Improving Adverse Drug Reaction Reporting in Hospitals

Abstract

Spontaneous reporting of adverse drug reactions (ADRs) is fundamental to drug safety surveillance (pharmacovigilance); however, substantial under-reporting exists and is the main limitation of the system. Several factors could favour under-reporting. The aim of this pilot study was to assess the effect of regular visits of a Clinical Research Assistant (CRA) on the improvement of ADR reporting in non-university hospitals. We set up an ADR report collecting system that involved regular visits by a CRA to non-university hospitals, which was similar to a system that already existed in university hospitals in Toulouse, France. Two areas in our region were chosen: Haute Garonne and Gers. We compared firstly the reporting rate (number of reports/number of beds) of total ADRs (i.e. spontaneously reported ADRs plus solicited ADRs collected by the CRA) and secondly, the percentage of serious ADRs reported by non-university hospitals in these two areas, in 2005 (the year prior to CRA visits) and after the start of CRA visits (2006 until the end of December 2008). We also compared the reporting rate of total ADRs in Haute Garonne and Gers non-university hospitals with those reported during the same period with a control group (the Ariège area, which has a similar number of beds to Gers and that was not visited by the CRA). The characteristics of ADRs collected by the CRA were also described. A total of 687 reports were collected by the CRA: 40% were classified as serious, including two deaths. The number of ADRs and the reporting rate increased significantly between 2005 and 2008 in non-university hospitals of Haute-Garonne and Gers, but not in Ariège. In Gers, the reporting rate was 3% in 2005 and 25% in 2008. In Haute-Garonne, the reporting rate was 11% in 2005 and 40% in 2008. The difference between the number of spontaneous and solicited reports also increased. This study shows that regular visits by a CRA increases the number of ADRs collected by a Regional Pharmacovigilance Centre. Another interesting consequence was the rise in spontaneous reporting by healthcare professionals following the set-up of this system. Further assessment of this procedure is necessary for the long-term evaluation of its effectiveness.