Country-specific differences in the association between obesity risk factors in two high-income countries: a cross-sectional analysis from the Prospective Urban Rural Epidemiology (PURE) study

Abstract

Abstract Background Although Sweden and Canada are high-income countries with similar social structures, obesity prevalence is significantly higher in Canada. This study explored country-specific differences in the association between socioeconomic and behavioural risk factors for obesity in Sweden and Canada, using data from the Prospective Urban Rural Epidemiological (PURE) study. Methods This cross-sectional study included 9 790 adults aged 34–60 years from Canada (n = 6 652, 55% women) and Sweden (n = 3 138, 54% women). The Boruta algorithm was used to identify relevant factors that were associated with obesity among individual with normal weight (body mass index [BMI]: 18.5-<25 kg/m2) compared to individuals with obesity (BMI > 30 kg/m2). Logistic regression models with 95% confidence intervals (CI) estimated odds ratios (OR) of obesity by country in relation to selected risk factors for obesity. Results More Canadians than Swedes had obesity (26 vs. 16%, p-value: <0.001). Ultra-processed food (UPF) and the Alternative Healthy Eating Index (AHEI-2010) score were identified as the main drivers of obesity risk. The highest UPF intake group was strongly associated with obesity in both Canada (OR = 2.31 [CI = 1.57–3.37]) and Sweden (OR = 2.83 [CI = 2.30–3.49]). Canadian men had higher UPF intake and were found to have a significantly higher risk of obesity (p for interaction: 0.032 and 0.021 for middle and highest tertiles), compared to Swedish males. Among women, low socioeconomic status (rural residence (OR = 2.30 [CI = 1.66–3.17] vs. OR = 1.42 [CI = 1.19–1.70]), low income (OR = 4.75 [CI = 2.71–8.95] vs. OR = 2.62 [CI = 2.16–3.18]), not working (OR = 2.24 [CI = 1.51–3.29] vs. OR = 1.32 [CI = 1.07–1.63]), and unskilled occupation (OR = 5.08 [CI = 3.05–8.45] vs. OR = 1.78 [CI = 1.38–2.28]) was more strongly associated with obesity in Sweden than in Canada. Conclusions This study highlights important differences in obesity-related risk factors between Sweden and Canada. UPF consumption, the AHEI-2010 score and socioeconomic disadvantage emerged as key drivers of obesity, with notable sex- and country-specific patterns. These findings underscore the importance of tailored, context-specific public health strategies to address obesity in different national settings.