Heart attacks and strokes rose for two days after the 2016 US election

Major elections are stressful. That is not a controversial claim, and surveys have measured the stress for years. The harder question is whether that stress is large enough, and sharp enough, to show up in one of the few places population data can detect it: the rate at which people are admitted to hospital with a heart attack or a stroke in the days immediately around the vote.

A study published in the Proceedings of the National Academy of Sciences in 2020 asked exactly that of the 2016 US presidential election [s1].

The design

Researchers using data from Kaiser Permanente Southern California compared hospitalisation rates for acute cardiovascular disease immediately before and after the 2016 election date among adults aged 18 and over in the health system [s1]. Cardiovascular disease here meant an inpatient or emergency department discharge diagnosis of acute myocardial infarction or stroke, identified with ICD-10 codes [s1].

The comparison was deliberately narrow. Rate ratios compared cardiovascular rates in the two days following the 2016 election to the rates in the same two days of the prior week [s1]. Using the week before as the control is the central move: it holds the day of the week, the season, and the background health of the population roughly constant, so that what is left is closer to the election itself. Rates are expressed per 100,000 person-years, a way of standardising counts to the time people were actually under observation rather than to a raw headcount.

What they found

In the two days after the 2016 election, the rate of cardiovascular events was 573.14 per 100,000 person-years, compared with 353.75 per 100,000 person-years in the window before — 1.62 times higher (95% CI 1.17 to 2.25) [s1].

Analysed separately, the two components moved in the same direction. The rate ratio was 1.67 for acute myocardial infarction (95% CI 1.00 to 2.76) and 1.59 for stroke (95% CI 1.03 to 2.44) [s1]. The results were similar across sex, age, and race or ethnicity groups [s1].

How much this establishes

Less than the single headline number suggests, and the authors say so. Their own framing is that transiently heightened cardiovascular risk around the election may be attributable to sociopolitical stress [s1] — "may be", because an observational before-and-after comparison cannot rule out everything else that differs between one two-day window and another.

Three cautions sit behind that hedge.

The confidence intervals are wide. For acute myocardial infarction the lower bound is 1.00, sitting exactly on the edge of statistical significance. A single health system over a handful of days yields a small number of events, and small numbers produce unstable estimates.

The two-day window is both the strength and the weakness of the design. It makes the comparison clean, but it also means the study captures an acute trigger rather than a lasting shift. Nothing here speaks to cardiovascular risk weeks or months after the vote.

And a before-and-after design attributes the whole difference to the election when anything else peculiar to those two days — a weather event, a holiday pattern, a reporting quirk — would be folded into the same estimate. The control week guards against slow-moving differences, not against something that happened to coincide with those particular days.

Why the mechanism is plausible anyway

The idea that an acute emotional stressor can trigger a cardiovascular event within hours to days is neither new nor exotic. It is one of the better-established ideas in cardiovascular medicine, described after earthquakes, major sporting losses, and bereavement. The election study is one more instance of a known pattern rather than a novel biological claim [s1].

The American Heart Association's 2021 scientific statement on psychological health and cardiovascular disease synthesises that wider literature, concluding there are clear associations between psychological health and cardiovascular disease and risk, with increasing evidence that the relationship may be causally linked to biological processes [s2]. A short spike around a national election is the kind of acute signal that general picture would predict.

What it does not say

It does not say the election harmed any individual's heart; no population study can reach a single person. It does not favour any party or outcome. The mechanism proposed is stress, which cuts across political lines, and the researchers studied the days around the vote, not who won or what policies followed. And it forecasts nothing about any later election.

What it offers is a measured estimate of a plausible, transient effect with honest uncertainty attached.

If an election season is leaving you anxious, distressed, or unwell, that is worth taking seriously. Step back from the coverage, and reach out to a clinician or a crisis line if you need support. Chest pain or sudden stroke symptoms — facial droop, arm weakness, slurred speech — are a medical emergency regardless of the calendar; call your local emergency number.

This article is informational and does not constitute medical advice.