The Nicotine Pregnancy Dilemma: What Should We Recommend When the Evidence Is Incomplete?

Smoking during pregnancy is one of the most well-characterized and serious environmental risks to fetal development. The evidence, accumulated over decades of epidemiological research, is unambiguous: maternal smoking increases the risk of miscarriage, preterm birth, low birth weight, placental abruption, and sudden infant death syndrome (SIDS). The mechanisms—carbon monoxide reducing fetal oxygen delivery, nicotine constricting placental blood vessels, the thousands of combustion products crossing the placental barrier—are well-understood. Cigarette smoking is, by a wide margin, the most harmful form of nicotine use during pregnancy, and smoking cessation is the single most impactful behavioral change a pregnant smoker can make for her baby's health. The challenge is that smoking cessation is difficult under any circumstances, and the circumstances of pregnancy—stress, hormonal changes, the disruption of established coping mechanisms—can make it harder, not easier. The pregnant smoker who cannot quit faces a dilemma: continue smoking (known catastrophic risk), use nicotine replacement therapy (uncertain risk, likely lower than smoking), switch to vaping (uncertain risk, likely lower than smoking but with even less pregnancy-specific data), or use no nicotine at all. The evidence to guide this decision is inadequate, and the default recommendation—'quit all nicotine'—is, for many pregnant smokers, not achievable.

The evidence on nicotine replacement therapy during pregnancy is more extensive than for any other nicotine product, but it is still limited and mixed. Randomized controlled trials of NRT during pregnancy have produced inconsistent results: some show improved cessation rates and improved birth outcomes (higher birth weight, fewer preterm deliveries), others show no benefit. The inconsistency may reflect dosing challenges—pregnancy accelerates nicotine metabolism, meaning that standard NRT doses may be inadequate for pregnant smokers, and the nicotine levels achieved with NRT may be insufficient to suppress the withdrawal that drives continued smoking. Observational studies suggest that NRT use during pregnancy is associated with better birth outcomes than continued smoking—as would be expected if NRT is substituting for the more harmful cigarette smoke—but the observational designs cannot exclude the possibility that NRT users differ from continuing smokers in other ways that affect birth outcomes. The evidence, on balance, supports the use of NRT during pregnancy when smoking cessation without pharmacological support has failed—but the support is qualified, and the uncertainty is genuine.

The evidence on vaping during pregnancy is essentially nonexistent. No randomized controlled trial has evaluated the safety or efficacy of vaping for smoking cessation during pregnancy, and the observational studies that exist are small, methodologically limited, and unable to separate the effects of vaping from the effects of the smoking that vaping is partially replacing. The absence of evidence is not evidence of absence of risk—nicotine itself is a vasoconstrictor that can reduce placental blood flow, and the non-nicotine constituents of vaping aerosol (propylene glycol, vegetable glycerin, flavor compounds) have unknown effects on fetal development—but the absence of evidence also does not mean that vaping is as harmful as smoking. The toxicological rationale—vaping delivers nicotine without the combustion products that are responsible for the vast majority of smoking's harm—applies during pregnancy as it does in the non-pregnant population, and the harm to the fetus from maternal vaping is almost certainly substantially less than the harm from maternal smoking. But 'almost certainly' and 'substantially less' are not the same as 'proven safe,' and the pregnant smoker making a decision under uncertainty deserves an honest account of what is known and what is not. The current public health messaging—'no amount of nicotine is safe during pregnancy, avoid all nicotine products'—provides certainty where the evidence does not, and withholds the comparative risk information that the pregnant smoker needs to make an informed decision.

The ethical framework for nicotine use during pregnancy is contested. The dominant framework—'protect the fetus from all potential harm, regardless of the consequences for the mother'—treats the pregnant woman as a fetal container whose behavior should be optimized for fetal health, without regard for her own autonomy, her own health, or the difficulty of the behavior change being demanded. This framework is ethically problematic—it denies the pregnant woman's agency and treats her as a means to the end of fetal health—but it is deeply embedded in the culture of prenatal care and in the public health communication about pregnancy. The alternative framework—harm reduction—acknowledges that some pregnant smokers cannot or will not quit all nicotine, and that the appropriate public health response is to support them in reducing harm to the fetus to the greatest extent possible, even if the reduction falls short of complete abstinence. The harm reduction framework treats the pregnant smoker as an agent making decisions under constraint, and the role of public health is to provide the information, support, and options that enable her to make the best possible decision for herself and her baby. The two frameworks lead to different recommendations: the 'protect the fetus' framework says 'abstain from all nicotine'; the harm reduction framework says 'if you cannot quit all nicotine, switching to NRT or vaping is likely much safer for your baby than continuing to smoke.' The evidence supports the harm reduction recommendation. The institutional culture of prenatal care supports the abstinence recommendation. The pregnant smoker is caught between them.

The research gaps that prevent evidence-based guidance are the product of an institutional failure, not a scientific impossibility. The reason there are no randomized controlled trials of vaping during pregnancy is not that such trials are methodologically impossible—trials of NRT during pregnancy exist, and the same methodology could be applied to vaping. The reason is that the institutional review boards, funding agencies, and research ethics committees that govern clinical research have applied a precautionary standard to vaping research that they do not apply to NRT research or, indeed, to the standard of care (continued smoking). The precautionary standard—'we cannot expose pregnant women and their fetuses to a product whose risks are unknown'—is ethically incoherent when the alternative to the research is that pregnant smokers continue to be exposed to a product (cigarettes) whose risks are known and catastrophic. The absence of evidence on vaping during pregnancy is not a justification for restricting vaping during pregnancy. It is a consequence of a research ethics framework that has, through excessive caution, denied pregnant smokers the evidence they need to make informed decisions—and the ethical responsibility for that denial rests with the institutions that have refused to generate the evidence.

For the pregnant smoker who cannot quit, the practical guidance—acknowledging the limitations of the evidence—is as follows. Complete cessation of all nicotine use is the ideal outcome and should be supported with all available resources (counseling, NRT at adequate doses, behavioral support). If complete cessation is not achievable, switching completely from smoking to NRT is likely to substantially reduce harm to the fetus, and should be encouraged and supported. If NRT is not acceptable or effective, switching completely from smoking to vaping is likely to substantially reduce harm to the fetus compared to continued smoking—the reduction in combustion products almost certainly outweighs the risks of nicotine and the unknown risks of vaping-specific constituents—but the evidence is indirect and the uncertainty is genuine. Dual use (smoking and vaping during the same period) should be avoided to the extent possible—the goal is to eliminate combustion exposure, not to supplement it. And throughout, the pregnant smoker should be treated as an agent making difficult decisions under uncertainty, with respect for her autonomy and compassion for the circumstances that make quitting so hard. The abstinence-only message that dominates prenatal care is not evidence-based, does not respect the pregnant smoker's agency, and does not serve the health of the fetus it claims to protect.

Shareable insight: Smoking during pregnancy is catastrophic for fetal health. But the pregnant smoker who cannot quit faces a dilemma: the evidence on NRT and vaping during pregnancy is limited, and the default recommendation—'abstain from all nicotine'—is not achievable for many. The appropriate guidance is harm reduction: if you cannot quit all nicotine, switching completely to NRT or vaping is almost certainly much safer for your baby than continuing to smoke. The abstinence-only message that dominates prenatal care is not supported by the evidence—and it fails the very women and babies it claims to protect.