Abstract

Objective: Many hospitalizations of nursing facility (NF) residents are thought to be avoidable. Unnecessary hospitalizations are burdensome for frail elders and wasteful of health care resources. Policymakers in the US have centered attention on public reporting and payment policies to prevent 30-day re-hospitalization of short-stay, post-acute NF residents with medical diagnoses, such as heart failure, COPD or pneumonia, that indicate preventability. Little attention has been given to hospitalization risk for long-stay NF residents who have unique needs and care requirements that go beyond a limited set of medical diagnoses and that could be addressed through better resource targeting and quality improvement strategies.

Our objectives are to identify risk factors for hospitalization of long-stay NF residents, and describe the application of findings within an innovative care model, the OPTIMISTIC project (Optimizing Patient Transfers, Impacting Medical Quality, and Improving Symptoms: Transforming Institutional Care).

OPTIMISTIC is part of a national demonstration funded by the US Centers for Medicare and Medicaid Services (CMS) to reduce avoidable hospital transfers among long-stay NF residents. The project’s clinical staff of advanced practice nurses (APNs) and registered nurses (RNs) promotes proactive chronic care management, urgent care and robust transitional care services for residents in 19 partnering NFs in central Indiana. A multidisciplinary team of geriatricians and nurses supports clinical staff. Independent evaluators found a significant reduction in hospitalization rates in OPTIMISTIC facilities between 2015 and baseline (2012), both an absolute reduction and compared to matched facilities.

Data and methods: Primary analysis focused on hospitalizations for a retrospective cohort of 1975 nursing facility residents enrolled in OPTIMISTIC, starting at a single time point (October 2014) and followed for 90 days. Health and functional risk factors came from resident Minimum Data Set (MDS) assessments. Random effect logistic regression was used to model hospitalizations as a function of candidate risk factors. Supplemental data came from root cause analysis by project RNs for 1200 hospitalizations from October 2014 to December 2015.

Results: Ten percent of the cohort was hospitalized within 90 days. Statistically significantly risk factors and (Odds Ratios) were: hospitalization in the prior 30 days (4.74) or 31-90 days (2.23); history of cancer (2.52), COPD (1.52), anemia (1.42) or pressure sores (1.74); and receiving dialysis (3.58). Residents enrolled in hospice had a substantially reduced risk (0.12). Root cause analysis of hospitalizations found other contributing factors: dementia, behavioral problems, poly-pharmacy, and poor communication between nursing staff and medical providers.

Application: Findings are being applied in refinement of OPTIMISTIC’s intervention strategies: targeting of high-risk residents for enhanced RN monitoring, APN interventions for hospitalized residents returning to the facility, medication reconciliation and poly-pharmacy review, advanced care planning, and special programs addressing behavioral problems and dementia.

Policy implications: Lessons from models of care, such as OPTIMISTIC, are shaping policy initiatives to reduce avoidable hospitalizations and promote better care for long-stay NF residents. Phase 2 of the demonstration project, which extends and expands on the clinical model, includes enhanced Medicare funding to medical providers and NFs to address changes in condition and prevent unnecessary hospitalizations.