Validation of Indian Diabetic Risk Score for detecting Undiagnosed Diabetics in A Semi-Urban Population of Mangalore-South India

Abstract

Background: Diabetes is considered as an "iceberg disease". According to latest WHO estimates the number of people with diabetes worldwide in 2000 is 171 million which is likely to increase to at least 366 million by 2030.Indian Diabetes Risk Score is useful for identifying undiagnosed diabetic subjects in India and includes four risk factors: Age, Abdominal obesity, Family history of diabetes and Physical activity in predicting diabetes. Objectives: To Validate the Indian Diabetic Risk score for detecting undiagnosed diabetics in a Semi urban population. Material and Methodology: The study was a validation study conducted in the field practice area of K.S. Hegde Medical Academy, Mangalore in 2015. Universal sampling was done among 80 households to obtain a minimum sample size of 160 participants. Two from each household one being the highest age and one being the middle age among the households were enrolled. Individuals aged 20 years and above were included to validate the Indian Diabetic Risk score. Informed consent was taken before introducing the participants to the study and two Random blood sugars were checked. Results: Out of 160 participants 31.9% were male and 68.1% were female. Indian diabetic risk score >60 were 60%. Mean risk scores was 57.4±2x1.7. Out of 160 respondents, 86.9% were non-diabetic and 13.1% were detected diabetic. There was a significant difference in Indian diabetic risk score among diabetic and non-diabetics (p<0.001). Indian diabetic risk score >60 had sensitivity of 95.2% and specificity of 45.3% in predicting diabetes. Area under ROC was 0.719 (95% CI:0.62-0.81). Conclusion: Indian diabetic risk score can be used as a simple screening tool in predicting and detecting undiagnosed diabetics in the community.