Dietary trends over a 15-y period (2007–2022) in the Prospective Urban Rural Epidemiology (PURE) study

Abstract

Background Prospective cohort studies documenting long-term dietary trends across multiple world regions are lacking. Objectives To evaluate 15-y (2007–2022) changes in food and beverage consumption among adults aged 35–70 y living in 626 urban and rural communities across 16 high-income (HIC), middle-income (MIC) and low-income (LIC) countries. Methods In the Prospective Urban Rural Epidemiology Study, dietary intake was assessed using an abbreviated food-frequency questionnaire approximately every 3 y across four follow-up visits over a 15-y calendar period (2007–2022). Linear mixed-effect models and mixed-effect multinomial logistic regression were used to evaluate changes in food and beverage consumption over time. Dietary changes '0.05 servings/d (s/d) are presented as servings/wk (s/wk). Changes of '0.2 s/d change over the study period were considered modest. Participant demographics (age, sex, urban/rural residence, and education) closely matched national distributions in each country. Results A total of 120 782 participants provided at least one dietary assessment (median follow-up: 12.1 y). Consumption of animal-based products has generally increased, with clear regional differences. Milk increased in China (+0.21 s/d, +0.15 to +0.26), and in LICs (+0.08 s/d, +0.05 to +0.11); decreased in HICs (−0.10 s/d, −0.12 to −0.07) and in MICs (−0.21 s/wk, −0.28 to −0.15). Eggs increased in HICs, MICs, and in China up to 0.12 eggs/d (+0.11 to +0.14); no change in LICs. All regions increased chicken intake. Only HICs increased cooked fish consumption (0.28 s/wk, +0.06 to +0.50). By contrast, plant-based food intake has generally decreased, but these changes were generally small, and there was some regional variation. Fruit consumption decreased in HICs (−0.16 s/d, −0.18 to −0.14) and MICs (−0.09 s/d, −0.10 to −0.07); increased in China (+0.13 s/d, +0.11 to +0.14) and LICs (+0.22 s/wk, +0.16 to +0.29).