[Detection and codification of adverse drug reactions in a mother-child university hospital].

INTRODUCTION: In view of the underdeclaration of adverse drug reactions (ADRs), there is a need for optimizing their reporting. To improve our declaration, a pharmacovigilance program was started in 2006 at our center and measures were progressively started (weekly patient rounds, monthly report, etc.). The objectives of the study were to describe and compare the annual average incidence of the ADRs coded by the medical archivists from 1 April 1989 to 31 March 2010 (period 1) and from 1 April 2010 to 31 March 2015 (period 2), to determine the therapeutic classes of drugs associated with ADRs according to age and determine the percentage of ADRs that are declared to Health Canada.

METHODS: This was a nonexperimental, descriptive, and retrospective study in a mother-child university hospital. All the ADRs that arose with patients less than 18 years old during period 2 were included. The results of a previous study, conducted during period 1 were used. A Student t-test was used to compare the mean annual incidence and the average number of ADRs per year between the two periods.

RESULTS: The average number of ADRs per year was 225.7±65.1 ADRs during period 1 versus 429.6±52.1 ADRs during period 2 (P<0.01). The mean annual incidence of the ADRs was 1.64%±0.41% during period 1 versus 3.17%±0.43% during period 2 (P<0.01). Throughout the study period, the mean annual gravity and mortality index were stable. The ADRs caused by antibiotics/antifungals, anticancer agents, opiates, and corticoids accounted for more than 50% of the total ADRs from 1989-1990 to 2014-2015. Only 37% of the ADRs were declared to Health Canada during period 1 versus 41% during period 2.

CONCLUSION: This study highlighted a progressive increase in the codification of the ADRs in the pediatric population from 1989-1990 to 2014-2015. No causal effect could be obtained between the measures that we implemented in 2006 and the improvement seen with the ADR codification, but we can reasonably suspect a positive impact.