Coffee and Mortality: Morning Timing Reduces Risk, All-Day Drinking Does Not

What did the 2025 study find?

Wang et al. (European Heart Journal, 2025) analyzed data from 40,725 adults in the U.S. National Health and Nutrition Examination Survey (NHANES) 1999–2018. The median follow-up was 9.8 years, during which 4,295 all-cause deaths were recorded.

The authors classified participants by coffee consumption pattern:

• 36% — morning drinkers (coffee consumed predominantly before noon).
• 14% — all-day drinkers (coffee distributed evenly throughout the day).
• 48% — non-coffee drinkers.

The primary outcome was all-cause mortality; the secondary outcome was cardiovascular disease mortality. After adjustment for age, sex, smoking, diet, physical activity, and several other confounders:

• Morning drinkers: HR = 0.84 (95% CI: 0.74–0.95) for all-cause mortality; HR = 0.69 (0.55–0.87) for cardiovascular mortality.
• All-day drinkers: HR = 0.96 (0.87–1.05) for all-cause mortality — a statistically non-significant difference.

External validation in a separate sample of 1,463 participants confirmed the direction of the findings.

What do earlier large cohorts show?

Freedman et al. (NEJM, 2012; PMID: 22591295) followed 400,000 participants in the NIH-AARP study aged 50–71. After adjusting for smoking and other factors, consuming 4–5 cups per day yielded a mortality hazard ratio of 0.88 in men (p < 0.001 for trend). The effect in women was comparable. This study did not account for the time of day of consumption.

Poole et al. (BMJ, 2017) conducted an umbrella review of meta-analyses covering more than 200 health outcomes. For all-cause mortality, the optimal level was 3–4 cups per day, HR = 0.83 (95% CI: 0.79–0.88). Stroke mortality at 3 cups — HR = 0.70 (0.57–0.86). Notably, decaffeinated coffee at 3 cups was also associated with lower risk — HR = 0.89 (0.85–0.93) — pointing to the role of non-caffeine coffee components.

Why might timing matter?

Wang et al. (2025) proposed several physiological explanations. Caffeine suppresses adenosine receptors — a mechanism that supports wakefulness and may contribute to cardiovascular protection. When coffee is consumed in the afternoon, caffeine's half-life (on average 5–6 hours) can delay sleep onset and impair sleep quality. Chronic sleep disruption is associated with inflammation, insulin resistance, and cardiovascular risk.

Morning coffee consumption aligns with the natural post-awakening cortisol peak and is least likely to affect nighttime sleep quality. In addition, a regular morning routine may itself be a marker of a more structured lifestyle — a classic confounding problem in observational research.

Limitations: what cannot be concluded from these data

All three studies are observational. Coffee consumption patterns in NHANES were based on self-report. Residual confounding is unavoidable: morning coffee drinkers may on average lead healthier lives. Reverse causality is also plausible: people with chronic illness often restrict or shift their coffee consumption timing.

The data on the morning pattern (Wang et al., 2025) represent a single study without independent replication. Until confirmatory data emerge, these findings should be considered preliminary.