Risk of mortality and nursing home institutionalization after injury

Abstract

Objectives: To evaluate the effects of unintentional injuries on the risks of nursing home institutionalization and mortality in older adults.

Design: A retrospective analysis of data from the Medicare Current Beneficiary Survey, a nationally representative survey of Medicare beneficiaries.

Setting: Noninstitutionalized community residents.

Participants: Older adult panel members (N = 12,031) with continuous Medicare eligibility not enrolled in managed care in a cohort starting between 1998 and 2001.

Measurements: Cox regression and competing risk survival models were estimated using time-varying injury-status dummy variables and control variables for outcomes measured as time until death and institutionalization, respectively.

Results: Almost 4% of persons were institutionalized, 15% died, 14% had a sentinel injury, and 3% had two or more minor nonsentinel injuries within 1-year period. Persons hospitalized for sentinel injury had elevated institutionalization and mortality risks during an injury episode and after the episode ended. Persons receiving outpatient treatment for sentinel injuries had elevated institutionalization risk during injury episodes (subhazard ratio [SHR] = 6.78, 95% confidence interval [CI] = 3.72-12.37) and elevated mortality risk after episodes (hazard ratio [HR] = 1.60, 95% CI = 1.28-2.00). Persons with multiple minor nonsentinel injuries within a year also had elevated mortality (HR = 1.56, 95% CI = 1.15-2.11) and institutionalization (SHR = 3.55, 95% CI = 2.25-5.67) risks.

Conclusion: Mortality and institutionalization risks extend well beyond the acute episode of treatment for sentinel and repeated minor injuries. More research is needed on longer-term health outcomes of injury survivors to inform development of evidence-based quality-of-care indicators.