Diabetes mellitus is a very common and serious disease in many American Indian tribes, Indians, and many other populations in the world. Several well-known risk factors such as parental diabetes, genetic markers, obesity, diet are considered as the main risk factors for diabetes mellitus, while the precise nature of the gene or genes remains unknown.

The Pimas, Indians, and many other population in the world now suffer from the high prevalence rates of diabetes. Epidemiological research often seeks to identify a causal relationship between the risk factors and the disease. In the present article, two data sets from two different study groups (one for the Pima Indian mothers (768 subjects) and the other for the Indian medical students (64 subjects) are analyzed to determine the causal factors of diabetes mellitus. This article aims to identify the determinants of diabetes mellitus in the Pima Indian mothers and Indian young medical students.

The causal factors for diabetes mellitus of the Pima Indian mothers and young Indian medical students are identified. Statistical significant causal factors, namely, triceps skin fold thickness (P-value < 0.01), serum insulin (Pvalue < 0.01), body mass index (or obesity) (P-value < 0.01), diabetes pedigree function (P-value = 0.06), age (P-value < 0.01) are identified as the determinants of diabetes mellitus in the Pima Indian mothers. In the young Indian medical students, age (P-value = 0.04), body mass index (P-value < 0.01), family history of diabetes mellitus (P-value < 0.01), sex (P-value < 0.01), low density lipoprotein (P-value = 0.01), total cholesterol (P-value = 0.11), serum triglyceride (P-value < 0.01), family blood pressure (P-value < 0.01), dietary habits like eating outside (P-value < 0.01) are identified as the determinants of diabetes mellitus. The effects of different causal factors on diabetes mellitus are explained based on probabilistic models.

Impacts of biochemical parameters, personal characteristics, family history, and dietary factors on human plasma glucose concentration are explained based on mathematical relationships. The results of the present analyses support many earlier research findings. However, these analyses also identify many additional casual factors that explain the mean and the variance of plasma glucose concentration, which earlier researches have not reported.