Prescribing in Pregnancy: How to Think About It, Using Ondansetron as the Worked Example

Interns are routinely asked whether a drug is “safe in pregnancy”, and the expected answer is yes or no. That framing is the problem, because it hides the comparison that actually matters: the risk of the drug against the risk of leaving the condition untreated.

At a glance

Why simple answers are usually wrong

Very few drugs have been formally trialled in pregnancy, for obvious ethical reasons. Most of what is known comes from observational data, registries and post-marketing surveillance. That evidence is genuinely useful, but it carries confounding that trials would have removed — most importantly confounding by indication, where the condition being treated is itself associated with the outcome being measured.

This is why the honest answer to many questions is a description of what is known and how confidently, rather than a verdict.

The worked example: ondansetron

Ondansetron is often repeated among students as a drug that is simply not given in pregnancy. That is a flattening of a real controversy, and it is worth understanding properly because it is a model for how to handle contested evidence generally.

Ondansetron is widely used for nausea and vomiting in pregnancy, including hyperemesis, in many countries. Some observational studies raised concerns about a possible association with orofacial clefting and with cardiac septal defects when used in the first trimester. Other studies did not replicate those findings. The absolute effect sizes discussed have been small, the studies differ in design and in how they controlled for confounding, and expert bodies in different countries have reached different practical conclusions.

The result is that guidance genuinely differs. In many settings ondansetron is used, commonly after first-line options such as doxylamine with pyridoxine have been tried, with the discussion documented. In other settings first-trimester use is discouraged.

What follows for you as an intern is not a fact to memorise but a method: check your own national and hospital guidance, because this is one of the questions where it varies; and if asked, describe the uncertainty accurately rather than asserting a rule that your examiner or your consultant may not share.

The other half: untreated hyperemesis is not benign

The reason ondansetron is used at all is that severe, persistent vomiting in pregnancy carries its own substantial risks — dehydration, electrolyte disturbance, weight loss, nutritional deficiency, and significant effects on the woman’s ability to function and on her mental health. Withholding all treatment is not a neutral act. It is a choice with its own consequences, and it is the side of the equation that gets forgotten when the question is framed as whether a drug is “safe”.

How to think it through in practice

1. Establish the gestation, because timing changes the concern.
2. Establish how severe the condition is and what it is doing to the patient.
3. Check your current national formulary and local guideline for that specific drug in pregnancy. Do this every time — this is a fast-moving area and memory is not adequate.
4. Consider whether a non-drug measure or a better-characterised alternative would suffice.
5. Where a drug is needed, use the option with the most reassuring data for the shortest period at the lowest effective dose.
6. Discuss it with the patient, including that the evidence is imperfect. Women are entitled to participate in decisions about their own pregnancy.
7. Document the reasoning and the discussion.
8. Escalate. Prescribing in pregnancy is rarely something an intern should finalise alone.

What seniors actually ask

• What is the gestation?
• What happens if we do not treat this?
• What does our own guideline say, and have you looked at it today?
• Have you discussed the uncertainty with her?

Last-minute checklist

• Ask what the untreated condition would do, not only what the drug might do
• Establish gestation before deciding anything
• Check the current formulary and local guideline rather than recalling a rule
• Where evidence is contested, say so — do not convert it into a rule
• Involve the patient in the decision and document it
• Escalate rather than deciding alone

References

• Your current national formulary’s pregnancy guidance
• Your hospital’s obstetric prescribing guideline
• National obstetric guidance on nausea and vomiting in pregnancy