Screening for and Prevention of Diabetes in India: Need for Simple and Innovative Strategies

O besity, type 2 diabetes mellitus (T2DM), and cardiovascular diseases are rapidly escalating all over the world. According to the International Diabetes Federation, 366 million people have diabetes globally, a figure that is expected to rise to 552 million by 2030. ¹ The global healthcare expenditure for diabetes was 465 billion USD in 2011, equivalent to 11% of total health spending. ¹ This is predicted to rise to 595 billion USD by 2030. The rise in prevalence has been especially pronounced in the developing countries, particularly in South Asia. During the previous three decades, the prevalence of T2DM has doubled in India. Along with communicable diseases, these noncommunicable diseases pose a “double burden” of diseases in South Asian countries.

Rapid economic, demographic, and nutritional changes are occurring in India and other developing countries. With the widespread presence of transnational food companies, there is an increasing trend of consumption of nontraditional “fast foods” in urban areas, ² most noticeable in children. ³ Physical inactivity has increased because of increasing mechanization. Genetic predisposition and fetal programming because of intrauterine malnutrition could be other possible contributory factors. ⁴ All these are prime drivers for obesity, the metabolic syndrome, and T2DM in South Asians. Such a huge disease burden gets translated into enormous economic costs. The economic burden of diabetes in India was estimated to be 25–38 billion USD (1,230–1,837 billion Rupees) in 2010. ⁵

One of the approaches to contain the diabetes epidemic in India and other developing countries effectively is to identify the prediabetes/diabetes early in the disease course, so that appropriate actions could be taken for prevention and management. In other words, accurate and reliable biomarker(s) for prediction of diabetes and a valid diagnostic test are needed. Although biomarker research is emerging, recent development on glycosylated hemoglobin (HbA1c) as a diagnostic test for diabetes is a progressive but debatable step. There are some data to show that HbA1c may discriminate prevalence of retinopathy better than fasting plasma glucose (FPG). ⁶ Furthermore, it is affected less in acute situations; day-to-day and within-person variations of HbA1c are <2%, whereas those for FPG are 12–15%. ⁷ These characteristics make HbA1c an attractive test for the diagnosis of diabetes.

The study by Nazir et al. ⁸ reported in this issue of Diabetes Technology & Therapeutics attempts to look at the use of HbA1c as the diagnostic tool for diabetes in Asian Indians. In this study, prevalence rates of diabetes were compared in 2,188 subjects without known diabetes using the FPG level, the 2-h plasma glucose (PG), or HbA1c criterion. The prevalence of diabetes was 12.8% using the HbA1c criterion, which was 110% higher than the FPG criterion and 27% higher than the 2-h PG criterion. As discussed by the authors, this would increase the estimated population with diabetes from 62.4 million to 79.2 million in India, with significant implications for human and financial resources. Furthermore, it was also observed that individuals diagnosed using the HbA1c criterion had significantly lower levels of FPG, PG, and serum triglycerides than those diagnosed using FPG and PG criteria, implying that they had possibly lower cardiometabolic risk. The latter observation is intriguing and similar to a study on a Japanese population, in which impaired fasting glucose and HbA1c identified people with different cardiovascular risk profiles. ⁹

Is it time to propagate HbA1c as a diagnostic test for diabetes in developing countries, including India? Based on current evidence and the socioeconomic situation, and in contradiction to an announcement by the World Health Organization, ¹⁰ this step seems premature. Interesting findings from the study of Nazir et al. ⁸ need further discussion specific to the Indian population. First, standard laboratory estimation of HbA1c is not available at most locations in India, and any estimation from most laboratories should be interpreted with caution. Second, about one-fourth of the adult Indian population is anemic ¹¹ ; also, 3–14% of the population has thalassemia, ¹² making interpretation of HbA1c fallacious. Furthermore, with a recent elevation of blood glucose, the HbA1c value may be misleading. The high cost of this test, as has been discussed in detail by the authors, ⁸ is particularly important because the prevalence of diabetes is increasing in middle and low socioeconomic strata in India, and these patients may have difficulty in paying for a relatively expensive diagnostic test. In India's National Program for Diabetes, Cardiovascular Disease, and Stroke, particular emphasis has been given for research to develop a simple, cost-effective, reliable, and quick screening test. Point-of-care capillary blood glucose testing has been shown to have a good predictive value for diabetes ¹³ ; however, there is a need to define the cutoffs for random blood glucose as a screening for diabetes. ¹⁴ In this context, it is interesting to note that a single determination of blood glucose (FPG or 2-h PG) was a more precise correlate of insulin resistance and secretion than HbA1c among non-Hispanic whites. ¹⁵ Overall, a single blood glucose determination is simple, low cost, and more convenient (in the case of random PG measurements) and should be researched more in developing countries. Finally, a mandate has been given to the Indian Council of Medical Research to invite various biotechnology researchers to develop methods of noninvasive determination of blood glucose levels for screening.

On a larger discussion thread, screening for diabetes should be a low-priority intervention in developing countries mainly because of its high cost per quality-adjusted life years and low diagnostic yield. ¹⁶ However, it could be useful in two settings: first, in high-risk populations (e.g., specific racial groups such as Asian Indians, those living in urban areas of India, or those with other high-risk characteristics); and second, to decrease cost. The existing pool of healthcare workers engaged in other healthcare activities in India could also be trained for screening for diabetes, and low-cost tools of measurement of blood glucose (e.g., cheaper blood glucose testing strips) could be adapted. The development of the latter has again been mandated by the Ministry of Health, Government of India. ¹⁷ However, it remains to be seen if general screening of the population, as has been started in India, will lead to reduction in morbidity and mortality. It is interesting that one of the offshoots of such a screening program and its publicity could be an increase in awareness of the population regarding diabetes.

Until more research data are available from developing countries, a rational approach would be to identify high-risk individuals for screening and institute lifestyle changes according to ethnic-specific guidelines for diet ¹⁸ and physical activity ¹⁹ at population-specific body mass index and waist circumference cutoffs (lower than those of non-Asian ethnic groups). ²⁰ It is estimated that application of India-specific body mass index and waist circumference cutoffs would identify an additional 10–15% of the population as overweight/obese, requiring appropriate screening and lifestyle management, and this “intervention” itself may result in a deceleration of T2DM and cardiovascular diseases. An approach to identify high-risk individuals in a rural setting could be to train primary healthcare workers to select those with a family history of diabetes and high waist circumference/body mass index, ²¹ both low-cost activities. Finally, it is important to research simple and cost-effective solutions for screening and for identification of high-risk Asian Indians from an early age, preferably in childhood itself.