Adverse events related to drugs and drug withdrawal in nursing home residents.

OBJECTIVE: To (1) develop and standardize explicit criteria to link clinical adverse events to drug withdrawal, (2) determine the incidence and severity of Adverse Drug Events (ADEs) and Adverse Drug Withdrawal Events (ADWEs) in a nursing home population, and (3) establish the contribution of demographic, clinical, and functional characteristics to ADEs and ADWEs.

DESIGN: Retrospective record review of an admission cohort.

SETTING AND SUBJECTS: Consecutive admissions of residents of an academic Veterans Affairs nursing home with available records and lengths of stay > 30 days (n = 175). Subjects were 96% men, aged 70 +/- 12 years, and took 7.0 +/- 3.4 medications.

METHODS: We applied standardized algorithms to determine incidence, probability, and severity of ADEs and ADWEs. Multiple regression techniques were used to identify factors associated with frequency and risk of events.

RESULTS: Ninety five residents experienced 201 ADEs. Twelve required hospitalization or prolonged hospitalization, and one resident died. Sixty two persons had 94 ADWEs. None were associated with death and one with hospitalization. The four most commonly prescribed drug classes accounted for 72% of ADEs and 80% of ADWEs. Results of multivariate analyses showed common risk factors for both ADEs and ADWEs: number of diagnoses, number of medications, and hospitalization during the nursing home stay.

CONCLUSIONS: ADEs and ADWEs were common in nursing home residents in this Veteran's Affairs setting. Explicit criteria developed and applied in this study should be applied prospectively in other settings, both to further define risk of drug discontinuation and to assist in development of specific drug discontinuation guidelines.