• Using records of the Saskatchewan Prescription Drug Plan, we determined the incidence of antidepressant use (a marker for depressive symptoms) in patients who received β-blockers or other treatments for chronic diseases (diuretics, antihypertensives, and hypoglycemics) during 1984, but not in the previous 6 months. Antidepressants initiated within 12 months after the study drug were counted. Of the 3218 new β-blocker users, 6.4% received concurrent prescriptions (ie, within 34 days) for an antidepressant and β-blocker. Only 2.8% of the reference group (no study drug use) received an antidepressant. A greater proportion of patients prescribed propranolol (9.5%) received an antidepressant than those prescribed other "lipophilic" (3.9%) or "hydrophilic" (2.5%) β-blockers. Incidence ratios for propranolol revealed the overall risk antidepressant use was 4.8 (95% confidence interval [Cl], 4.1 to 5.5) times that of the reference group and 2.1 (95% Cl, 1.7 to 2.5) times that of all other study drug users. For propranolol, relative risk of antidepressant use (drug/reference group) varied with age and was greatest in the 20- to 39-year-old group (17.2; 95% Cl, 13.7 to 21.5).
(Arch Intern Med. 1990;150:2286-2290)