Last Updated on March 16, 2022 by David Vause
Came across an interesting study in PLOS Medicine https://journals.plos.org/plosmedicine/ on diet and longevity: Estimating impact of food choices on life expectancy: A modeling study. Based on data from the Global Burden of Disease study (2019), published in The Lancet, the researchers modeled the effect of three diets on life expectancy (LE) using multivariate regression. The diets compared were the extant typical Western diet (TW), an optimized diet (OD) where longevity was maximized, a feasibility approach diet (FA). This last eating pattern represents a mid-point between TW and FA which is considered more feasible for European and American consumers.
| Food | Western Diet (oz) | Feasibility Approach (oz) | Optimal Diet | Comments on Optimal Diet Portions |
|---|---|---|---|---|
| Whole grains (fresh weight) | 1.75 | 4.8 | 7.9 oz (2.6 oz dry weight) | ~ 7 servings: 2 thin slices of rye bread, 1 small bowl of whole-grain cereal, some whole grain rice |
| Vegetables | 8.8 | 11.5 | 14.1 | ~ 5 servings: 1 big tomato, 1 sweet pepper, mixed salad leaves, half an avocado, small bowl of vegetable soup |
| Fruits | 7 | 10.6 | 14.1 | ~ 5 servings: banana, orange, kiwi, a handful of berries |
| Nuts | 0 | 0.45 | 0.89 | 1 handful of nuts |
| Legume | 0 | 3.5 | 7.1 | 1 big cup of soaked lentil/beans/peas |
| Fish | 1.8 | 4.4 | 7.1 | 1 big slice of fish |
| Eggs | 1.8 | 1.3 | 0.9 | 1/2 egg |
| Milk/dairy | 10.6 | 8.8 | 7.1 | 1 cup of yogurt |
| Refined grains | 5.3 | 3.5 | 1.8 | e.g., refined grains in bread if mixed whole/refined wheat |
| Red meat | 3.5 | 1.8 | 0 | |
| Processed meat | 1.8 | 0.9 | 0 | |
| White meat | 2.6 | 2.2 | 1.8 | |
| Sugar-sweetened beverage | 17.6 | 8.8 | 0 | |
| Added plant oils | 0.9 | 0.9 | 0.9 |
The health gains from FA and OD arise mostly from reductions in cardiovascular disease, cancer, and diabetes.
The impact of diet on LE is affected by the age at which the diet is adopted. Below is a table of the age-specific gains in LE at different ages for the diets compared to the TE. The LE values are for persons living in the U.S.
| Age | Male FA Gain | Female FA Gain | Male OD Gain | Female OD Gain |
|---|---|---|---|---|
| 20 | 7.3 | 6.2 | 13.0 | 10.7 |
| 40 | 6.5 | 5.7 | 11.7 | 10 |
| 60 | 4.8 | 4.5 | 8.8 | 8 |
| 80 | 1.9 | 2.0 | 3.4 | 3.4 |
I find a few things interesting about this study. Note that it essentially describes the Mediterranean Diet as its optimal diet. This is unremarkable to me. There are no recommended amounts of red or processed meats. Two items that contribute substantially to healthier diets are nearly absent from the average daily consumption of the TW: legumes and nuts. The amount of added oil is minimal; I presume this means salad dressing too.
I acknowledge that correlation does not mean causation. However, the data sets used are worldwide and across all age populations. The dose-response is in direct proportion for all age groups. I strongly suspect causation is indicated.
