Deprescribing and Polypharmacy for Medicare Beneficiaries Under Guardianship in Long-Term Care Facilities

Abstract Local judicial courts vary in the amount of supervision they provide guardians, which makes the practice of guardianship uneven. To begin to address the evidence gap to inform best practices for persons under guardianship care, this study examines the issues of polypharmacy and prescribing patterns for four therapeutic classes most commonly targeted for deprescribing for older adults. The Medicare Current Beneficiary Survey (MCBS) for 2015 and 2016 is used to examine facility-dwelling Medicare beneficiaries under guardianship compared with those that are not. Logistic regression is used to examine association of polypharmacy outcomes and guardianship care controlling for patient and facility characteristics. Statistical models are adjusted using Fay’s Method with replicate weights for the MCBS complex survey design. Approximately 12% of the facility-dwelling Medicare population in 2015 and 2016 are persons under guardianship care. Persons under guardianship were more likely to have polypharmacy or to be prescribed 5 or more medications (Odd Ratio (OR)=1.168, 95% Confidence Interval (CI)=1.156 to 1.180, p<0.001) than facility-dwelling Medicare beneficiaries not under guardianship care. Medicare beneficiaries under guardianship were more likely to be prescribed PPIs (OR=1.229, 95% CI=1.222 to 1.237, p<0.001) or antipsychotic medications (OR=1.240, 95% CI=1.232 to 1.247, p<0.001) but less likely to be prescribed benzodiazepines (OR=0.920, 95% CI=0.913 to 0.927, p<0.001) or antihyperglycemics (OR-0.726, 95% CI=0.721 to 0.731, p<0.001). Medical decision support services, such as guardianship care, are increasing in importance as shared decision making between patients and physicians evolves to address polypharmacy and deprescribing for older adults.


CORRELATES OF SKILLED NURSING FACILITY (SNF) PERFORMANCE IN THE SNF VALUE-BASED PURCHASING PROGRAM
Dan Andersen, 1 Sherly Binu, 2 and Mike Sacca, 3 1. RELI Group,Sykesville,Maryland,United States,2. RELI Group,Baltimore,Maryland,United States,3. RELI Group,Baltimore,United States We examined the results of the 2020 skilled nursing facility (SNF) value-based purchasing (SNF VBP) program to identify correlates and potential drivers of SNF performance in this program. The SNF VBP program provides incentive payments to SNFs based on their performance on a risk-adjusted hospital readmission measure (i.e., the rate at which SNF residents are admitted back to the hospital within 30-days of being admitted to the SNF). SNFs are assessed on this measure for both improvement compared to their historical baseline and overall achievement compared to their peers. All SNFs that are covered under Medicare's prospective payment system are included in the SNF VBP program. We performed analyses to assess the correlation between individual SNFs' performance in the 2020 SNF VBP (n=15,201), which is based on actual performance in fiscal year 2018, with contemporaneous matched data related to SNF health inspection results, staffing, and performance on quality measures (these data form the basis of the five-star quality rating system on the Nursing Home Compare website). We also examined longitudinal trends in these non-SNF VBP program variables and their association with changes in SNF performance in the SNF VBP program. We controlled for important SNF-specific factors (e.g., for-profit status, connected to a hospital). We found strong contemporaneous and longitudinal associations between SNF VBP program performance and some, but not all, of these factors. Our findings are supported by decades of empirical research in SNF quality and highlight potential policy alternatives that could further incentivize high quality care in SNFs.

DEPRESCRIBING AND POLYPHARMACY FOR MEDICARE BENEFICIARIES UNDER GUARDIANSHIP IN LONG-TERM CARE FACILITIES Tami Swenson, Des Moines University, Des Moines, Iowa, United States
Local judicial courts vary in the amount of supervision they provide guardians, which makes the practice of guardianship uneven. To begin to address the evidence gap to inform best practices for persons under guardianship care, this study examines the issues of polypharmacy and prescribing patterns for four therapeutic classes most commonly targeted for deprescribing for older adults. The Medicare Current Beneficiary Survey (MCBS) for 2015 and 2016 is used to examine facility-dwelling Medicare beneficiaries under guardianship compared with those that are not. Logistic regression is used to examine association of polypharmacy outcomes and guardianship care controlling for patient and facility characteristics. Statistical models are adjusted using Fay's Method with replicate weights for the MCBS complex survey design. Approximately 12% of the facility-dwelling Medicare population in 2015 and 2016 are persons under guardianship care. Persons under guardianship were more likely to have polypharmacy or to be prescribed 5 or more medications (Odd Ratio (OR)=1.168, 95% Confidence Interval (CI)=1.156 to 1.180, p<0.001) than facility-dwelling Medicare beneficiaries not under guardianship care. Medicare beneficiaries under guardianship were more likely to be prescribed PPIs (OR=1.229, 95% CI=1.222 to 1.237, p<0.001) or antipsychotic medications (OR=1.240, 95% CI=1.232 to 1.247, p<0.001) but less likely to be prescribed benzodiazepines (OR=0.920, 95% CI=0.913 to 0.927, p<0.001) or antihyperglycemics (OR-0.726, 95% CI=0.721 to 0.731, p<0.001). Medical decision support services, such as guardianship care, are increasing in importance as shared decision making between patients and physicians evolves to address polypharmacy and deprescribing for older adults.

DOES LONG-TERM CARE SUBSTITUTE FOR HOSPITAL CARE? EVIDENCE FROM THE CHINA HEALTH AND RETIREMENT LONGITUDINAL SURVEY Wei Yang, King's College London, London, United Kingdom
Publicly funded long-term care (LTC) support is shrinking in many countries despite continuing increases in the number of older people who need care. Evidence has shown that the LTC services have an impact on the efficient use of the resources in the health care sector by reducing rates of admission and associated costs through assisting older people with daily living. This paper seeks to examine whether and to what extent these services are substitutes. We use a fixedeffect instrumental variable GMM model to predict the effect of long-term care services on the utilisation of outpatient and inpatient care services. Data are drawn from China Health and Retirement Longitudinal Survey 2011, 2013 and 2015. Our findings suggest that LTC significantly reduces the use of outpatient care but not inpatient care. We have also found LTC use is concentrated among the rich, but the substitution effects are stronger among the poor compared to the rich. This indicates that the poor would benefit more from subsided LTC services. We urge the Chinese government to take action to develop its formal LTC system and to channel more resources to its LTC system, which will benefit the older population in general, and the poor in particular. This study examined the effect of registered nurse (RN) staffing level, work environment, and education on adverse events experienced by residents in nursing homes. A cross-sectional study was conducted with 216 RNs working in nursing homes who were selected using random stratified sampling by location and bed size. Self-reported questionnaires regarding staffing level, work environment, education level, adverse events, and nurse characteristics were administered. Data from the National Health Insurance Service were used to describe nursing home characteristics. Both multiple and multinomial logistic regressions were used to control for the characteristics of nurses and nursing homes, and investigate the effects of nursing staffing level (number of older adults assigned to a nurse), work environment (Practice Environment Scale of the Nursing Working Index), and level of nursing education on the adverse events experienced by residents. An increase of one resident per RN was significantly associated with a higher incidence of pressure ulcers (OR= 1.019, 95% CI=1.004-1.035). Poor work environment increased the incidence of adverse events such as pressure ulcers (OR= 3.732, 95% CI=1.155-12.056) and sepsis (OR=3.871, 95%CI=1.086-13.800). Compared to RNs with a baccalaureate or higher, RNs with diplomas reported increased incidence rates of pressure ulcers (OR=2.772, 95%CI= 1.173-6.549). RN staffing, work environment, and education level affect the incidence of pressure ulcers, and the work environment affects the incidence of sepsis among residents in nursing homes. Policy-wise, improving the level of nurse staffing, nursing work environment, and nursing education will improve health outcomes of residents.

EXPLORING THE COMPLEXITIES OF QUALITY MEASUREMENT IN ASSISTED LIVING IN WASHINGTON STATE
Carolyn Ham, 1 Cathy McAvoy, 2 Roger Gantz, 2 and Lindsay Schwartz, 3 1. Washington State Department of Health,Olympia,Washington,United States,2. Washington State Department of Social and Health Services,Olympia,Washington,United States,3. American Health Care Association/ National Center for Assisted Living,Pittsboro,North Carolina,United States In Assisted Living (AL) choice and quality of life are highly valued by residents. Prior research has found many interrelated components contribute to AL residents' experiences of quality. There are 543 licensed AL facilities in Washington state, with 33,830 licensed beds; in 2019, resident cost per year ranged from $44,000 to $78,000. In 2018, the Washington Legislature identified a need for consumer access to unbiased information on AL quality and directed the Department of Social and Health Services (DSHS) to form a workgroup on quality measures. The workgroup consists of representatives from state agencies, AL provider associations, advocacy organizations, AL providers and AL residents. To inform the workgroup, DSHS conducted a study of AL quality measures. Five statewide programs were identified and analyzed in 10 categories. The WA workgroup then assessed a series of measures across six domains: community participation and quality of life; consumer satisfaction; equity, diversity and inclusivity; informed choice and decision making; person-centered planning; and, resident safety. The workgroup assessed and chose to adopt CoreQ, a satisfaction measure developed by the American Health Care Association/National Center for Assisted Living (AHCA/ NCAL) and endorsed by the National Quality Forum (NQF). In evaluating potential measures of quality, workgroup members weighed regulatory, industry, provider, consumer and family perspectives. Outcomes of this process include assessment of AL quality measurement in other states, selected measures for Washington and key insights into relative prioritization of quality domains by different stakeholder types.

LONG-TERM CARE MARKET TREND AND PATTERNS OF CAREGIVING IN THE UNITED STATES Rashmita Basu, East Carolina University, Greenville, North Carolina, United States
Objectives: The current study aims to: 1) identify patterns of the use of long-term care services and supports (LTSS) among community-dwelling individuals; 2) examine if the changes in supply of formal care predict the use of informal care (IC). Methods: Linking the market supply of formal LTSS to individual level Health and Retirement Survey data from (N=7,781), descriptive and regression analysis were