by Sean Philpott, Director of the Center for Bioethics and Clinical Leadership
In a study published earlier this week, researchers found that pregnant women who take acetaminophen — a widely used drug found in such over-the-counter painkillers as Tylenol and Excedrin — are at increased risk of having children with hyperkinetic disorders like ADHD (attention deficit hyperactivity disorder).
Although the results of the study are not conclusive, children whose mothers used acetaminophen frequently while pregnant were nearly twice as likely to be diagnosed with ADHD by the time they were 7 years old.
This finding is surprising and troubling. Unlike other painkillers like aspirin and ibuprofen, acetaminophen has largely been considered safe for use during pregnancy. Nearly 50% of women in the United States take this drug at some point during pregnancy.
Pregnant women now face a difficult decision: whether or not to use a drug like acetaminophen when clinically indicated, despite the potential risks to the child. But this is not the first time that women have been faced with this choice. For example, antidepressant use during pregnancy raises a number of concerns, including the possibility of certain congenital defects. But stopping their use is also risky, particularly for the mental and physical wellbeing of the mother.
In an ideal world, pregnant women and their doctors would rely on objective information about the safety of drugs and the risks of forgoing treatment to make appropriate decisions. However, the truth is that we have almost no such data. Physicians and their patients are flying blind when it comes to the use of various medications during pregnancy. Moreover, the US Food and Drug Administration (FDA) — the federal agency tasked with ensuring the safety and effectiveness of drugs sold in the US — can do little to help.
Currently, the FDA classifies drugs according to possibility that they may birth defects if used during pregnancy. Drugs are sorted into one of five categories – A, B, C, D and X – based on the amount of animal and human safety data available.
Category A drugs are those that are safe for use during pregnancy, based on extensive animal and human testing. Category X drugs are those where there is conclusive animal and human data demonstrating fetal risk. Only a handful of drugs fall into category A. Far more drugs fall in category X.
The vast majority of drugs fall into categories B, C and D. For these compounds there may or may not be data from animal models or human studies to suggest that the drug is safe. Use of these drugs by pregnant women is essentially a crapshoot; women and their physicians have little guidance to help them weigh the risks and benefits of these treatments.
The problem is that there is a bit of a Catch-22 with respect to prescription drug use and pregnant women. We would like to collect data on the safety and effectiveness of new drugs during pregnancy, but these women are typically excluded from clinical trials.
For decades, regulatory agencies like the FDA have recommended that pregnant women be excluded from clinical trials out of fear that experimental drugs may pose a risk to the developing fetus. The images of severely deformed children following the thalidomide scandal of the late 1950s and early 1960s still resonate strongly among regulators and researchers alike.
Under existing guidelines, studies in pregnant women can be considered only when drug safety and efficacy has been demonstrated in the general population, when the drug has potentially significant therapeutic benefit to pregnant women, and when drug exposure not expected to pose undue risk to the woman or fetus. Those are very difficult requirements to meet, and there is little incentive for pharmaceutical companies to conduct additional studies in pregnant women once a drug has been approved for general use.
The vast majority of the data on the safety and effectiveness of drugs during pregnancy thus comes from two sources: pregnancy exposure registries in which women who use drugs are monitored for the effect of these compounds on fetal development, and retrospective studies like the acetaminophen trial.
Unfortunately, data from pregnancy exposure registries is biased. Physicians are more likely to report adverse birth outcomes than they are to report healthy deliveries, raising concerns about the safety of drugs during pregnancy where issues don’t really exist. These registries are also largely ineffectual in identifying problems that arise later in life, such as ADHD. Retrospective studies like the acetaminophen trial have similar problems, including the difficulty of proving a direct link between drug exposure and health issues in children.
It doesn’t make sense that we rely on such “natural experiments” rather than collect the necessary safety and effectiveness data by conducting controlled clinical trials using informed and willing participants. It’s time for us the reconsider existing guidelines on the inclusion of pregnant women in clinical trials: the mothers and their children deserve better.
[This blog entry was originally presented as an oral commentary on Northeast Public Radio on February 27, 2014. It is also available on the WAMC website. Its contents are solely the responsibility of the author alone and do not represent the views of the Bioethics Program or Union Graduate College.]