After colon cancer, coffee showed no clear link. Decaf was different.

Cohort study. 1,098 people. post-diagnosis survey 2003-2008, followed to end of 2019; mean follow-up 10.4 years (SD 4.9).

People were watched for years. Nobody was told what to do. A clue, not proof.

Who: Multiethnic Cohort (Hawaii and Los Angeles, California) participants diagnosed with colorectal cancer 1993-2008; mean age at diagnosis 69, 55% men, 73% colon cancer, 60% localized / 36% regional / 4% distant; 14% Black, 42% Japanese American, 16% Latino, 7% Native Hawaiian, 21% White

• Coffee: longer lifeno cleartotal or regular
• Fewer cancer deathsno clearany coffee type
• Decaf, 1+ cup a day-29%could be chance

More to know, from our read of the paper:

• HEAVY TOPIC: calm tone, no 'coffee beats cancer'; the PRIMARY result (total coffee, all-cause death) is null: say 'no clear link', not 'coffee does nothing'
• Decaf finding (>=1 vs 0 cups/day, HR 0.71 in Model 3 / 0.74 in Model 2) is a secondary exposure; only 142 people drank >=1 cup/day of decaf (33 drank >=2); authors flag chance from many exposure-outcome tests and that decaf drinkers may be more health conscious. Always pair it with 'could be chance'
• Decaf showed NO link with colorectal-cancer deaths (Model 3 HR 0.99, 0.51-1.89)
• Survival bias: people had to survive to the 2003-2008 follow-up survey (735 who died earlier were excluded); excluded participants more often had distant disease
• Self-reported coffee at one time point; few heavy drinkers (no >=4 cups category); no recurrence data

1,098 people with colorectal cancer in the Multiethnic Cohort (Hawaii and California) reported their coffee intake after diagnosis and were followed for about 10 years. Total and regular coffee were not clearly linked to death from any cause or from colorectal cancer. One or more cups of decaf a day was linked to 29% lower risk of death from any cause, but not to fewer colorectal cancer deaths, and the authors say it could be chance or healthier habits. Observational: linked, not proven.

We read the abstract and the full text of this paper and checked every number on this page against it (16 facts, checked on 2026-10-29). The study design is named above. Cohort and other observational studies are described as linked, not proven. Found an error? Email theis@onestudyaday.com.