Role of pharmacoepidemiology in evaluating prescription drug safety in pediatrics

Abstract

Background: Pharmacoepidemiology is a discipline that applies epidemiological methodologies to the study of drug use, especially the safety of drug use in a large group of people. But this methodology has not been widely utilized to assess drug safety in pediatrics for children aged between 0 and 18 years.

Objectives: (1) To describe current gaps in evaluating pediatric drug safety; (2) to use published examples demonstrating the value of pharmacoepidemiology in detecting and quantifying adverse drug events in pediatrics; (3) to review pharmacoepidemiological study designs, their strengths and limitations, and examples of their applications in the field of drug safety evaluation; and (4) to provide recommendations about selecting an appropriate study design and minimizing a major methodological limitation.

Overview: Only about 20% of approved drugs are licensed for pediatric use, and many prescription drugs are used as 'off label' or 'unlicensed' pediatric treatments. Among this small percentage of licensed pediatric drugs, pre-marketing trials may not adequately address all drug-related safety issues in pediatrics. Consequently, an important gap exists in pediatric drug safety evaluation. Pharmacoepidemiology offers a number of advantages in detecting and quantifying adverse drug events in pediatrics. Each pharmacoepidemiological study design has its own strengths and limitations in the evaluation of drug safety. An appropriate study design depends on the underlying research question being addressed. Confounding by indication, the most serious and challenge form of confounding in pharmacoepidemiological designs, could lead to misleading claims about the relationship between drug use and the development of potential adverse drug events, and should be adequately controlled.

Conclusions: Pharmacoepidemiology is a powerful methodology that offers many advantages to assess pediatric drug safety such as large sample size, long period of follow-up, and good generalizibility. But this methodology should be applied judiciously to maximize its potential.