Most human dietary studies gather data via participant self-reporting, which has its limitations. Using biomarkers to examine levels of specific dietary components is a step up from that, but it doesn't remove the core issues inherent in looking at specific dietary components in isolation. For example, we know that overall calorie intake level is enormously important in determining health and longevity, and that correlates with levels of different dietary components. Also, people who make better efforts to take care of their health tend to have better diets, but that effort extends beyond just diet. So levels of specific dietary components in human studies are going to correlate with all sorts of other line items that can impact health and longevity, such as exercise, calorie intake, amount of visceral fat tissue, conscientiousness in use of medical resources, and so on.
Correlation is not causation, but invariably when it comes to diet there are all sorts of vested interests willing to sell you the idea that you should believe otherwise. This study reports a large enough result to awake that contingent:
Polyphenols are naturally occurring compounds found largely in fruits, vegetables, coffee, tea, nuts, legumes and cereals. More than 8,000 different phenolic compounds have been identified in plants. Polyphenols have antioxidant, antiinflammatory, anticarcinogenic, etc. effects.
Polyphenols might have a role in the prevention of several chronic diseases, but evaluating total dietary polyphenol (TDP) intake from self-reported questionnaires is inaccurate and unreliable. A promising alternative is to use total urinary polyphenol (TUP) concentration as a proxy measure of intake. The current study evaluated the relationship between TUPs and TDPs and all-cause mortality during a 12-y period among older adult participants. The study population included 807 men and women aged 65 y and older from the Invecchiare in Chianti study, a population-based cohort study of older adults living in the Chianti region of Tuscany, Italy.
In conclusion, the research proves that overall mortality was reduced by 30% in participants who had rich-polyphenol diets (greater than 650 mg/day) in comparison with the participants who had low-polyphenol intakes (less than 500 mg/day). "[The] results corroborate scientific evidence suggesting that people consuming diets rich in fruit and vegetables are at lower risk of several chronic diseases and overall mortality." Moreover, the research stresses the importance of evaluating - if possible - food intake by using nutritional biomarkers, not only food frequency questionnaires.