In the prospective population-based cohort study of adults aged 80 years or older, about three-fourths drank coffee, having started on average at the age of 19 years.
Among those who were dementia-free at baseline, coffee drinkers were 32% less likely to develop dementia over the next 5.5 years, compared with their counterparts who had never consumed the beverage, reported lead investigator Ugo Lucca, M.Sc., head of the laboratory of geriatric neuropsychiatry at the Mario Negri Institute for Pharmacological Research in Milan.
"Recent findings have suggested that coffee consumption may have neuroprotective effects," he noted. "Long-term caffeine consumption could, in fact, reduce amyloid beta production and aggregation through suppression of both beta- and gamma-secretase activities."
"What about reverse causality, with dementia causing a decrease in coffee consumption?" one attendee asked.
Lucca acknowledged that coffee consumption did indeed fall with increasing severity of dementia at baseline, but he also noted that 41% of participants with severe dementia still drank at least some. "The only explanation could be that, even in this population, drinking coffee is protective for survival," he said. "In a subpopulation of those with dementia, probably drinking coffee is not protective for survival."
The investigators studied 2,198 participants in the Monzino 80-Plus Study, conducted in the Varese province of Italy. Participants or relatives reported the number of cups of coffee consumed daily, in the past and at present.
Study participants were 90 years old and had five years of education, on average, and 73% were women. Their mean Mini-Mental State Examination score was 20.5.
Overall, about three-fourths of participants reported drinking coffee, with about one-third drinking two or more cups daily.
"Though the average daily coffee intake tends to decrease from midlife, coffee consumption remains common in the very old," Mr. Lucca observed. The average intake was 2.3 cups daily at the age of 50 years, but still 1.5 cups daily at the age of 87 years.
In the cross-sectional part of the study, the prevalence of dementia was about 51% among participants who drank less than one cup of coffee daily, 34% among those drinking one cup daily, and 23% among those drinking two or more cups daily, he reported.
In a multivariate analysis, participants who drank coffee were 29% less likely to have dementia (odds ratio, 0.71; P = .008).
In the longitudinal part of the study, the incidence of dementia over 5.5 years among the 1,097 participants who were free of dementia at baseline was significantly lower for coffee drinkers than for never-drinkers (P less than .0001), with a widening gap over time.
The incidence was 45% among participants drinking less than one cup of coffee daily, 31% among those drinking one cup daily, and 29% among those drinking two or more cups daily, Mr. Lucca reported.
In a multivariate analysis, compared with never-drinkers, coffee drinkers as a whole were about one-third less likely to develop dementia (hazard ratio [HR], 0.68; P = .006). The benefit was similar whether they consumed one cup daily (HR, 0.68; P = .01) or two or more cups daily (HR, 0.66; P = .02).