Objective: To assess the preventability of adverse drug reactions (ADRs) leading to hospital admissions and to investigate the feasibility of the use of a standardised preventability scale in clinical practice.
Design: The study was a prospective pharmacovigilance study. All patients more than 15 years old admitted to an emergency department during a period of 4 weeks were included. Characteristics of patients admitted for a suspected ADR (cases) were compared to those admitted for other reasons (controls). Preventability was assessed in two different ways: (i). by using a standardised preventability scale; and (ii). by the assessment of four reviewers without the scale. Results of the two methods were compared.
Patients: In total, 671 patients were admitted to an emergency department during the study period.
Results: Overall, 44 ADRs were identified involving 41 patients. The incidence of hospital admissions for ADRs was 6.1 per 100 admissions (95% CI 4.4-8.3). According to the French causality assessment method, 71% of ADRs were 'possible', 18% were 'plausible' and 11% were 'likely'. Using the standardised preventability scale, one-third of all ADRs were considered as being preventable (9% 'definitely' and 25% 'potentially' preventable). Reviewers found that 54.5% of ADRs were 'preventable'. Discrepancies between the two methods concerned mainly cases defined as not preventable by the scale. In general, reviewers overestimated the preventability of ADR compared with the scale.
Conclusions: These results emphasise that ADRs leading to hospitalisation are frequent, with one-third of them likely to be preventable. Moreover, the risk of ADRs mainly involved a small number of drugs. Our experience suggests that there is a need for further studies to validate the French standardised scale of preventability assessment.