HPV vaccination before pregnancy is linked to fewer adverse birth outcomes in Sweden

The case for human papillomavirus vaccination has always been built on cancer: it prevents the infections that cause most cervical cancer, and a growing body of population data now shows that protection turning into fewer cancers and fewer deaths. A study published in The BMJ on October 7 asks a different question — whether the vaccine leaves any mark on how a later pregnancy goes — and reports an association that points, if anything, the right way [s1].

What the researchers did

The team used Swedish national registries covering births from 2006 to 2023 [s1]. They assembled 624,713 singleton births among nulliparous women — women having a first baby — aged 16 to 35 [s1]. Of these, 92,620, or 14.8%, had received the quadrivalent HPV vaccine before becoming pregnant [s1].

The design was a matched case-control study analysed with conditional logistic regression [s1]. The primary outcomes were preterm birth, before 37 weeks, including its subtypes, and spontaneous preterm birth [s1]. Secondary outcomes included preterm prelabour rupture of membranes, prelabour rupture of membranes, small-for-gestational-age and severe small-for-gestational-age infants, stillbirth, and neonatal death [s1]. The researchers also tested whether the age at which a woman was vaccinated changed the picture [s1].

What they found

After adjustment for confounders, HPV vaccination before pregnancy was associated with lower odds of every adverse outcome the study examined [s1]. Several reached statistical significance: preterm birth before 37 weeks (adjusted odds ratio 0.95, 95% confidence interval 0.91 to 0.99), very preterm birth at 28 to 31 weeks (0.85, 0.75 to 0.97), spontaneous preterm birth (0.95, 0.91 to 0.99), preterm prelabour rupture of membranes (0.93, 0.87 to 1.00), and severe small-for-gestational-age infants (0.92, 0.87 to 0.97) [s1].

The reductions were generally more pronounced among women vaccinated at younger ages [s1]. The authors frame this as extending the benefits of HPV vaccination "beyond cancer prevention" [s1]. With nearly 15% of the 624,713 births occurring in vaccinated women, the study had the scale to detect associations this small with confidence, which is precisely why the effect sizes deserve scrutiny before the conclusion does [s1].

How much weight the numbers carry

Read the effect sizes before the headline. An odds ratio of 0.95 is a 5% relative reduction in the odds of preterm birth — real at this enormous scale, but small for any individual woman. The sturdiest finding is the 15% reduction in very preterm birth, the outcome that matters most for a newborn, though it rests on a smaller number of events and a wider interval that nearly touches 1.0.

This is observational, so the central worry is confounding. Women who received HPV vaccine before a first pregnancy are not a random slice of the population. In Sweden, organised HPV vaccination began for schoolgirls around 2012, so vaccinated mothers skew younger, more recently pregnant, and plausibly different in health behaviour, screening uptake and socioeconomic position — all of which independently track with how a pregnancy ends. The investigators adjusted for measured confounders, but adjustment cannot reach what the registries do not record.

The dose-by-age pattern cuts both ways. A stronger signal in those vaccinated youngest is biologically coherent if the mechanism runs through preventing cervical HPV infection, treatment of precancerous lesions, and the cervical procedures that are themselves a known risk factor for later preterm birth. But younger age at vaccination is also a near-perfect marker of the school-based programme, and therefore of birth cohort and era — exactly the kind of variable that can manufacture a gradient on its own.

Why the mechanism is plausible

There is an established pathway here. Persistent HPV infection leads to cervical intraepithelial neoplasia; the excisional treatments used to remove it, such as loop electrosurgical excision, shorten and weaken the cervix; and a treated cervix carries a higher risk of preterm birth in a subsequent pregnancy. A vaccine that heads off the infection heads off the lesion and the procedure. On that logic, a downstream reduction in preterm birth is not a surprise so much as a prediction — which is also why the association, while it cannot prove cause, is not easy to dismiss as noise.

What to watch

Whether the finding replicates in other countries with long-running vaccination programmes and linkable birth data, where the confounding structure differs; whether studies can isolate the cervical-procedure pathway from general cohort effects; and whether the nonavalent vaccine, now standard in most programmes, shows the same pattern. None of this changes the reason the vaccine exists. Its job is cancer prevention, and the evidence for that is far stronger than anything in this paper.

This article describes a single observational study and is informational only. It is not medical advice.

Sources

• [s1] Quadrivalent human papillomavirus vaccination before pregnancy and pregnancy outcomes: nationwide, population based case-control study, BMJ, 2026;395:e100739, published 2026-10-07. PMID 42843853.