Risk of a new benzodiazepine prescription in relation to recent hospitalization

OBJECTIVE: To determine if recent hospital admission was associated with ne w outpatient prescribing of benzodiazepines among community-dwelling older people. DESIGN: Nested case-control study using administrative data sets of the pro vincial health insurance board. SETTING: Province of Quebec. PARTICIPANTS: Cases were 4127 community-dwelling older people who were newl y dispensed a benzodiazepine during an g-month period in 1990. Controls wer e 16,486 community-dwelling older people who were dispensed any drug (excep t a benzodiazepine) on the same day as the case-defining index prescription . EXPOSURE AND OUTCOME MEASURES: Admission to an acute care hospital within a 30-day period before a new dispensing of a benzodiazepine. Other variables measured were patient age, gender, number of ambulatory physician visits, healthcare region, Chronic Disease Score (CDS), and use of drugs for depres sion and psychosis. RESULTS: Cases were more than three times as likely as controls to have bee n hospitalized in the 30-day period before the index date (adjusted odds ra tio (OR) 3.09; 95% CI, 2.78-3.45). The use of prescription drugs for physic al health problems modified this association in that cases who used more me dication were also more likely to receive a new benzodiazepine prescription following a recent hospital admission (adjusted OR 4.09; 95% CI, 3.59-4.65 when the CDS was equal to 5 vs adjusted OR 1.96; 95% CI, 1.66-2.31 when th e CDS was equal to 0). CONCLUSIONS: Recent hospitalization confers an increased risk of a new outp atient benzodiazepine prescription among community-dwelling older people in Quebec. Those who use more medication, and who may be more vulnerable to d rug-related adverse events, are more likely to be newly dispensed a benzodi azepine following a recent, acute-care hospital admission.