HOSPITAL CARE USE IN OLDER ADULTS: THE ROLE OF PSYCHOLOGICAL AND SOCIAL FACTORS

Abstract Although older people’s health status is the main determinant of healthcare use, there has been little research on how psychosocial factors relate to healthcare utilization. We explored the extent to which psychological and social aspects predict the use of hospital care in an older Swedish population. 2867 people ≥60 years from the Swedish National study on Aging and Care in Kungsholmen (SNAC-K) were followed from baseline (2001-2004) for four years. We created standardized indexes of psychological well-being, and social well-being. Binomial negative mixed models were used to estimate the association of psychological and social indexes with hospital care use (i.e. unplanned hospital admissions [UHA], 30-day readmissions [30DR] and length of stay [LOS]). Individuals with a psychological well-being score above the median had less UHA (IRR 0.43, 95%CI 0.20-0.93) and lower LOS (IRR 0.18, 95% 0.06-0.58), even after full adjustment. High levels of social well-being were also protective for UHA and LOS in the minimally adjusted model, but not after adjusting by life style and personally traits. Relative to individuals with poor well-being on both indexes, those with rich psychological and poor social well-being had reduced hospital care use (IRR 0.44 95%CI 0.24-0.84; IRR 0.23, 95%CI 0.08-0.67, respectively), and even further in those with rich psychological and social well-being (IRR 0.33 95%CI 0.14-0.75; IRR 0.10, 95% 0.02-0.45, respectively). No statistically significant association was found with 30DR. Provided the importance of psychosocial aspects in predicting UHA and LOS, targeting the former could be a strategy for reducing healthcare use and, eventually, costs.

Older adults with disabilities commonly rely on family caregivers' help, yet effects of caregiver factors on patient outcomes are poorly understood. Within this population, dementia is common. Our objective was to evaluate the association between caregiver factors and risk of hospitalization in disabled older adults with and without dementia. We examined 2,589 community-living older adults with mobility/self-care disability and their primary family caregiver in four waves of the National Long-Term Care Survey and National Health and Aging Trends Study. We used Cox proportional hazards models to examine risk of one-year, Medicare claims-derived, all-cause hospitalization as a function of caregiver factors, adjusting for older adult characteristics (sociodemographics, comorbidities, healthcare utilization) and survey year, considering dementia a characteristic of interest. Among disabled older adults, 38% were hospitalized over one year, and 31% had probable dementia. Hospitalization rates were similar for older adults with and without dementia (39.5% and 37.3% respectively); dementia was not associated with hospitalization risk ( Although older people's health status is the main determinant of healthcare use, there has been little research on how psychosocial factors relate to healthcare utilization. We explored the extent to which psychological and social aspects predict the use of hospital care in an older Swedish population. 2867 people ≥60 years from the Swedish National study on Aging and Care in Kungsholmen (SNAC-K) were followed from baseline (2001)(2002)(2003)(2004) for four years. We created standardized indexes of psychological well-being, and social well-being. Binomial negative mixed models were used to estimate the association of psychological and social indexes with hospital care use (i.e. unplanned hospital admissions [UHA], 30-day readmissions [30DR] and length of stay [LOS]). Individuals with a psychological well-being score above the median had less UHA (IRR 0.43, 95%CI 0.20-0.93) and lower LOS (IRR 0.18, 95% 0.06-0.58), even after full adjustment. High levels of social well-being were also protective for UHA and LOS in the minimally adjusted model, but not after adjusting by life style and personally traits. Relative Innovation in Aging, 2019, Vol. 3, No. S1 to individuals with poor well-being on both indexes, those with rich psychological and poor social well-being had reduced hospital care use (IRR 0.44 95%CI 0.24-0.84; IRR 0.23, 95%CI 0.08-0.67, respectively), and even further in those with rich psychological and social well-being (IRR 0.33 95%CI 0.14-0.75; IRR 0.10, 95% 0.02-0.45, respectively). No statistically significant association was found with 30DR. Provided the importance of psychosocial aspects in predicting UHA and LOS, targeting the former could be a strategy for reducing healthcare use and, eventually, costs.

IMPLEMENTING A GERIATRIC FRACTURE PROGRAM IN A PLURALISTIC ENVIRONMENT REDUCES LENGTH OF STAY AND TIME TO SURGERY
Sonja L. Rosen, 1 Kathy Breda, 1 Carol Lin, 1 Jeanne Black, 1 Jae Lee, 1 Aaron Chiang, 1 Mark Vrahas, 1 and Brad Rosen 1 ,

Cedars-Sinai Medical Center, Los Angeles, California, United States
Geriatric-orthopaedic co-management models have been demonstrated to improve patient outcomes, but are typically implemented in closed, non-pluralistic medial systems. The Cedars-Sinai Geriatric Fracture Program (GFP) was developed through collaboration amongst a multi-disciplinary group. Cedars-Sinai is an academic medical center with a pluralistic medical staff that includes faculty, several hospitalist groups, and private practitioners. The GFP was introduced in July 2018 as a quality improvement pilot to provide standardized treatment for geriatric fracture patients. We hypothesized GFP enrollment would reduce time to surgery (TTS) and length of stay (LOS). Geriatric fracture patients were prospectively enrolled from July -December 2018. The Wilcoxon Rank-Sum test was used to compare TTS and LOS between the two patient groups. A p < 0.05 was considered significant. 190 operative fractures in patients over 65 years-old were prospectively followed.56 (30%) were enrolled in the GFP, 54 (28%) were admitted to other hospitalist groups (OH), and 80 (42%) were managed by their primary care physician (PCP). There were no demographic differences between groups. Patients enrolled in the GFP had a significantly shorter LOS compared to the OH and PCP groups (4 days v 5 days v 5 days, p = 0.039) as well as a significantly shorter TTS (19.7hrs v 22.4 hrs vs 23.3 hrs, p = 0.037). Our data shows that a multi-disciplinary geriatric fracture program can be successfully implemented in a complex pluralistic environment resulting in improved patient metrics. Adherence to evidence-based protocols and close multidisciplinary teamwork are critical to program success.

IN-HOSPITAL NURSE CARE CONTINUITY: DOES IT MATTER?
Orly Tonkikh, 1 Anna Zisberg, 2 and Efrat Shadmi 1 , 1. University of Haifa,Haifa,Israel,2. The Cheryl Spencer Department of Nursing Faculty of Social Welfare and Health Science,University of Haifa,Israel,Haifa,Isreal,Israel In-hospital cognitive decline affects up to 40% of hospitalized older adults and is associated with posthospitalization worsening of medical and functional status. Studies pointed to the substantial role of the interpersonal relationship between older adults with cognitive impairment and the nurses who care for them. We investigated the association between nursing interpersonal continuity and cognitive outcomes in a cohort of 646 older adults aged 70 or older admitted to internal units for non-disabling conditions. Cognitive decline was defined as at least one point decline in the Short Portable Mental Status Questionnaire from at admission to discharge assessments. Nursing interpersonal continuity was measured using continuity of care index (CoC). CoC assesses the extent of different nurses assigned to take care of each patient during the hospital stay (2 shifts per day) and ranges from 0 (none of the nurses is the same) to 0.4 (highest feasible score according to full time standard shift plan and length of stay (LOS)). Multivariate logistic regression showed that achieving 25% of the highest feasible in-hospital nursing CoC was associated with lower odds of cognitive decline (OR=0.67, 95% CI=0.47-0.97), controlling for age, sex, premorbid activities of daily living status, at admission cognitive status, comorbidities, severity of illness and LOS. This study shows that in-hospital nursing continuity is negatively associated with older adults' cognitive decline, even in low-continuity levels. Future studies should investigate in-hospital continuity patterns and interventions maintaining continuity in larger and more heterogenic samples.

IT'S THE LITTLE THINGS THAT COUNT: IMPLICATIONS OF DAILY EXPERIENCES FOR WELL-BEING AND BIOLOGICAL INDICATORS
Chair: Kira S. Birditt, Institute for Social Research, University of Michigan, Ann Arbor, Michigan, United States Middle-age and older adults vary widely in their physical health. This symposium describes studies that identify diverse daily experiences that account variation in health using multiple indicators of well-being (self-reported, biological). Fingerman et al. assessed daily TV viewing among older adults and found that more frequent television watching was associated with poorer physical health, worse health behaviors, and less energy expenditure (via actical watch). Leger et al., examined links between daily affect and sleep. Greater variability in daily positive affect is associated with fewer hours of sleep and greater morning tiredness even after adjusting for mean levels of affect. Luong et al. examined links between daily stress and affect. Interpersonal stressors were associated with greater affect reactivity than non-interpersonal stressors and links were reduced among older adults. Birditt et al. assessed links between daily social interactions and cardiovascular reactivity. More frequent social interactions and negative social interactions were associated with increased heart rate and links varied by gender and race. Polenick et al. examined links between daily social interactions and salivary DHEA-S (a marker of the stress response). Positive interactions predicted greater DHEA-S over the course of the day and links between negative interactions and DHEA-S varied by age group such that younger individuals appeared to be more reactive. These studies offer important clues regarding how daily experiences get under the skin to influence health and well-being.