Re-prescription of penicillin in patients who have had allergic-like reactions: Investigation using a large electronic medical record

Abstract

Rationale To determine the frequency of re-prescription of penicillin to penicillin-allergic individuals and the risk of a second reaction in those who had a previous reaction. Methods A retrospective cohort study was conducted within the UK General Practice Research Database, an electronic medical record. All patients who had received a prescription for penicillin, 1987 through 9/2001, and those who had received a second prescription for a penicillin at least 60 days later, were selected. Patients with events consistent with a hypersensitivity (allergic-like) reaction within 30 days after the first prescription were compared to those without such events. Results 3,375,162 patients received at least one prescription for penicillin. Of 6212(0.18%) who experienced an allergic-like reaction to the initial penicillin prescription, 48.5% were given a second prescription for penicillin, compared with 59.8% of those without an initial allergic reaction (OR 0.6, 95% CI 0.6, 0.7). 2,017,957 patients were given two or more prescriptions for penicillin. 3014 (0.15%) experienced an allergic-like reaction with the first prescription and 57 (1.89%) of those had another reaction after the second prescription. The unadjusted OR of an allergic-like reaction with the second prescription for those who experienced an allergic-like reaction following the first prescription, compared with those who had no initial reaction, was 11.2 (95% CI 8.6, 14.6). Adjusting for confounding had no substantive effect on this result. Conclusions The risk of an allergic-like reaction to penicillin is markedly increased in those who have had a prior event. Re-prescription of penicillin to such patients is more frequent than anticipated.