Cardiovascular risk assessment among type-2 diabetic subjects in selected areas of Bangladesh: concordance among without cholesterol-based WHO/ISH, Globorisk, and Framingham risk prediction tools.

Introduction: Cardiovascular disease (CVD) risk is higher among the subjects with type-2 diabetes mellitus (T2DM) in low- and middle-income countries, like Bangladesh. However, there is no relevant available online published data for this country. We aimed at assessing the 10-year CVD risk among T2DM subjects in selected areas of Bangladesh using the without cholesterol-based joint World Health Organization/International Society of Hypertension (WHO/ISH), Globorisk, and Framingham Risk Score (FRS) risk prediction tools, and also evaluating the concordance among these tools.

Methods And Materials: In this paper, we extracted a total of 327 subjects (40-60 years aged) from an observational study with 356 subjects, excluding those with diagnosed CVDs. The subjects were selected conveniently from purposively selected respective diabetic hospitals of Pirojpur and Dinajpur districts. We used the required respective variables of WHO/ISH, Globorisk, and FRS tools to predict CVD risks. The risks were categorized as low (<10%), moderate (10-<20%), high (20-<30%) and very high (≥30%).

Results: Subjects at moderate CVD risk were much higher identified by Globorisk (37.0%) and FRS (38.8%) compared to WHO/ISH (15.3%), and the same scenarios have also been observed for high (13.5%, 19.3% and 2.4%, respectively) and very high (5.5%, 17.4% and 1.8%, respectively) risks. There was fair level of concordance between WHO/ISH and Globorisk (PABAK-OS = 0.37; 95% CI 0.33-0.42; < 0.001), and Globorisk and FRS (PABAK-OS = 0.34; 95% CI 0.30-0.39; < 0.001). And, between WHO/ISH and FRS, it was none to slight level (PABAK-OS = 0.09, 95% CI 0.04-0.14; = 0.001).

Conclusions: A significant proportion of the selected study subjects is at moderate to very high risk of developing CVDs predicted especially by Globorisk and FRS compared to WHO/ISH, indicating low concordance. With and without cholesterol-based studies can answer the problem more clearly.