<![CDATA[Dietary patterns may contribute to diabetes burden, but evidence from Indonesian provincial populations remains limited. This study assessed associations between selected processed/instant food and beverage indicators and doctor-diagnosed diabetes mellitus (DM) in West Sumatera. We analysed 31,921 survey respondents from West Sumatera using complex-sample weights. DM was defined as self-reported doctor diagnosis. Exposures were ever vs never consumption of salty foods, grilled foods, preserved meat/fish with additives, carbonated soft drinks, and instant noodles. Multivariable survey logistic regression produced adjusted odds ratios (AORs) with 95% confidence intervals (CIs). The weighted prevalence of diagnosed DM was 1.4% (≈71,486 of 5,272,303). In adjusted models, respondents reporting ever consumption of preserved meat/fish (AOR = 0.64; 95% CI: 0.47–0.89), carbonated soft drinks (AOR = 0.38; 95% CI: 0.28–0.52), and instant noodles (AOR = 0.56; 95% CI: 0.40–0.78) had lower odds of doctor-diagnosed diabetes mellitus than those reporting never consumption. Salty foods (AOR = 0.97; 95% CI: 0.64–1.46) and grilled foods (AOR = 1.27; 95% CI: 0.88–1.82) were not significantly associated. These inverse associations should be interpreted cautiously because they are likely to reflect reverse causality, post-diagnosis dietary modification, and residual confounding rather than protective effects of these foods. The prevalence of doctor-diagnosed diabetes mellitus in West Sumatra was low. Although several processed-food indicators showed inverse associations with diagnosed diabetes mellitus, these findings should not be interpreted as evidence of protective effects. Instead, they are more likely explained by reverse causality, behavioral modification following diagnosis, underdiagnosis, and residual confounding. Prospective longitudinal studies are warranted to clarify the causal relationship between dietary behaviors and diabetes risk.]]>
