The Pregnancy Paradox: Why Most Drugs Prescribed During Pregnancy Were Never Tested in Pregnancy
Millions of pregnant women take medications for depression, epilepsy, high blood pressure, diabetes, and infections. Most of those drugs were never properly studied in pregnancy. That is the pregnancy paradox, and it is the subject of the Clarity Clinical Solutions video: we have been so afraid of studying drugs in pregnancy that pregnant patients are routinely treated with drugs that have no pregnancy-specific evidence behind them.
The scale is striking. According to the video, over 90% of pregnant women take at least one medication during pregnancy, with the average being three to five different drugs. Fewer than 1% of approved drugs have adequate pregnancy safety data from clinical trials, and more than 80% of medication use during pregnancy is off-label. In other words, the same drugs carefully studied for the general population get prescribed to pregnant women without any pregnancy-specific evidence on safety or dosing.
How we got here
The exclusion of pregnant women has two roots. The historical root is the thalidomide tragedy of the 1950s and 1960s, when a morning sickness drug caused severe birth defects in more than 10,000 infants. That was followed by the DES disaster, where a drug given to prevent miscarriage caused cancers in the daughters of treated women decades later. These catastrophes produced a protective reflex that effectively shut pregnant women out of drug research.
The contemporary barriers are more mundane: fear of liability, the physiological complexity of pregnancy, ethical concerns about fetal risk, and cautious institutional review boards. The video's point is that well-intentioned protection created an evidence vacuum. Exclusion sounds protective, but it has its own costs.
The costs of exclusion
First, when pregnant patients need medication, doctors prescribe based on animal data and expert opinion rather than human trial data. That is the opposite of evidence-based medicine. Second, some pregnant women avoid necessary medications because of unknown risks. Untreated depression, uncontrolled epilepsy, and poorly managed hypertension harm both mother and baby. The video cites the numbers: roughly 30% of pregnant women with depression stop their antidepressants during pregnancy, and their relapse rate increases fivefold. Third, pregnancy changes how the body handles drugs, so even when a drug is used, the dose may be wrong.
Pregnancy changes drug handling
Pregnancy rewires pharmacokinetics. Blood plasma volume increases by about 50%, which dilutes drug concentrations and often requires higher loading doses. Kidney filtration increases by about 50%, so renally cleared drugs are eliminated faster and doses must go up. Liver enzyme activity changes dramatically. CYP3A4, the enzyme that metabolizes roughly half of all drugs, is induced during pregnancy. Albumin levels drop, leaving more free active drug for protein-bound medications. The video notes that these changes can reduce drug concentrations at the target site by 30 to 60%. Without pregnancy-specific pharmacokinetic data, dosing is a guess.
The regulatory response
In 2014, the FDA finalized the Pregnancy and Lactation Labeling Rule, which replaced the old letter categories that had been in use since 1979 (FDA: Pregnancy and Lactation Labeling Rule, final rule). The letters were widely misunderstood. A category B label of "no evidence of risk" often meant no human studies had been done at all. The new rule requires narrative summaries with data from pregnancy exposure registries, risk summaries, clinical considerations, dose adjustments during pregnancy, and information about labor and delivery. Labels must be updated as new data arrives.
The FDA has also published guidance on including pregnant women in clinical trials when appropriate, laying out the scientific and ethical considerations for sponsors (FDA: Pregnant Women: Scientific and Ethical Considerations for Inclusion in Clinical Trials). At the federal level, the Task Force on Research Specific to Pregnant Women and Lactating Women, known as PRGLAC, made 15 recommendations to the NIH and FDA addressing ethical frameworks, trial design, and regulatory incentives (NIH: PRGLAC Task Force).
The ethical framework for inclusion
US regulations already offer a path. The federal regulations for protecting human subjects include a special subpart for pregnant women, which permits research when it presents no more than minimal risk to the fetus, or when the research offers direct benefit to the woman or the fetus (45 CFR 46 Subpart B). The video gives a concrete example: a single extra blood draw for pharmacokinetic analysis is minimal risk, and many useful pregnancy PK studies could be done under this standard. When a pregnant woman has a medical condition, she can also enroll in treatment trials that offer direct benefit, such as an HIV medication trial for a pregnant woman with HIV.
The ethical case for inclusion rests on more than regulation. Pregnant patients can make informed decisions, there is an ethical obligation to study drugs that millions of pregnant women take, and the concept of "dignity of risk" holds that excluding people from potentially beneficial research is itself an ethical problem.
Registries are necessary but not enough
Most pregnancy safety information comes from pregnancy exposure registries, observational studies that follow women who take specific medications during pregnancy and compare outcomes against a control group. The Antiretroviral Pregnancy Registry has followed over 20,000 pregnancies and provides essential safety data for HIV medications. But registries have limits. They can only detect relatively large increases in birth defects. They cannot determine appropriate dosing. And they only start collecting data after the drug is already on the market and thousands of women have been exposed. The video sums it up: registries are necessary but not sufficient.
Smarter designs for the future
Several trial designs can advance pregnancy research with minimal risk. Adaptive pharmacokinetic studies use sparse blood sampling, just two or three samples per patient, in pregnant women already receiving the drug clinically. Physiologically based pharmacokinetic modeling uses computer simulations to predict drug behavior in pregnancy, and combined with a small amount of real PK data can support dosing recommendations. Staggered enrollment starts with third-trimester patients, then second, then first, building evidence step by step.
Regulatory incentives modeled on the Pediatric Exclusivity Program, where studying a drug in pregnancy would earn extended market exclusivity, could change the economics for sponsors. Decentralized trial designs suit pregnancy studies well because they reduce clinic-visit burden. And the broader movement is toward default inclusion: pregnant women included in research unless there is a specific scientific or safety reason for exclusion.
The bottom line
The reflex to exclude pregnant women came from a genuine desire to protect. After sixty years, the evidence shows exclusion has caused its own harms, from unstudied dosing to millions of off-label prescriptions. The path forward is not reckless inclusion. It is careful, incremental research that starts with minimal-risk pharmacokinetic studies, uses innovative designs, and respects the autonomy of pregnant patients to make informed decisions. Protecting pregnant patients means studying them, not excluding them.
This article is based on the Clarity Clinical Solutions video "Most Drugs Prescribed to Pregnant Women Have Never Been Tested in Pregnancy." Watch it here: Most Drugs Prescribed to Pregnant Women Have Never Been Tested in Pregnancy
References
- Clarity Clinical Solutions — "Most Drugs Prescribed to Pregnant Women Have Never Been Tested in Pregnancy" (framework for this article). https://www.youtube.com/watch?v=sckYSbmZ_mo
- Federal Register — FDA final rule: Content and Format of Labeling for Human Prescription Drug and Biological Products; Requirements for Pregnancy and Lactation Labeling (PLLR). https://www.federalregister.gov/documents/2014/12/04/2014-28241/content-and-format-of-labeling-for-human-prescription-drug-and-biological-products-requirements-for
- FDA — Guidance: Pregnant Women: Scientific and Ethical Considerations for Inclusion in Clinical Trials. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/pregnant-women-scientific-and-ethical-considerations-inclusion-clinical-trials
- NIH / NICHD — Task Force on Research Specific to Pregnant Women and Lactating Women (PRGLAC). https://www.nichd.nih.gov/about/advisory/PRGLAC
- eCFR — 45 CFR 46 Subpart B, Additional Protections for Pregnant Women, Human Fetuses and Neonates Involved in Research. https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-A/part-46/subpart-B