Underreporting in pharmacovigilance: an intervention for Italian GPs (Emilia-Romagna region)

Abstract

Purpose: Underreporting is a major limitation of spontaneous reporting systems for suspected adverse drug reactions (ADRs). Several interventions to increase the ADR reporting rate have been proposed, but their efficacy remains poorly investigated.

Methods: This was a questionnaire study aimed at assessing the knowledge, attitudes, and behavior of general practitioners (GPs) regarding ADR reporting and at evaluating whether a monthly e-mail-based newsletter on drug safety could affect the rate and the quality of the ADR reports submitted by these GPs. Three local health authorities (LHAs) of the Emilia-Romagna region were chosen on the basis of their ADR reporting rate during the period preceding the study: Rimini (high), Ferrara (average), and Piacenza (low reporting rate). All GPs (n = 737) associated with these three LHAs were recruited. The pooled number of ADR reports sent by GPs in the remaining seven LHAs of the region was used as controls. The study covered a period of 3 years and was divided into: (1) identification of the reasons leading to underreporting through a questionnaire (Phase I); (2) the intervention, i.e., sending a newsletter for a 10-month period (Phase II); (3) evaluation of the intervention outcomes during the 10 months following the period in which the newsletter had been received (Phase III).

Results: Among GPs involved, 22.8 % returned the questionnaire. Over 94 % of the respondents considered the spontaneous reporting of suspected ADRs to be part of their professional obligations, but only 6.5 % had submitted at least one report in the previous 6 months. Following the completion of Phase II, the overall number of reports coming from the LHAs subjected to the intervention rose by 49.2 % compared to 2009, while the number of reports coming from the control LHAs increased by 8.8 %. Rimini and Piacenza showed a 200 % increase in the number of ADR reports submitted by GPs, while the number of ADR reported submitted by the control group decreased by 25.5 %. In 2011, the number of overall ADRs reports from the LHAs subjected to the intervention decreased by 6.8 %; this decrease reached 50.0 % of the GPs. Control HLAs showed an overall decline of 4.3 %, while the total number of ADRs from GPs increased by 63.3 %. Ferrara was excluded from the analysis due to confounding factors.

Conclusions: The periodic e-mail update on the safety of drugs represents an effective and inexpensive way to raise the awareness of GPs on the importance of spontaneous ADR reporting. Since the outcome of the intervention seemed to disappear after the intervention was stopped, there appears to be a need to adopt a policy of regular updates and educational strategies for health professionals.