Our study aimed to explore the impact of different home‐ and community‐based service (HCBS) use patterns on older adults' physical function. The cohort data were drawn from two national datasets, the National Ten‐Year Long‐Term Care Plan 1.0 database and the National Health Insurance Program claims data. Participants were care recipients ages 65 and over, first evaluated and prescribed HCBS from 2010 through 2013 and evaluated again after 6 months (n = 32,392). Latent class analysis was used to identify subgroups with different HCBS use patterns. Multiple regression was used to examine the impact of different HCBS use patterns on change over time in disability related to activities of daily living (ADLs) and instrumental activities of daily living (IADLs). The analysis was stratified by respondents' levels of disability. Four subgroups of HCBS recipients were identified, with patterns of home‐based personal care, home‐based personal care and medical care, home‐based medical care and community care services. Older adults in the Home‐based MpC had significantly more improvement in both ADL (p < 0.05) and IADL (p < 0.001) scores compared with adults in the other three groups, while the community care group regressed the most. In the stratified analysis of the severely disabled, the IADL outcome of the Home‐based MC group was better than the home‐based PC group (p < 0.001). Study findings shed light on the benefits of promoting the use of integrated HCBS that combines personal and medical care, especially for community care services.

Background Long-Term Care (LTC) in Taiwan has been implemented over the last 15 years, focusing on providing home- and community-based services (HCBS). Previous studies have shown that care recipients’ level of unmet needs is associated with adverse health outcomes. This study aimed to identify HCBS use patterns among LTC care recipients and examine the impact of different patterns of HCBS on unmet needs. Methods Surveys of service users were conducted in eight counties. Latent class analysis was used to identify the underlying subgroups of LTC recipients. Multiple regression analysis was used to assess the impact of HCBS patterns and care recipients’ predisposing, enabling, and need factors on the level of unmet needs. Results A total of 952 participants from northern, middle, and southern Taiwan completed the surveys. Three subgroups based on HCBS use were identified, including home-based personal care (HB-PC), home-based personal care and medical care (HB-PC/MC), and community care (CC). Care recipients in the HB-PC group (β = −1.922, p = 0.028) and the CC group (β = −2.177, p = 0.019) had lower unmet needs than those in the HB-PC/MC group. A lower disability level, living in a highly urbanized city, and lower levels of quality of life were associated with higher unmet needs (p &lt; 0.05). Conclusions Our results demonstrating that care recipients with lower disability levels had higher unmet needs may indicate insufficient funding support from Taiwan’s LTC plans. However, multiple service users had high unmet needs, which warrants further investigation.

Background Taiwan’s Long-Term Care (LTC) Plans 1.0 and 2.0 aimed to provide accessible LTC services since 2007. Currently, &gt;40% of the people needing LTC are covered by these plans. Past studies have indicated that high unmet needs among LTC users may lead to adverse outcomes. Thus, whether the LTC services meet the needs of both the care recipients and their family caregivers merits further study. The purpose of this study was to develop a suitable “unmet needs appraisal scale (UNAS) for long-term care service users” and pilot test this tool on older Taiwanese adults. Methods The tool was developed through a systematic literature review. The tool underwent two rounds of the Delphi method with 26 experts, and the tool reached a high consensus among the experts. The UNAS includes 11 items assessing activities that need more assistance, including daily activities, psychological health, and spiritual health. We pilot tested the tool in older Taiwanese adults and examined the tool properties, including reliability and validity. We included short-form 8 as a concurrent validity measure. Result: The survey was conducted in 6 counties, and 255 service user questionnaires were collected. The overall Cronbach's α value was 0.91, indicating good internal consistency. Our confirmatory factor analysis also showed good construct validity [χ2 = 146.514, p &lt; 0.001, CFI = 0.98, RMSEA = 0.06, SRMR = 0.02]. The UNAS scores showed a moderate negative correlation with quality of life scores (r = –0.452, p &lt; 0.001), demonstrating good concurrent validity of the unmet needs scale.