Socio-behavioral determinants of undiagnosed type 2 diabetes in middle-aged adults: a cross-sectional analysis

Late diagnosis of type 2 diabetes mellitus causes preventable complications and higher morbidity. Obesity, inactivity, poor diet, and smoking increase the risk of diabetes; however, their relationship with undiagnosed diseases in middle-aged adults requires further clarification.

Between January and June 2024, we enrolled 200 adults aged 35-45 years from outpatient clinics. The staff measured the height, weight, and body mass index of each participant. Participants completed the International Physical Activity Questionnaire and Mediterranean Diet Adherence Screener and reported their smoking histories. After overnight fasting, we measured the fasting blood glucose and HbA1c levels. Undiagnosed diabetes was defined according to the 2024 American Diabetes Association criteria (FBG ≥ 126 mg/dL or HbA1c ≥ 6.5%) in patients without a prior diagnosis. Statistical analyses included independent t-tests, chi-square tests, and multivariable logistic regression.

Undiagnosed diabetes was found in 29 participants (14.5, 95% CI 9.8-19.2). Those with undiagnosed diabetes had a higher body mass index (30.6 ± 2.6 vs. 27.6 ± 3.4 kg/m², p < 0.001), lower physical activity prevalence (10.3% vs. 21.6%, p < 0.018), and higher smoking rates (48.3% vs. 26.3%, p < 0.024). After adjusting for physical activity, smoking, and diet quality, body mass index remained independently associated with undiagnosed diabetes (OR 1.36 per kg/m², 95% CI 1.16-1.70, p < 0.001). Physical activity level (OR 1.54 per category decrease, 95% CI 1.30-1.92, p < 0.024) and smoking (OR 1.33, 95% CI 1.13-1.66, p < 0.005) were also independently associated with undiagnosed status. Model discrimination was good (area under the curve 0.74, 95% CI 0.67-0.81), with a 93.3% negative predictive value.

Approximately one in seven middle-aged adults in outpatient settings has undiagnosed type 2 diabetes. A higher body mass index, lower physical activity, and smoking were independently associated with an undiagnosed status. These readily assessable socio-behavioral factors should inform screening strategies and risk assessments in primary care.