Is coffee good or bad for your heart? Across 2.86 million people, genuinely both.

Coffee's relationship with heart health has been studied for decades, with results conflicting enough that a definitive public answer has never quite landed. An updated meta-analysis published this month in BMC Cardiovascular Disorders pools the largest body of evidence yet — and the honest answer it arrives at is that coffee is not uniformly good or bad for the cardiovascular system. It appears to be both, depending on which condition you're asking about [s1].

The scale of the analysis

Researchers searched five databases — PubMed, Web of Science, the Cochrane Library, Embase, and Scopus — for studies published between January 2000 and December 2025 examining coffee's relationship to cardiovascular disease [s1]. After exclusions, 38 studies covering 2,856,002 participants were included in the pooled analysis [s1], with study quality assessed using the Newcastle-Ottawa Scale for cohort studies and the Joanna Briggs Institute tool for case-control studies [s1]. The review was registered in advance with PROSPERO, the international registry for systematic reviews [s1].

Where coffee showed no effect

Comparing people in the highest coffee-consumption category against those in the lowest, the pooled analysis found no significant association between coffee consumption and total coronary heart disease (hazard ratio 0.98, p = 0.80) or heart failure (HR 1.03, p = 0.62) [s1]. In plain terms: across this pooled dataset, how much coffee someone drank didn't meaningfully move their risk of these two conditions in either direction.

Where coffee looked harmful

The analysis found a significant positive association between higher coffee consumption and the risk of myocardial infarction — heart attack — with an odds ratio of 1.48 (p < 0.0001) [s1]. That's the meta-analysis's clearest signal in the direction of harm, and it's a large enough effect size, on a large enough pooled sample, that it's unlikely to be a statistical fluke, though the pooled odds ratio doesn't establish causation on its own — it reflects an association across observational studies of varying design.

Where coffee looked protective

Set against that heart attack finding, the same analysis found an inverse relationship between coffee consumption and both stroke (HR 0.89, p = 0.01) and cardiac arrhythmias (HR 0.94, p = 0.04) [s1] — meaning higher coffee intake was associated with a lower risk of both. A separate finding specific to caffeine intake, rather than coffee consumption broadly, showed a non-linear relationship with cardiovascular disease mortality among hypertensive patients, with a hazard ratio of 0.68 (p = 0.001) [s1] — a fairly substantial reduction in cardiovascular death risk specifically within that patient population.

How to read a genuinely mixed result

The study's own conclusion resists the urge to round its findings into a single verdict: higher coffee intake might increase the risk of heart attack, while also offering protective effects against stroke and cardiac arrhythmias, and higher caffeine intake specifically may reduce cardiovascular mortality risk in people with high blood pressure [s1]. Those aren't contradictory findings so much as evidence that coffee's cardiovascular effects aren't uniform across different conditions — a substance can plausibly raise risk for one outcome through one mechanism (caffeine's acute effects on heart rate and blood pressure, for instance) while lowering risk for another through a different one (coffee's antioxidant compounds and their association with vascular and metabolic effects), and this analysis captures that complexity rather than resolving it into a simple headline.

What the analysis doesn't settle

This is a meta-analysis of observational studies, not a randomized controlled trial — it can identify associations at scale but can't fully rule out that people who drink more or less coffee differ in other health-relevant ways that the pooled analysis doesn't fully control for. The "non-linear relationship" reported for caffeine and CVD mortality in hypertensive patients also signals that the benefit isn't simply "more is better" — a non-linear relationship typically means the effect changes direction or magnitude at different levels of intake, a nuance the single hazard ratio reported here doesn't fully capture. And none of this analysis translates into a specific recommended amount of coffee for any individual reader; it describes population-level associations across a wide range of study designs and coffee consumption patterns, not a personalized dosing guideline.

Sources

• Coffee, caffeine, and cardiovascular health: navigating risks and benefits—an updated systematic review and meta-analysis — BMC Cardiovascular Disorders, 2026-05-07