Protecting pregnant people means more than limiting research exposure. It means generating evidence that makes treatment decisions safer, during pregnancy and beyond.
A clinical trial can exclude pregnancy. A health service cannot.
Pregnant people still need treatment for HIV, tuberculosis, hypertension and other conditions. When pregnancy-specific evidence is missing, clinicians must make decisions anyway: whether to prescribe, which dose to use, and how to weigh the risks of treatment against those of untreated disease.
Exclusion does not remove those decisions. It leaves patients and clinicians to make them with greater uncertainty.
The answer is not to include pregnant participants in every trial. It is to stop treating exclusion as an adequate endpoint. Where an intervention is likely to be used during pregnancy, its development programme should explain how the evidence needed for that use will be generated.
What Botswana’s experience teaches
The history of dolutegravir, an HIV medicine, shows why maternal research needs both prospective studies and sustained safety surveillance.
In 2018, Botswana’s Tsepamo birth-outcomes surveillance study identified a possible association between dolutegravir use at conception and neural tube defects, serious abnormalities affecting the developing brain or spine. Continued surveillance changed the interpretation as more pregnancies were recorded. A 2025 reanalysis of Tsepamo data described how the initially large difference diminished, with updated estimates of 0.10% for dolutegravir exposure at conception and 0.11% for other antiretroviral exposure (1). Surveillance in Eswatini also provided reassuring evidence (2).
The lesson is not that the initial concern should have been ignored. It is that a safety signal must be investigated, communicated with its uncertainty intact, and reassessed as evidence accumulates.
African maternity services and research teams made that possible. Their contribution was not simply enrolling participants; it was maintaining the systems needed to detect a concern and help determine what it meant.
There is an equally important limit to this example. A modest-sized trial would not necessarily have resolved a question about a rare birth defect. Nor can a study beginning late in pregnancy establish safety around conception. Earlier trials and large-scale surveillance answer different questions. Responsible maternal research needs both.
Replace blanket exclusion with a plan for evidence
The ethical direction is already clear. CIOMS Guideline 19 calls for research addressing the health needs of pregnant and breastfeeding women, following careful assessment of the available evidence (3). WHO’s 2024 clinical-trial guidance similarly supports planning for their participation and assessing safety early (4).
Neither position makes pregnancy irrelevant to risk. Both require more careful reasoning than a blanket prohibition.
Decisions should consider reproductive toxicology, how the intervention works, previous human exposure, disease severity, available treatments, gestational timing and potential benefit. They should also distinguish procedures that offer a prospect of individual benefit from those undertaken only to generate knowledge. The acceptable risk is not identical in each case.
Sometimes exclusion from a particular trial is justified. The next question should then be explicit: what evidence or safeguards would allow pregnancy research to proceed, and who is responsible for securing them?
This turns exclusion from an indefinite default into a decision that can be reviewed. It also prevents a common conceptual mistake: comparing research participation with an imagined state of no risk, rather than with the real alternatives of treatment, delayed treatment or untreated disease.
Design around the clinical question
Pregnancy changes how the body handles medicines. Drug exposure may differ across gestation and after birth, so dosing cannot always be assumed from studies in non-pregnant adults. Yet not every uncertainty requires a large efficacy trial. The study design should match the question.
WHO’s antiretrovirals-in-pregnancy research toolkit describes approaches ranging from pharmacokinetic studies, which examine drug concentrations over time, to randomized trials and observational follow-up (5).
DolPHIN-2, conducted in South Africa and Uganda, studied HIV treatment initiated late in pregnancy (6). IMPAACT 2010/VESTED enrolled participants between 14 and 28 weeks’ gestation and evaluated both virologic efficacy and pregnancy-related safety outcomes (7). These studies demonstrate that clinically useful questions can be investigated through protocols designed specifically for pregnancy. Their findings must still be interpreted within the gestational periods and outcomes studied.
Such protocols need obstetric, midwifery, pharmacology and neonatal expertise from the outset, alongside ethics, statistical and community input. Maternal illness, pregnancy complications, fetal outcomes and infant health should be assessed distinctly rather than compressed into a single reassuring label of “safety.”
Comparisons also matter. An adverse outcome after treatment is not automatically caused by treatment. Underlying disease, other medicines and differences in care can influence outcomes. Relevant comparison groups, reliable gestational dating and consistent outcome definitions are essential to interpreting the evidence.
Safeguards must extend beyond data collection. Obstetric and neonatal referral arrangements should be funded, workable and tested before enrollment. Consent should explain known risks, unresolved questions and alternatives to participation. Partner or family involvement may support a participant’s decision, but it must not replace her own informed consent.
Birth is not the end of the study
A healthy appearance at delivery cannot answer every question about an intervention used during pregnancy.
Depending on the product and its potential risks, follow-up may need to assess maternal health after birth, neonatal complications, exposure through breastfeeding, infant growth or later development. The duration should be scientifically justified, not determined solely by the end of a grant or the convenience of a study visit schedule.
This is where protocol design meets health-service reality. A participant may receive antenatal care at one facility, deliver at another and take her infant elsewhere for postnatal care. A credible follow-up plan must account for those transitions.
That means agreeing who retains responsibility, how maternal and infant records will be linked securely, how clinically important findings will be communicated, and what support makes follow-up practical for families. It also means obtaining appropriate consent for infant assessments and protecting the confidentiality of both mother and child.
The same principle applies when pregnancy occurs during another trial. Stopping an investigational product, when clinically indicated, need not mean stopping all evidence collection. With consent and appropriate oversight, pregnancy and infant outcomes can still be followed.
Make responsibility visible in budgets and decisions
For African research systems, the priority is not another general commitment to inclusion. It is making maternal evidence an explicit responsibility in research planning and financing.
Sponsors should justify pregnancy exclusions and identify the remaining evidence needs. Funders should budget for specialist input and appropriate follow-up beyond delivery. Investigators and maternity services should agree referral and data-sharing arrangements before recruitment. Regulators and ethics committees should examine whether these plans are scientifically sound and operationally credible.
Pregnancy registries and medicine-safety surveillance should complement this work, particularly for uncommon outcomes and exposures around conception. They are not substitutes for trials when prospective research is appropriate, just as trials are not substitutes for continued surveillance.
Preparedness requires the same discipline. The PREVENT Working Group’s ethics guidance argues for pregnant women’s interests to be incorporated into epidemic vaccine research and response before an emergency forces decisions under pressure (8). Waiting for an outbreak to begin is too late to start designing every safeguard.
For ACRN, the position is straightforward: pregnancy should prompt a reasoned research decision, not an automatic exclusion. Inclusion must be scientifically justified, supported by proportionate safeguards and respectful of participant autonomy. Where inclusion is not appropriate, the unanswered clinical question should remain visible.
Success is not simply more pregnant participants in trials. It is better answers about which interventions to use, when to use them, at what dose, and with what consequences for mothers and children.
A responsible research programme should be able to explain not only whom it protects from research risk, but how it will reduce the uncertainty they face in care.
References
- Brummel SS, Swanson S, Caniglia E, Lockman S, Zash R, Shapiro R. A lesson in embracing uncertainty in early safety signals from the Tsepamo study. AIDS. 2025;39(15):2266–73.
- Gill MM, Khumalo P, Chouraya C, Kunene M, Dlamini F, Hoffman HJ, et al. Strengthening the Evidence: Similar Rates of Neural Tube Defects Among Deliveries Regardless of Maternal HIV Status and Dolutegravir Exposure in Hospital Birth Surveillance in Eswatini. Open Forum Infect Dis. 2023;10(9):ofad441.
- Council for International Organizations of Medical Sciences. Guideline 19: Pregnant and breastfeeding women as research participants. International Ethical Guidelines for Health-related Research Involving Humans: Prepared by the Council for International Organizations of Medical Sciences (CIOMS) in collaboration with the World Health Organization (WHO) [Internet]. 4th ed. Geneva: Council for International Organizations of Medical Sciences; 2016.
- World Health Organization. New global guidance puts forward recommendations for more effective and equitable clinical trials Geneva: World Health Organization; 2024 Available from: https://www.who.int/news/item/25-09-2024-new-global-guidance-puts-forward-recommendations-for-more-effective-and-equitable-clinical-trials.
- World Health Organization. Clinical trials and observational studies Geneva: World Health Organization; Available from: https://www.who.int/tools/antiretrovirals-in-pregnancy-research-toolkit/clinical-trials-and-observational-studies.
- Kintu K, Malaba TR, Nakibuka J, Papamichael C, Colbers A, Byrne K, et al. Dolutegravir versus efavirenz in women starting HIV therapy in late pregnancy (DolPHIN-2): an open-label, randomised controlled trial. The Lancet HIV. 2020;7(5):e332–e9.
- Lockman S, Brummel SS, Ziemba L, Stranix-Chibanda L, McCarthy K, Coletti A, et al. Efficacy and safety of dolutegravir with emtricitabine and tenofovir alafenamide fumarate or tenofovir disoproxil fumarate, and efavirenz, emtricitabine, and tenofovir disoproxil fumarate HIV antiretroviral therapy regimens started in pregnancy (IMPAACT 2010/VESTED): a multicentre, open-label, randomised, controlled, phase 3 trial. The Lancet. 2021;397(10281):1276–92.
- Krubiner CB, Faden RR, Karron RA, Little MO, Lyerly AD, Abramson JS, et al. Pregnant women & vaccines against emerging epidemic threats: Ethics guidance for preparedness, research, and response. Vaccine. 2021;39(1):85–120.