HealthProtocols
← All sources

Association of Coffee Consumption With Total and Cause-Specific Mortality Among Nonwhite Populations

Multiethnic Cohort prospective study (185,855 adults, Hawaii and Los Angeles, mean follow-up 16.2 years): higher coffee consumption associated with lower total mortality versus non-drinkers after smoking adjustment (e.g., ≥4 cups/day HR 0.82; 95% CI 0.78–0.87; P for trend < 0.001), with similar patterns for caffeinated and decaffeinated coffee and inverse associations for several major causes of death.

Design

  • Cohort: Multiethnic Cohort (MEC) — 185,855 adults (45–75 y at recruitment; 1993–1996)
  • Exposure: coffee intake from validated FFQ (non-drinker referent)
  • Follow-up: through 2012 (58,397 deaths)

Total mortality (multivariable Cox; abstract)

Compared with no coffee (adjusted for smoking and other covariates):

  • 1 cup/day: HR 0.88 (95% CI 0.85–0.91)
  • 2–3 cups/day: HR 0.82 (95% CI 0.79–0.86)
  • ≥4 cups/day: HR 0.82 (95% CI 0.78–0.87); P for trend < 0.001
  • Caffeinated vs decaffeinated: abstract reports similar trends

Cause-specific patterns (abstract summary)

Inverse associations reported for deaths from heart disease, cancer, respiratory disease, stroke, diabetes, and kidney disease.

Evidence hygiene

  • Observational — residual confounding possible, though sensitivity analyses summarized as reassuring in abstract.
  • Use alongside EPIC (Gunter et al. 2017, PMID 28693038) and umbrella summary (Poole et al. 2017, PMID 29167102).

Publication

Park SY, Freedman ND, Haiman CA, et al. Ann Intern Med. 2017 Aug 15;167(4):228-235. doi: 10.7326/M16-2472. PMID 28693036.

Outcomes

  • All-Cause Mortality Risk
    Events: /
  • All-Cause Mortality Risk
    Events: /
View original paper →