“Coffee extends lifespan”

The strongest single piece of evidence pools 21 long-running population studies covering 997,464 people and 121,915 deaths [1]. The lowest risk of dying from any cause came at 4 cups a day, about 16% below non-drinkers, and the lowest risk of dying from heart and blood-vessel disease came at 3 cups a day, about 21% lower [1]. The curve flattens beyond that rather than continuing to fall, so more coffee is not better coffee.

The same pooled analysis found no link at all between coffee and cancer deaths [1]. That is worth stating plainly, because coffee is often marketed as a blanket protector: the survival signal here sits in heart and circulatory causes, not in cancer.

A separate study followed 521,330 adults across ten European countries for an average of 16 years [2]. The heaviest coffee drinkers had lower death rates than non-drinkers - about 12% lower in men and 7% lower in women - and the pattern held in every country, even though brewing habits across Europe range from Scandinavian boiled coffee to Italian espresso.

That study also found much lower death rates from digestive disease among coffee drinkers, and blood tests in a smaller subgroup showed lower levels of liver enzymes, the proteins that leak into the blood when liver cells are damaged [2]. It flagged one association pointing the other way too: among women, higher coffee drinking went with slightly more deaths from ovarian cancer.

Caffeine does not look like the active ingredient. A pooled analysis of cohort studies covering more than 10 million people found decaffeinated coffee linked to lower death rates much like caffeinated coffee [3], which points to the other compounds in the bean - chlorogenic acids and related polyphenols, plant chemicals with antioxidant and anti-inflammatory effects.

All of this is observational, and people who drink coffee differ from people who do not in ways that are hard to strip out - smoking habits in particular have historically travelled with coffee drinking and can bend the numbers. Nobody has run, or realistically could run, a decades-long trial that randomly assigns people to coffee. The honest reading is that moderate coffee drinking is very unlikely to be shortening lives and may be doing some good, but the evidence cannot show that the coffee itself is what adds the years.