**Background:** Type 2 diabetes (T2D) prevalence is rising globally, with Southeast Asia—particularly Indonesia—among the most affected regions. Prediabetes, defined as impaired fasting glucose (IFG) and/or impaired glucose tolerance (IGT), often precedes T2D. While dietary patterns are known to influence glucose control, the specific association between types of unhealthy food consumption and glucose metabolism in overweight or obese individuals remains unclear. This study aimed to determine the association between unhealthy food consumption and impaired glucose metabolism in Indonesian adults with overweight or obesity using nationally representative data.
**Methods:** This analysis used data from the 2018 Indonesia Basic Health Research (Riskesdas 2018), a population-based, cross-sectional, nationally representative survey conducted by the National Institute of Health Research Development, Ministry of Health, Indonesia. A total of 8,752 adults aged 18–50 years with BMI >23 kg/m² (WHO Asian criteria) and complete data on food frequency, physical activity, fasting glucose, and 2-hour postprandial glucose were included. Dietary intake was assessed using a validated questionnaire and food card, with food frequency categorized as frequent (≥1x/day) or rare (<1x/day). Blood glucose markers included fasting plasma glucose and 2-hour postprandial glucose, classified per ADA criteria: IFG (100–125 mg/dL with normal OGTT <140 mg/dL), IGT (OGTT 140–199 mg/dL with normal fasting glucose <100 mg/dL), or combined glucose intolerance (CGI, both IFG and IGT). Statistical analysis used Pearson's chi-square tests and multiple logistic regression across five models: unadjusted (model 1); adjusted for BMI (model 2); adjusted for physical activity (model 3); adjusted for BMI and physical activity (model 4); and fully adjusted for BMI, physical activity, age, and sex (model 5).
**Key Results:** Of the 8,752 participants, 16.6% had IFG, 25.6% had IGT, 13.9% had CGI, and the remainder had normal glucose regulation. Overweight accounted for 49.6% and obesity for 22.6% of participants. IGT was the most common impairment across nearly all subgroups. Sweet food consumption was significantly associated with IFG across all models; after adjustment for physical activity (model 2) and BMI plus physical activity (model 4), frequent sweet food consumption conferred approximately 15% higher odds of IFG (OR=1.153, 95% CI=1.047–1.268). Frequent grilled food consumption showed higher odds of IFG (model 2: OR=1.350, 95% CI=1.056–1.725). Processed food consumption had the strongest association with IFG: in the fully adjusted model (model 5), consumption >1x/day increased IFG risk by approximately 72% (OR=1.729, 95% CI=1.311–2.280). For IGT, high-fat food consumption was significantly associated in the unadjusted model (OR=1.124, 95% CI=1.032–1.225). Sweet food consumption was associated with CGI across models 2–5 (model 2: OR=1.173, 95% CI=1.034–1.331). Processed food consumption was consistently associated with CGI across all models; in model 5, frequent consumption increased CGI odds by approximately 62% (OR=1.620, 95% CI=1.150–2.283).
**Clinical Implications:** This study demonstrates that specific unhealthy food groups—particularly processed foods, high-fat foods, sweet foods, and grilled foods—are independently associated with impaired glucose metabolism in overweight or obese Indonesian adults. Processed food consumption emerged as the strongest risk factor for both IFG and CGI, while high-fat foods were most strongly linked to IGT. These findings support targeted dietary interventions focusing on reducing processed, high-fat, and sweet food intake as a strategy for prediabetes prevention in this high-risk population. The authors note that intensive lifestyle interventions can reduce progression from prediabetes to T2D by 40–70%, and that dietary patterns such as the Mediterranean and DASH diets may be protective. Limitations include the cross-sectional design (precluding causal inference), reliance on food frequency data without precise gram or calorie quantification, and lack of data on additional confounding biomarkers such as IGF-1 levels. Longitudinal studies and randomized controlled trials are needed to confirm these associations and establish causality.