LONGITUDINAL PRESCRIBING PATTERNS IN A NURSING-HOME POPULATION

Objective: This study documents the patient characteristics associated with prescribed medications on entry to a nursing home and the change in prescribing patterns after 3 months. Design: One-year admission cohort. Setting: Three university-affiliated community nursing homes in Albuquerque, NM. Patients: All new admissions (n = 81) to a University geriatrics team, covering intermediate and skilled levels of care during 1 year (July 1, 1988-July 1, 1989). Methods: Outcome measures were scheduled and as-needed (PRN) medications prescribed at entry and 3 months. Data collected at entry included patient demographics, activities of daily living index, mental status score, and medical diagnoses. Results: Older persons were prescribed fewer scheduled medications than younger ones, and women fewer than men. There was a positive association between the number of diagnoses and the number of scheduled medications (r = 0.25, P = 0.02). No associations were found between medications prescribed and mental status or functional level. There were no associations between as-needed (PRN) medications and any of the variables studied. Overall, there was a significant increase in the average total number of medications prescribed between admission (4.7) and 3 months (6.2). This was due to an increase in the number of PRN medications from 1.3 at admission to 3.0 at 3 months (P < 0.001). Conclusions: Measuring medications at consistent points in a person's nursing home stay may be more informative than using cross-sectional sampling. Future studies on medications in nursing home populations should distinguish between PRN and scheduled medications because medication prescribing patterns may be different in these categories.