HCBS recipients64 and another study found no difference between the groups.57 Compared with AL residents, more NH residents were depressed in two studies45,54 and fewer in two.40,62 Comparisons of other measures of mental health are also mixed. Three studies38,43,59 found no difference in psychological well-being or life satisfaction between HCBS or AL recipients and NH residents, while two studies42,66 found HCBS or AL recipients to have better psychological well-being.

Table 8. Description of participant mental health: U.S. cross-sectional and longitudinal peer-reviewed studies.

Table 8

Description of participant mental health: U.S. cross-sectional and longitudinal peer-reviewed studies.

Table 9. Description of participant mental health: international cross-sectional and longitudinal peer-reviewed studies.

Table 9

Description of participant mental health: international cross-sectional and longitudinal peer-reviewed studies.

Tables 10 and 11 summarize mixed information on clinical status. In one study,36 the number of conditions was lower for NH residents than for HCBS recipients, while another study found that NH residents had higher case-mix scores than HCBS recipients.61 One study52 found that NH residents had more of three specific conditions than did HCBS recipients, but the differences were not significant. The comparison is likewise mixed with AL. The rate of conditions among NH residents was higher in two studies,36,62 but lower in a third.40

Table 10. Description of participant clinical status: U.S. cross-sectional and longitudinal peer-reviewed studies.

Table 10

Description of participant clinical status: U.S. cross-sectional and longitudinal peer-reviewed studies.

Table 11. Description of participant clinical status: international cross-sectional and longitudinal peer-reviewed studies.

Table 11

Description of participant clinical status: international cross-sectional and longitudinal peer-reviewed studies.

Parallel information from the grey literature reflects a similar pattern. As shown in Table 12, physical function was more impaired in NH residents than in HCBS recipients or AL residents. Likewise, cognitive function was more impaired in NH residents than in those receiving HCBS or in AL (Table 13), although one study showed a similar rate of cognitive impairment between residents in AL and NH.69 The prevalence of depression (Table 14) was higher for NH residents than for HCBS recipients.72 Rates of mental disorders were similar for those receiving care in NHs and HCBS settings.73 As shown in Table 15, the rates of comorbidities were similar for HCBS recipients and NH residents.72 NH residents generally had more comorbidities than those in AL.73

Table 12. Description of participant physical function: U.S. cross-sectional and longitudinal grey literature studies.

Table 12

Description of participant physical function: U.S. cross-sectional and longitudinal grey literature studies.

Table 13. Description of participant cognitive function: U.S. cross-sectional and longitudinal grey literature studies.

Table 13

Description of participant cognitive function: U.S. cross-sectional and longitudinal grey literature studies.

Table 14. Description of participant mental health: U.S. cross-sectional and longitudinal grey literature studies.

Table 14

Description of participant mental health: U.S. cross-sectional and longitudinal grey literature studies.

Table 15. Description of participant clinical status: U.S. cross-sectional and longitudinal grey literature studies.

Table 15

Description of participant clinical status: U.S. cross-sectional and longitudinal grey literature studies.

KQ 1b. Outcomes in Populations Served by HCBS Versus NHs

Key Points
  • Evidence was low strength or insufficient for all of the outcomes examined.
  • Low-strength evidence suggested that the rate of change in physical function did not differ between AL and NH residents over time. Evidence was insufficient for the HCBS versus NH comparison.
  • Low-strength evidence suggested that the rate of change in cognitive function did not differ between AL and NH residents over time. Evidence was insufficient for the HCBS versus NH comparison.
  • Low-strength evidence suggested that the rate of change in mental health did not differ between AL and NH residents over time. Evidence was insufficient for the HCBS versus NH comparison.
  • Low-strength evidence suggested that mortality did not differ between AL and NH residents. Evidence insufficient for the HCBS versus NH comparison.
  • Evidence was insufficient for acute care utilization for the AL versus NH comparison and for the HCBS versus NH comparison.
  • Other outcomes were rarely examined.
  • No studies conducted subgroup analysis.
Detailed Synthesis

Tables 1618 describe the analyses performed and list the outcome domains addressed in the modest collection of longitudinal studies. Outcomes were measured in terms of functional change (physical or cognitive), mental health, mortality, and use of acute care services. Evidence was low strength or insufficient for the outcomes examined in detail.

Table 16. Description of analysis and outcome domains: U.S. longitudinal peer-reviewed studies.

Table 16

Description of analysis and outcome domains: U.S. longitudinal peer-reviewed studies.

Table 17. Description of analysis and outcome domains: international longitudinal peer-reviewed studies.

Table 17

Description of analysis and outcome domains: international longitudinal peer-reviewed studies.

Table 18. Description of analysis and outcome domains: U.S. longitudinal grey literature studies.

Table 18

Description of analysis and outcome domains: U.S. longitudinal grey literature studies.

Table 19 summarizes the results for physical function over time for HCBS recipients or AL residents with NH residents for peer-reviewed studies from the United States. Frytak et al. 200142 found that although AL residents were more functional at the outset of the study, their rate of decline in physical function was the same as that of NH residents. In the study by Pruchno and Rose 2000,54 AL residents maintained better function throughout, but the rate of change in functional ability did not differ between AL and NH residents over time. Sloane et al. 200562 found no significant differences in the mean rate of decline in function between AL and NH residents for the cohort with mild dementia or for the cohort with moderate or severe dementia. These three studies provided low-strength evidence (due to high risk of bias of the studies) that the rate of change in physical function did not differ between AL and NH residents over time. Marek et al. 200548 found that NH residents remained more disabled than HCBS clients over 24 months, but they did not evaluate rate of change in function between NH and HCBS recipients. This study provided insufficient evidence for the HCBS versus NH comparison of change in physical function.

Table 19. Physical function outcomes: U.S. longitudinal peer-reviewed studies.

Table 19

Physical function outcomes: U.S. longitudinal peer-reviewed studies.

Table 20 summarizes the U.S. longitudinal peer-reviewed studies that examined changes in cognitive function. Two studies54,62 showed no significant differences in trends in cognition between AL and NH residents. Due to high risk of bias, these two studies provided low-strength evidence that the rate of change in cognitive function did not differ between AL and NH residents over time. Marek et al. 200548 showed that NH residents remained more impaired than HCBS recipients at most time points, but they did not evaluate rate of change in cognition between the two groups. This high-risk-of-bias study provided insufficient evidence for the HCBS versus NH comparison of change in cognitive function.

Table 20. Cognitive function outcomes: U.S. longitudinal peer-reviewed studies.

Table 20

Cognitive function outcomes: U.S. longitudinal peer-reviewed studies.

Table 21 summarizes the results from the U.S. peer-reviewed studies in terms of mental health outcomes. Two studies54,62 found that changes in depression scores did not differ significantly between AL and NH residents. Frytak et al. 200142 found no difference in psychological well-being scores from the SF-36 between AL and NH residents and no change over time for either group. The three studies provided low-strength evidence that the rate of change in mental health did not differ between AL and NH residents over time. Marek et al. 200548 found that depression was higher for NH residents than for HCBS clients at 6 and 12 months, but not at baseline. The difference disappeared for the 18- and 24-month followup times. However, this study did not analyze the rate of change in depression between HCBS recipients and NH residents. This high-risk-of-bias study provided insufficient evidence for the HCBS versus NH comparison for change in mental health.

Table 21. Mental health outcomes: U.S. longitudinal peer-reviewed studies.

Table 21

Mental health outcomes: U.S. longitudinal peer-reviewed studies.

Two U.S. peer-reviewed studies that compared mortality in residents of AL versus NH (Table 22)54,62 found no differences. Another study found that median survival was lowest for NH residents and highest for PACE enrollees, with HCBS recipients falling between the two.64 Two peer-reviewed international studies (Table 23) showed that individuals in nursing homes had a higher risk of mortality.50,58 A study from the grey literature (Table 24) showed the highest odds of mortality for NH residents, followed by those in HCBS and AL, compared with individuals not yet receiving long-term care (LTC).71 The two published studies from the United States provided low-strength evidence (due to high risk of bias of the studies) that mortality did not differ between AL and NH residents; evidence was insufficient for mortality for the HCBS versus NH comparison.

Table 22. Mortality outcomes: U.S. longitudinal peer-reviewed studies.

Table 22

Mortality outcomes: U.S. longitudinal peer-reviewed studies.

Table 23. Mortality outcomes: international longitudinal peer-reviewed studies.

Table 23

Mortality outcomes: international longitudinal peer-reviewed studies.

Table 24. Mortality outcomes: U.S. longitudinal grey literature studies.

Table 24

Mortality outcomes: U.S. longitudinal grey literature studies.

As shown in Table 25, two peer-reviewed studies from the United States examined the use of acute care services. Mitchell et al. found no significant difference in the odds of hospitalization for HCBS recipients compared with NH residents.52 NH residents, however, had lower odds of using hospice services. Sloane et al. found that among persons with mild dementia, those in AL used hospitals more than those in NHs,62 but they found no difference for those with moderate or severe dementia. Due to the high risk of bias of both studies, evidence was insufficient for the use of acute care services for AL versus NH residents and for HCBS recipients versus NH residents.

Table 25. Utilization outcomes: U.S. longitudinal peer-reviewed studies.

Table 25

Utilization outcomes: U.S. longitudinal peer-reviewed studies.

Other outcomes are presented in Appendix G, Tables G1 and G2.

No studies conducted subgroup analysis to determine whether outcomes differed between subgroups of HCBS recipients and NH residents for characteristics such as socioeconomic status.

KQ 1c. Harms in Populations Served by HCBS Versus NHs

Key Points
  • The strength of evidence for harms was low for the HCBS versus NH comparison, and it was insufficient for AL versus NH.
  • Harms differed between HCBS recipients and NH residents. HCBS recipients experienced higher rates of some harms and NH residents experienced higher rates of other harms. Evidence was insufficient for harms for the AL versus NH comparison.
Detailed Synthesis

Three U.S. peer-reviewed studies measured harms (Table 26). The strength of evidence for harms was low for the HCBS versus NH comparison and insufficient for AL versus NH. Rigler et al. examined inappropriate medication use;56 the overall unadjusted rate was higher for HCBS recipients than for NH residents. Mitchell et al. compared end-of-life care for older individuals with advanced dementia.52 Compared with HCBS recipients, NH residents had lower odds of pain, shortness of breath, and use of anti-anxiety medication, but higher odds of having a feeding tube, experiencing pneumonia, having a pressure ulcer, or using oxygen therapy. These two studies provided low-strength evidence that harms differed between HCBS recipients and NH residents, since HCBS recipients experienced higher rates of some harms and NH residents experienced higher rates of other harms. Another study found that neither pain nor discomfort differed at baseline or over time between individuals in AL and NHs.42 This high-risk-of-bias study offered insufficient evidence for the AL versus NH comparison.

Table 26. Harms: U.S. longitudinal peer-reviewed studies.

Table 26

Harms: U.S. longitudinal peer-reviewed studies.

KQ 2. Costs for Populations Served by HCBS Versus NHs

Key Points

  • The strength of evidence for costs (or expenditures) for HCBS versus NHs was insufficient.
  • Although Medicaid expenditures for HCBS recipients and NH residents were analyzed, other program and individual expenditures were not examined. Costs related to family burden were not addressed.

Detailed Synthesis

Table 27 compares data on expenditures from one U.S. peer-reviewed study.60 This study analyzed individuals' Medicaid expenditures and found that HCBS was less expensive. However, the study did not analyze other program and individual expenditures. This high-risk-of-bias study provided insufficient evidence on costs for HCBS recipients compared with NH residents. No studies analyzed costs for AL versus NH residents.

Table 27. Expenditures: U.S. longitudinal peer-reviewed studies.

Table 27

Expenditures: U.S. longitudinal peer-reviewed studies.

Table 28 presents expenditure data from grey literature studies. One study on private LTC insurance users compared monthly expenditures on LTC for HCBS recipients and residents of AL and NHs. It found that expenditures were highest for NH residents.71 Three studies measured annual per-client Medicaid expenditures and found that total Medicaid expenditures were higher for NH residents compared with HCBS recipients or AL residents.69,70 These studies did not include other program or individual expenditures, nor did they address costs related to family burden.

Table 28. Expenditures: U.S. longitudinal grey literature studies.

Table 28

Expenditures: U.S. longitudinal grey literature studies.

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (984K)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...