LOOKING UPSTREAM: THE ASSOCIATION BETWEEN AFFORDABLE HOUSING, NURSING HOME USE, AND UNMET CARE NEEDS IN THE COMMUNITY

Abstract In the U.S., population aging is coinciding with a growing affordable housing crisis. Evidence suggests that housing security contributes to health, but less is known about how affordable housing affects aging in place. We use a nationally representative sample (n=5,117) of older community-dwelling Medicare beneficiaries from the 2015 National Health and Aging Trends Study to test the association between housing cost burden (HCB) and moving to a nursing home, death, or remaining in the community by 2017. Among 2017 community-stayers (n=4,836), we also test the association between HCB and unmet care need, defined as experiencing a consequence related to 12 mobility (e.g., stayed in bed), self-care (e.g., skipped meals) and household (e.g., no clean laundry) activities. HCB is the proportion of income spent on rent or mortgage: low (<30%), moderate (30-50%), severe (≥50%), or home paid off (referent). Among nursing home movers, 26% had moderate or severe HCB in 2015 compared to 16% of community-stayers. Informed by the person-environment fit perspective, weighted stepwise regression models (multinomial and logistic) adjust for race, age, sex (Model 1), self-rated health, probable dementia (Model 2), living with others and high income (Model 3). Severe HCB is significantly associated with nursing home entry (RRR=2.66, SE=0.89) and this association is only partially mediated by health factors (RRR=2.16, SE=0.72) and resources (RRR=1.95, SE=0.64). Among community-stayers, severe HCB is significantly associated with unmet care need across all models. This study suggests that affordable housing is an important protective factor for older adults to age well in the community.

less available in rural regions of the country than in more urban settings. This paper reports on a needs assessment for an ADS program in a small city, which serves as a health and human services hub for a large rural area; a particular focus of the study was to assess the feasibility and interest in intergenerational programming. Family caregivers were surveyed (n = 84) about their use and knowledge of and interest in ADS. Less than one in five respondents were using or had ever used ADS. Cost (20%) and ignorance of such programs (20%) were primary reasons for not using ADS; reduction of stress was the most frequently cited reason for using ADS (73%). Ten in-person interviews were conducted with ADS program directors and service providers who refer clients to ADS. Funding issues emerged as the key challenge given lack of private insurance coverage and poor reimbursement levels from public insurance programs. Challenges around transportation, stigma, and marketing of services also surfaced in the interviews. Nonetheless, all ten informants spoke of the positive impact of ADS for both consumers and their caregivers, and generally endorsed intergenerational activities, though with caveats. Implications will be discussed, including the need for greater financial support for this valuable aspect of our long-term supports and services system.

INS, OUTS, AND UNINTENDED CONSEQUENCES OF THE CMS NURSING HOME QUALITY MEASURES
Heather Davila, 1 Tetyana P. Shippee, 2 Young Shin Park, 2 Daniel Brauner, 3 and R Tamara Konetzka 3 , 1. VA Boston Healthcare System,Boston,Massachusetts,United States,2. University of Minnesota,Minneapolis,Minnesota,United States,3. University of Chicago,Chicago,Illinois,United States Ongoing concerns about the quality of care provided to nursing home (NH) residents have led the federal government to develop quality measures (QMs) for NHs. Many of these QMs are included in the NH 5-star ratings and reported online via Nursing Home Compare. However, we know little about how NH providers view the QMs, challenges they experience in addressing the measures, and strategies they use to achieve better scores. As part of a broader mixed-methods study to understand how NHs are responding to the 5-star ratings, we conducted interviews with NH personnel (n=110) and observed organizational processes in 12 NHs in three states. We also interviewed policy and industry leaders (n=34) to gain their perspectives. Interviews focused on perceptions of the 5-star ratings, organizational strategies to improve 5-star scores, experiences with the survey/regulatory process, and perceptions and responses to individual QMs. Key themes show that a) NH providers view the QMs as important indicators of quality, but there is variability across indicators; b) providers face challenges related to measurement and definitions for certain QMs (e.g., pain, restraints); and c) there are potentially conflicting goals, where some QMs aim to promote safety at the expense of resident autonomy and quality of life and vice versa. This work provides organizational context to the 5-star measures and the balancing act providers engage in to assess and improve their scores. The findings also identify several potentially unintended consequences related to the QMs, which can adversely affect residents, particularly those with more complex care needs.

THE ROLE OF POLICY CONTEXT IN INDIVIDUAL LONG-TERM CARE DECISIONS IN OECD COUNTRIES Selena Caldera 1 , 1. LBJ School of Public Affairs -University of Texas at Austin, Austin, Texas, United States
While family caregiving of the elderly has long been part of the cultural life of most OECD countries, longer life expectancy combined with low fertility rates has increased the share of the population dependent on current workers and minimized the available population of informal caregivers. The demand for expanded public provision of long-term care (LTC) resulting from this demographic shift prompted reforms in many OECD countries in the 1990s and 2000s. Differences in these reforms provide an opportunity to examine how individual choices between formal and informal care types are shaped by the policy context. I use longitudinal data on elders in three OECD countries, Sweden, Germany, and Japan, to examine LTC decisions under three varied approaches to population aging. The direction of LTC reforms in each country has been shaped by the existing model of care provision and financial constraints. In response to cost pressures, Sweden introduced need-based provision, financial devolution, and market-based approaches to its universal care model. Germany and Japan, in contrast, widely expanded restricted LTC coverage through public LTC insurance models. I use three multinomial logistic models of the LTC decision to test how differing policy schemes influence choices between formal and informal care. Using longitudinal Global Gateway to Aging data for each country, I model the LTC decision in each country as a factor of demographic and need characteristics of the elder experiencing limitations, characteristics of their family, and eligibility for publicly-provided LTC.

LOOKING UPSTREAM: THE ASSOCIATION BETWEEN AFFORDABLE HOUSING, NURSING HOME USE, AND UNMET CARE NEEDS IN THE COMMUNITY
Meghan Jenkins Morales, 1 and Stephanie Robert 1 , 1.

University of Wisconsin-Madison, Madison, Wisconsin, United States
In the U.S., population aging is coinciding with a growing affordable housing crisis. Evidence suggests that housing security contributes to health, but less is known about how affordable housing affects aging in place. We use a nationally representative sample (n=5,117) of older communitydwelling Medicare beneficiaries from the 2015 National Health and Aging Trends Study to test the association between housing cost burden (HCB) and moving to a nursing home, death, or remaining in the community by 2017. Among 2017 community-stayers (n=4,836), we also test the association between HCB and unmet care need, defined as experiencing a consequence related to 12 mobility (e.g., stayed in bed), self-care (e.g., skipped meals) and household (e.g., no clean laundry) activities. HCB is the proportion of income spent on rent or mortgage: low (<30%), moderate (30-50%), severe (≥50%), or home paid off (referent). Among nursing home movers, 26% had moderate or severe HCB in 2015 compared to 16% of community-stayers. Informed by the person-environment fit perspective, weighted stepwise regression models (multinomial and logistic) adjust for race, age, sex (Model 1), self-rated health, probable dementia (Model 2), living with others and high income (Model 3).
Severe HCB is significantly associated with nursing home entry (RRR=2.66, SE=0.89) and this association is only partially mediated by health factors (RRR=2.16, SE=0.72) and resources (RRR=1.95, SE=0.64). Among community-stayers, severe HCB is significantly associated with unmet care need across all models. This study suggests that affordable housing is an important protective factor for older adults to age well in the community.

BARRIERS AND FACILITATORS TO IMPLEMENTING THE BALANCING INCENTIVE PROGRAM (BIP) Lisa Beauregard, 1 and Edward A. Miller 1 , 1. University of Massachusetts Boston, Boston, Massachusetts, United States
The Balancing Incentive Program (BIP) was an optional Medicaid program within the Affordable Care Act. States spending less than 50% of Medicaid long-term services and supports on home and community-based services (HCBS) were eligible for the program and could participate from 2011 to 2015. Participating states received an enhanced federal match in exchange for rebalancing LTSS spending and adopting structural changes to their long-term services and supports system. The purpose of this study is to understand the barriers and facilitators to implementing the BIP in two states. Data was collected through semi-structured interviews with individuals involved in HCBS policy nationally and in Maryland and Texas, including government bureaucrats, consumer advocates, and provider representatives. Findings indicate that factors that facilitated Maryland and Texas' implementation of the BIP were regular communication with the Centers for Medicare and Medicaid Services and their consultants, Mission Analytics Group, merging the BIP with existing HCBS programs, and the substantial amount of funding associated with the program. On the other hand, the short duration of the BIP presented a challenge for states because they needed to enact multiple changes within a limited period of time. In addition, state procurement and contracting processes impeded the speed with which BIP requirements could be met. Key stakeholders, including consumer advocacy and provider organizations, often felt as though their state implemented the BIP with minimal input from interested groups. The findings indicate that the structure of the Balancing Incentive Program as well as internal state factors influenced the program's implementation.

THE IMPACT OF CULTURE CHANGE ON FINANCIAL PERFORMANCE OF HIGH MEDICAID NURSING HOMES
Justin C. Lord, 1 Ganisher K. Davlyatov, 2 Akbar Ghiasi, 2 and Robert Weech-Maldonado 2 , 1. Louisiana State University at Shreveport, Shreveport, United States, 2. University of Alabama at Birmingham, Birmingham, Alabama, United States This study examines the association between culture change artifacts and financial performance among under-resourced nursing homes (70% or higher Medicaid census). Culture change represents a transformational process to become person-centered, through staff and resident empowerment. Cultural artifacts represent the physical evidences that culture change is occurring. In this study, we focus on the workplace (nurse staffing consistent assignments) and leadership (residents engagement) artifacts to assess the relationship between culture change practices and performance. Survey data came from 387 nursing home directors from 2016-2018, merged with secondary data from LTCFocus, Area Health Resource File, and Medicare Cost Reports. The dependent variable consisted of the total profit margin (%), while the independent variables comprised composite scores for leadership (0-25) and workplace artifacts (0-15). Control variables included organizationallevel (ownership, chain affiliation, size, occupancy rate, and Medicare and Medicaid payer mix), and county-level factors (Medicare Advantage penetration, per capita income, educational level, unemployment rate, poverty level and competition). Multivariate regression was used to model the relationship between cultural change artifacts and financial performance. Workplace artifacts in nursing homes were found to be associated with significantly higher profit margin (β = 0.30, p < 0.05), while leadership artifacts were not. Culture change practices aimed at improving nursing staff consistent assignments are associated with better financial performance. Given increasing nursing home market competition and declining resources for high Medicaid nursing homes, facilities with a greater emphasis on workplace culture may be able to perform better financially among these under-resourced facilities.

NAO TAKES A BOW: USING SOCIAL ROBOTS TO ENHANCE THE MOOD OF OLDER ADULTS LIVING IN RESIDENTIAL CARE SETTINGS
Noelle L. Fields, 1 Julienne Greer, 1 and Ling Xu 1 , 1.

University of Texas at Arlington, Arlington, Texas, United States
Social robots have been utilized in research with older adults, however, few studies have integrated participatory arts (e.g. theatre) into social robotic platforms with this population. An interdisciplinary team designed an intervention integrating theater and social robotics with the aim of improving the mood of older adults. A purposive sample of persons age 65 and older (N = 15) participated in this 3-session pilot study that involved a Shakespeare activity using the robot, NAO. Mixed methods included interview questions as well as short survey measures of depression, loneliness, and a simplified face scale for mood pre and post each session. Results from Repeated Measurement Analysis of Variance (ANOVA) showed that face scale scores significantly decreased across six time periods (F = 2.72 (5, 50), p <. 05) and this decrease marginally differed between participants with dementia (M = 2.50, SD = 1.73) and those without dementia (M = 1.62, SD = 0.52). In addition, depression scores marginally significantly decreased after intervention (F = 2.28 (5, 40), p <. 10) and these declines were also marginally significantly different for participants with (M = 0.67, SD = 0.58) or without dementia (M = 0.86, SD = 0.69). Qualitative findings suggest that participants were highly engaged and responsive to the intervention. We discuss the promising aspects of using social robotics as a platform for participatory arts interventions with older adults and offer recommendations for future interdisciplinary studies involving the use of innovative technology in residential care settings.