Traditional health service use and cardiometabolic biomarker profiles among Indonesian adults with diabetes mellitus or hypertension: a national cross-sectional study

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Abstract Background Traditional health services (THS) are widely used in Indonesia, but their sociodemographic patterning and relationship with cardiometabolic biomarkers remain unclear. THS use among adults with physician-diagnosed diabetes mellitus and/or hypertension was assessed, and its associations with glycemic, lipid, and blood-pressure profiles were examined. Methods Data from the 2023 Indonesian Health Survey were analyzed for adults aged ≥ 15 years with complete exposure, biomarker, covariate, and survey-design data ( n = 2,703). THS use during the previous 12 months was defined as the primary exposure. The outcomes assessed were HbA1c, total cholesterol, LDL-C, HDL-C, systolic blood pressure, and diastolic blood pressure. Survey-weighted multinomial or binary logistic regression was applied, with stratification, clustering, and sampling weights incorporated. The false discovery rate was controlled using the Benjamini–Hochberg procedure, and propensity-score weighting was performed as a sensitivity analysis. Results THS use was reported by 1,309 respondents, corresponding to a weighted prevalence of 53%. Herbal preparations were used by 76.2% of users, whereas manual therapies were used by 52.9%. Higher odds of THS use were observed among urban residents (AOR1.28; 95% CI 1.05–1.56), self-employed respondents (AOR 1.49; 95% CI 1.10–2.02), and agricultural or fisheries workers (AOR 1.44; 95% CI 1.09–1.90). No association was detected between THS use and any biomarker. No modality-specific or pharmacotherapy-stratified association survived correction for multiple testing. Conclusions THS use was common and socially patterned, but no association with cardiometabolic biomarker profiles was detected. Because information on indication, product, dose, frequency, duration, and treatment timing was unavailable, the findings should be interpreted as population-level patterns rather than evidence of benefit or harm.