Preterm births to US Latina women rose above expected after the 2016 election

One of the more striking claims in the literature on elections and health is that a national vote can register in birth outcomes — that the stress attributed to a political event might show up, months later, in the gestational age of babies born to the groups most affected. A 2019 study in JAMA Network Open tested that claim for the 2016 US presidential election and the US Latina population [s1].

The design

The study used an interrupted time series, a method for asking whether a trend in an outcome is interrupted at the moment of some event [s1]. The researchers took monthly counts of preterm births — gestational age under 37 weeks — among Latina women and compared the counts observed after the 2016 election with the number that would have been expected had the election not occurred, based on the preceding trend [s1].

The data were national and large: singleton births by month and race or ethnicity from 1 January 2009 through 30 July 2017, drawn from the Centers for Disease Control and Prevention's WONDER database, which recorded 32,860,727 live births over the period [s1]. The analysis was run separately for male and female births [s1].

What was found

Across the full period, 11.0% of male and 9.6% of female births to Latina women were preterm, compared with 10.2% and 9.3% respectively among other women [s1].

The key result concerns the gap between observed and expected. In the nine-month window beginning with November 2016, the study estimated an additional 1,342 male preterm births (95% CI 795 to 1,889) and 995 female preterm births (95% CI 554 to 1,436) to Latina women above the number expected had the election not taken place [s1]. That is roughly 2,300 more preterm births than the pre-election trend predicted, concentrated in the months that would have been in progress on election day.

A second study, a different city

The 2019 analysis does not stand alone. A 2018 study in the Journal of Epidemiology and Community Health examined all singleton births in New York City from 1 September 2015 to 31 August 2017 — 230,105 births — and asked the same kind of question of immigrant, Hispanic, and Muslim populations [s2].

Comparing the period before the presidential nomination with the post-inauguration period, the overall preterm birth rate rose from 7.0% to 7.3% (relative risk 1.04; 95% CI 1.00 to 1.07) [s2]. The sharpest increase was among foreign-born Hispanic women with Mexican or Central American ancestry, at a relative risk of 1.15 (95% CI 1.01 to 1.31) [s2]. Rates were also higher for women from the Middle East and North Africa and from the travel-ban countries, though those estimates were not statistically significant because the number of events was small [s2].

Two independent teams, two different datasets, one national and one a single city, found increases pointing the same way and concentrated in the same populations. Convergence like that is harder to dismiss than a single result.

How much this establishes

Both studies are careful, and both are explicit that this is association, not demonstrated causation.

An interrupted time series is strong against slowly changing confounders — it models the existing trend and tests for a break — but it cannot distinguish the election itself from anything else that changed at the same moment and affected the same group. Preterm birth also has many drivers, and a model that attributes a deviation to "the election" is really attributing it to everything that coincided with that date.

The outcome is a count of recorded births, not a direct measurement of stress or its biology. The proposed pathway — that sociopolitical stress and fear in targeted communities can raise the risk of early delivery — is biologically plausible and consistent with a broader literature on severe stressors and preterm birth [s2]. But plausibility of mechanism is not the same as having measured it, and neither study has an individual-level marker of the stress it invokes.

What it is not

It is not a statement about who won, and not a partisan claim. The populations studied were defined by who the authors judged to be targeted by the rhetoric and policy of the period, and the mechanism proposed is stress in those communities — not a verdict on any party. And it does not predict anything about any later election.

What the two studies offer, read together, is a measured and corroborated estimate: in the months around the 2016 election, preterm births to US Latina and targeted immigrant women ran above what the prior trend predicted, by an amount large enough to see and uncertain enough to state carefully.

This article is informational and does not constitute medical advice.