Associations of diet quality and dietary adherence with metabolic control in adults with type 2 diabetes: A cross-sectional analysis
Mardiana Mardiana, Oktia Woro Kasmini Handayani, Eko Farida, Intan Zainafree, Safrina Oksidriyani, Safira Chairani Dimarti, Farida Dwi RokhmahObjectives
This study examined whether diet quality and dietary adherence were associated with metabolic biomarkers among older adults with type 2 diabetes mellitus (T2DM) enrolled in Indonesia’s Prolanis programme, after accounting for measured anthropometric and demographic covariates.
Methods
A cross-sectional study included 71 purposively sampled adults aged ≥60 years with T2DM at two primary healthcare centres in Semarang, Indonesia (October-December 2024). Diet quality and dietary adherence were assessed using a modified Alternative Healthy Eating Index 2010 (AHEI-2010) and the Perceived Dietary Adherence Questionnaire (PDAQ). Outcomes included HbA1c, fasting plasma glucose, and lipid parameters. Exploratory backward stepwise linear regression models included the dietary variables and measured demographic, anthropometric, blood pressure, and physical activity covariates.
Results
The study did not detect statistically significant associations between diet quality or dietary adherence and any metabolic biomarker (all p > 0.05). In exploratory models, age and hip circumference were inversely associated with HbA1c (R 2 = 0.207). Other retained variables were age for fasting plasma glucose (R 2 = 0.062), BMI for total cholesterol (R 2 = 0.049), sex for HDL cholesterol (R 2 = 0.132), waist-to-hip ratio and physical activity for triglycerides (R 2 = 0.078), and systolic blood pressure and physical activity for the cholesterol/HDL ratio (R 2 = 0.134). All variance inflation factors were below 2.0.
Conclusion
In this Prolanis-based cross-sectional sample, the study did not detect associations between diet quality or dietary adherence and glycaemic or lipid biomarkers after adjustment for measured covariates. This finding should not be interpreted as evidence that dietary factors are unimportant. The modest sample, cross-sectional dietary assessment, measurement error, and unmeasured treatment-related confounding may have limited the ability to detect associations.