Background: Short-term clinical trials have shown the effectiveness of low-carbohydrate diets (LCDs) and low-fat diets (LFDs) for weight loss and cardiovascular benefits. We aimed to study the long-term associations among LCDs, LFDs, and mortality among middle-aged and older people.
Methods: This study included 371,159 eligible participants aged 50-71 years. Overall, healthy and unhealthy LCD and LFD scores, as indicators of adherence to each dietary pattern, were calculated based on the energy intake of carbohydrates, fat, and protein and their subtypes.
Results: During a median follow-up of 23.5 years, 165,698 deaths were recorded. Participants in the highest quintiles of overall LCD scores and unhealthy LCD scores had significantly higher risks of total and cause-specific mortality (hazard ratios [HRs]: 1.12-1.18). Conversely, a healthy LCD was associated with marginally lower total mortality (HR: 0.95; 95% confidence interval: 0.94, 0.97). Moreover, the highest quintile of a healthy LFD was associated with significantly lower total mortality by 18%, cardiovascular mortality by 16%, and cancer mortality by 18%, respectively, versus the lowest. Notably, isocaloric replacement of 3% energy from saturated fat with other macronutrient subtypes was associated with significantly lower total and cause-specific mortality. For low-quality carbohydrates, mortality was significantly reduced after replacement with plant protein and unsaturated fat.
Conclusions: Higher mortality was observed for overall LCD and unhealthy LCD, but slightly lower risks for healthy LCD. Our results support the importance of maintaining a healthy LFD with less saturated fat in preventing all-cause and cause-specific mortality among middle-aged and older people.
Keywords: aging; cohort study; low-carbohydrate diet; low-fat diet; mortality.
© 2023 The Association for the Publication of the Journal of Internal Medicine.