How drug safety monitoring works when paediatric trials are small
Rare adverse events rarely show up in trials of a few hundred children. Post-marketing surveillance has to do the heavy lifting.
In brief
Rare side effects seldom appear in small paediatric trials. How spontaneous reporting, registries and FDA reviews monitor children’s drug safety.
- An adverse reaction affecting one child in ten thousand will almost never appear in a pre-approval trial.
- Spontaneous reporting finds signals but cannot measure how common they are.
- Off-label use makes paediatric signals slower to detect and harder to attribute.
A paediatric trial may enrol a few dozen or a few hundred patients. An adverse reaction that occurs in one child in ten thousand will almost certainly not appear before approval. That puts more weight on what happens after a medicine reaches the market.
Spontaneous reporting
Systems such as the FDA Adverse Event Reporting System and the EU’s EudraVigilance collect reports from clinicians, patients and companies. In the UK, the MHRA’s Yellow Card scheme plays the same role.
These systems are good at raising signals: an unexpected cluster of liver injury, say, or seizures after a particular product. They are poor at measuring how often something happens, because nobody knows how many children were exposed. Under-reporting is substantial and uneven.
Off-label use complicates the picture
Because so much paediatric prescribing is off-label, many children receive medicines whose labels say nothing about their age group. Reports may not record the child’s age or weight. Safety signals in those populations can be harder to attribute and slower to reach the label.
Dedicated paediatric review
In the US, the FDA’s Pediatric Advisory Committee reviews adverse event data for products after paediatric labelling changes made under the Best Pharmaceuticals for Children Act and the Pediatric Research Equity Act. In the EU, risk management plans must address paediatric use, and post-authorisation safety studies can be required as a condition of approval.
Better tools
Registries, electronic health records and claims databases increasingly supplement spontaneous reports. They can provide denominators, so that a signal can be turned into a rate, and comparison groups, so that it can be judged against background risk. Children’s hospital networks that pool data have made it possible to study rare events that no single centre would see.
