Does diet map with mortality? Ecological association of dietary patterns with chronic disease mortality and its spatial dependence in Switzerland

The aim of the study was to investigate the associations between dietary patterns and chronic disease mortality in Switzerland using an ecologic design and to explore the spatial dependence of these associations, i.e. the tendency of near locations to present more similar values than randomly expected and distant locations to present more different values. Data of the cross-sectional National Nutrition Survey menuCH (n = 2,057) were used to compute hypothesis- (Alternate Healthy Eating Index, AHEI) and data-driven dietary patterns. District-level standardised mortality ratios (SMR) were calculated using the Swiss Federal Statistical Office mortality data and were linked to dietary data by the participants’ place of residence. Quasipoisson regression models were fitted to investigate the associations between dietary patterns and chronic disease mortality; Moran’s I statistics were used to explore the spatial dependence. Compared to the first, the fifth AHEI quintile (representing the highest diet quality) was associated with district-level SMR (95% confidence interval) of 0.95 (0.93-0.97) for cardiovascular disease, 0.91 (0.88-0.95) for ischemic heart disease, 0.97 (0.95-0.99) for stroke, 0.99 (0.98-1.00) for all-cancer, 0.98 (0.96-0.99) for colorectal cancer, and 0.93 (0.89-0.96) for diabetes. Additionally, the Swiss traditional and Western-like patterns were associated with significantly higher district-level SMR for cardiovascular disease, ischemic heart disease, stroke, and diabetes (ranging from 1.02 to 1.08) compared to the Prudent pattern. Finally, significant global and local spatial dependence was identified, with very similar results observed across hypothesis- and data-driven dietary patterns. To conclude, significant associations were observed between dietary patterns and mortality from various chronic diseases, and chronic disease mortality clustered locally, with high similarities between hypothesis- and data-driven dietary patterns. Further analyses including spatial components in regression models would allow to better understand local factors influencing diet-disease associations and to identify regions where nutritional interventions are needed the most