Dietary patterns modify the association between body mass index and mortality in older adults.

BACKGROUND

The potential modifying roles of dietary patterns in the association between body mass index (BMI) and mortality in older adults remain unclear. This study aimed to examine the stratified and combined associations of dietary patterns and BMI with all-cause, cancer and cardiovascular disease (CVD) mortality.

METHODS

This prospective cohort study included 3982 Chinese community-dwelling older adults between 2001 and 2003. A 280-item validated food frequency questionnaire was used to calculate five dietary indies: Diet Quality Index-International (DQI-I), Dietary Inflammatory Index (DII), Mediterranean Diet Score (MDS), Dietary Approaches to Stop Hypertension (DASH), and Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND) diet. Higher DQI-I scores indicate better overall diet quality, while higher DII scores represent a pro-inflammatory diet. Higher MDS, DASH, and MIND scores indicate greater adherence to these healthy dietary patterns. Mortality data were obtained from official records. Hazard ratios (HRs) and 95 % confidence intervals (CIs) were estimated using Cox proportional hazards models.

RESULTS

Over a median follow-up of 16.8 years, there were 1879 all-cause deaths, 561 cancer deaths, and 386 CVD deaths. The J-shaped associations between BMI with all-cause and cancer mortality were weaker in healthier dietary patterns (DQI-I, MDS, DASH, MIND ≥median or DII <median) than in unhealthier dietary patterns (DQI-I, MDS, DASH, MIND <median or DII ≥median). Lower DQI-I scores were associated with increased all-cause mortality risk in participants with underweight (HR: 1.78; 95 % CI: 1.40-2.26) or obesity (HR: 1.25; 95 % CI: 1.02-1.53), while higher DQI-I scores did not have significantly higher risk in participants with underweight (HR:1.24, 95 % CI: 0.94-1.65) or obesity (HR:1.03, 95 % CI: 0.82-1.28), with those having higher DQI-I scores and normal weight as the reference. Higher DQI-I scores attenuated the excess risk of cancer mortality in the underweight or the obese. Similar trends were observed for other dietary patterns. For CVD mortality, lower DII and higher MIND scores reduced the elevated risk associated with obesity, but this modifying effect was not observed in other dietary patterns.

CONCLUSIONS

Higher diet quality attenuated the increased risks of all-cause and cancer mortality associated with underweight or obesity in older adults. Anti-inflammatory and antioxidative diets may protect against CVD mortality associated with obesity.