1 Introduction

Chronic non-communicable disease Diabetes and related metabolic non-communicable diseases (NCDs), such as obesity and hypertension, are no longer exclusive to rich and industrialized countries and India is no exception to this trend.1 Type 2 diabetes might be viewed as the model for chronic NCDs. While type 1 diabetes and other forms make up 5% to 10% of all cases of diabetes, type 2 diabetes continues to be by far the most prevalent type (over 90%) and has reached epidemic levels throughout the world. Globally, in 2021, it is estimated that 537 million people have diabetes, and this number is projected to 783 million by 2045. The second highest number of people with diabetes in the world currently is in India (74.2 million) and these numbers are expected to increase to 124.9 million by 2045 (Table 1).2 Given that diabetes is becoming increasingly prevalent in India, reliable epidemiological data is essential for estimating the effects and determinants of diabetes and facilitating the development of prevention and control methods.

Table 1: Countries/territories ranked by numbers of diabetes in 2021 and 2045.2
Full size table

2 Studies on Diabetes in India (Prior to ICMR-INDIAB Study)

There were significant data gaps in India regarding the distribution, trends, determinants, and effects of diabetes and related metabolic NCDs. Even when data are available, there is significant regional variation, and the inconsistent quality of the data reduces their usefulness. Numerous epidemiological studies have been conducted in India over the past 20 years to gauge the prevalence of NCDs such diabetes, hypertension, obesity, and others. The earlier studies carried out in India, however, have a number of drawbacks, including being primarily regional, having small sample sizes, low response rates, using a variety of diagnostic criteria, having issues with sample design, lacking standardization, having measurement errors, and reporting results insufficiently.3 As a result, a lot of the information that is available is either not usable or given in an incomplete manner, giving policymakers erroneous estimates.

There were not many multicenter studies that have looked at the prevalence of diabetes in India. In early 1970s, the Indian Council of Medical Research (ICMR), utilizing standardized methods and sample methodologies, conducted a systematic collaborative study on diabetes in 34,194 adults (aged > 15 years) across six regions of the nation.4 Diabetes was diagnosed using capillary blood glucose and the overall prevalence was reported to be 2.1% in urban areas and 1.5% in rural regions, while the prevalence among adults aged 40 years and above was 5% in urban areas and 2.8% in rural areas. Later, in six major metropolitan areas, the National Urban Diabetes Survey5 found an overall age-standardized prevalence of 12.1% for diabetes and 14% for impaired glucose tolerance (IGT). According to the Prevalence of Diabetes in India Study (PODIS),6 which was conducted in 49 urban and 59 rural locations, small towns and rural areas had diabetes prevalence rates of 5.9% and 2.7%, respectively. In the WHO-ICMR NCD Risk Factor Surveillance Study conducted between 2003 and 2005, the prevalence of self-reported diabetes was reported to be 4.5% among individuals aged 15–64 years in six different states in urban and rural settings.7 Another cross-sectional study of workers and their families from 11 medium-to-large enterprises in five urban locations, including Delhi, Hyderabad, Bangalore, Chennai, and Trivandrum, reported a 10.1% overall prevalence of diabetes.8

Therefore, a large representative population-based study was required to obtain estimates of diabetes and other related NCDs such obesity, hypertension, and dyslipidemia at the state and rural–urban levels. This requirement is specifically addressed by the Indian Council of Medical Research-India Diabetes (ICMR-INDIAB) research, which offers precise and thorough state- and national-level data on the prevalence of diabetes and other metabolic NCDs in India. The ICMR, New Delhi and the Department of Health Research (DHR), Government of India supported the large multicentre ‘ICMR- India Diabetes [INDIAB] Study’, which is nationally coordinated by the Madras Diabetes Research Foundation (MDRF), Chennai. The Northeast component of the ICMR-INDIAB was coordinated by the Regional Medical Research Centre, Dibrugarh along with MDRF. This chapter will provide a compendium of the findings reported from the ICMR-INDIAB study.

3 The ICMR-INDIAB Study

The ICMR-INDIAB study is a cross-sectional, population-based survey of adults aged ≥ 20 years.[9,10,11,12,13,14,15] The study sampled 33,537 urban and 79,506 rural residents (overall n = 113,043) in 30 states/Union Territories (UTs) and the NCT using a stratified multistage sampling design.9 To obtain a truly representative sample of the population, a three-level stratification based on geography, population size, and socioeconomic status (SES) of each state was used. The primary sampling units were villages in rural areas and census enumeration blocks in urban areas. Using a systematic sampling method, 24 and 56 households were selected from urban and rural areas, respectively. Door-to-door assessment was done and from each household, one individual was selected based on the WHO Kish method, to avoid selection bias with respect to sex and age. The primary objectives of the study were to determine the prevalence of diabetes mellitus and prediabetes in India by estimating the state-wise prevalence of the same and to compare the prevalence rates in urban and rural areas across the country. The additional objectives of the study were to determine the prevalence of metabolic NCDs including hypertension, dyslipidemia, obesity and coronary artery disease (CAD) among subjects with and without diabetes and to assess the level of glycemic control among individuals with self-reported diabetes.

In view of the complexity of the study and the logistics involved, the study was planned in phases. The different phases of the study are presented in Fig. 1. The study was conducted in a phased manner between 2008 and 2020. Phase I, was conducted from 2008 to 2010, while between 2011 and 2020, the remaining phases were conducted as follows: Phase II survey period, 2012–2013; North East Phase survey period, 2011–2017; Phase III survey period, 2017–2018; Phase IV survey period, 2018–2019 and Phase V survey period between 2019 and 2020.

Figure 1:
figure 1

Phases of the ICMR-INDIAB study.

Full size image

4 Results from the ICMR-INDIAB Study

4.1 Diabetes

Data on diabetes and prediabetes from 15 states/UT have been published and the state-wise prevalence reported.10 In these 15 states/UT, of the 60,918 individuals approached, 57,117 (16,909 urban and 40,208 rural) participated (response rate, 94%), of whom 54,128 gave blood samples. An interviewer-administered questionnaire was used to gather data on all study participants demographics, behaviors, and medical history as well as their awareness of diabetes. As capillary blood glucose (CBG) has been demonstrated to be a feasible alternative for screening for diabetes and prediabetes in epidemiological studies in developing countries where obtaining venous samples is difficult or even impossible in some circumstances, fasting and 2-h post-glucose CBG were used for the diagnosis of diabetes. Additionally, a fasting venous sample was taken from each fifth individual, and lipids (total cholesterol, triglycerides, and HDL cholesterol) were measured. Diabetes was defined as individuals diagnosed by a physician and on glucose-lowering medications (self-reported) and/or those who had a fasting CBG ≥ 126 mg/dl and/or a 2 h post-glucose CBG value ≥ 220 mg/dl.16

The overall prevalence of diabetes was 7.3% throughout all 15 states/UT, ranging from 4.3% in Bihar to 10.0% in Punjab. The mainland has a significantly higher prevalence (8.3%) than the northeast (5.9%). In general, the prevalence of diabetes in urban areas (11.2%) was double that in rural areas (5.2%). Figure 2 presents the weighted prevalence of diabetes in the ICMR-INDIAB study population. The ratio of self-reported diabetes to newly diagnosed diabetes was 1:0.9 overall; however, it was lower in rural areas (1:1.5) compared to urban areas (1:0.7). The ratios were comparable between the northeast states (1:0.9) and the mainland states (1:0.8). Among both urban and rural settings, diabetes prevalence began to increase between the ages of 25 to 34 years.

Figure 2:
figure 2

Weighted prevalence of diabetes in the ICMR- INDIAB study population (14 States & 1 Union Territory of India). 10

Full size image

ICMR-INDIAB study reported that the prevalence of diabetes was higher in the more economically developed states, and even within states, individuals with medium or high SES were more likely to have diabetes than individuals with low SES. However, individuals with low SES were more likely to have diabetes in seven of the more economically developed states. In contrast, in rural areas of all the states surveyed, individuals with higher SES had a higher prevalence of diabetes. This research implies that the diabetes epidemic has advanced farther in more rich states urban areas, where those from less affluent backgrounds are more likely to develop diabetes than those from wealthier backgrounds. The epidemiological transition appears to be less complete in rural areas of India, where diabetes continues to be a disease of the wealthier social classes. The extension of the diabetes epidemic to these socially and economically marginalized and vulnerable groups has major implications for the nation's health and socioeconomic advancement. This warrants immediate adoption of efficient preventative measures to curb the epidemic.10

4.2 Prediabetes

Isolated impaired fasting glucose was diagnosed if individuals had fasting CBG of at least 110 mg/dl and less than 126 mg/dl, and 2 h post-glucose CBG less than 160 mg/dl.17 Isolated impaired glucose tolerance was diagnosed if individuals had 2 h post-glucose CBG of at least 160 mg/dl and less than 220 mg/dl, and fasting CBG less than 110 mg/dl.16 Prediabetes was defined as individuals with impaired fasting glucose [IFG] or impaired glucose tolerance [IGT] or both. The overall prevalence of prediabetes was 10.3% (95% CI 10.0–10.6%) across all 15 states/UT analyzed, ranging from 6.0% (5.1–6.8%) in Mizoram to 14.7% (13.6–15.9%) in Tripura. With the exception of Punjab and the northeast, all of the states examined in Phase II had higher rates of prediabetes than diabetes. In all states surveyed, with the exception of Bihar, Manipur, and Meghalaya, the prevalence of isolated impaired fasting glucose was 6.5% (6.3–6.7%), which was more than twice that of isolated impaired glucose tolerance (2.8%, 2.7–3.0%).

Except in the older age group of 65 years and above, the prevalence of prediabetes was higher in urban areas among all age groups. There were no significant differences in prevalence of prediabetes between men and women.

4.3 Knowledge of Diabetes

Evidence shows that raising awareness of diabetes and its consequences has several benefits, such as improving adherence to treatment and reducing the complications related to diabetes.18 The ICMR-INDIAB research provides data at the national level on the awareness of diabetes. In Phase I of the study involving four states/UT, both the general public and those with diabetes had their awareness and knowledge of diabetes evaluated.19 Only 43.2% of the participants in the survey had ever heard of diabetes. Residents in urban areas were more aware of diabetes than those in rural areas (58.4 vs. 36.8%). In all four regions, urban inhabitants had higher awareness rates than rural ones, with Tamil Nadu having the highest rates, followed by Maharashtra, Jharkhand, and Chandigarh. Overall, 46.7% of men and 39.6% of women said they were aware of the condition called diabetes. In all four regions, with the exception of Chandigarh, this study found that men were more aware of diabetes than women were.19

Among the general population, 41.5% knew about a condition called diabetes, of whom, over 80.0% knew that the prevalence of diabetes was increasing. Among the self-reported diabetes, 93.0% were aware that the prevalence of the disease was rising. In the general population and among self-reported diabetes, 56.3% and 63.4%, respectively, were aware that diabetes could be prevented, and 51.5% and 72.7% knew that it may have an impact on other organs. When asked whether they believed diabetes could harm other organs, the general population most frequently mentioned their feet (54%) followed by their eyes (52.3%), kidneys (36.3%), heart (33.6%), and nerves (18.7%). It is worrying that even individuals with diabetes had relatively low levels of awareness regarding diabetic complications (eyes: 73.5%, feet: 61.3%, kidneys: 47.9%, heart: 45.1%, and nerve issues: 26.8%), albeit better than among those without diabetes.19

4.4 Achievement of Diabetes Treatment Targets in Indians

The ICMR-INDIAB study reported for the first time the status of diabetes control in the whole of India based on a nationally representative sample of 5,789 individuals with known diabetes.15 Achievement of treatment goals were defined as glycemic control (A) A1c < 7·0%; (B) blood pressure control < 140/90 mmHg; and (C) cholesterol: LDL cholesterol < 100 mg/dl. Only a third of individuals with known diabetes in India have good control of diabetes, fewer than half have good control of blood pressure and LDL cholesterol, and only 7.7% meet all three targets (A1c, BP, and cholesterol targets). The study revealed that 36.3% of the population with diabetes achieved good glycemic control, 48.8% achieved good blood pressure control and 41.5% achieved LDL cholesterol control (Fig. 3). Higher education, male sex, rural residence and shorter duration of diabetes were associated with better achievement of the combined goals of diabetes treatment. 16.7% of the population reported using a blood glucose monitor to check their blood sugars at home. Only 36.9% of those on insulin performed any self-monitoring of blood glucose, notwithstanding the guidelines that all such individuals should self-monitor their glucose levels regularly. Figure 4 shows the lifestyle health habits among individuals with self-reported diabetes. Fewer than 20% of the population with diabetes reported consuming three servings of fruits and vegetables per day (compared to the WHO recommendation of five servings a day). Fewer than 25% of people with diabetes performed moderate to vigorous physical activity. The study results suggest that there is considerable scope for improving adoption of health habits in the Indian population.

Figure 3:
figure 3

Diabetes ABC goals of treatment among individuals with self-reported diabetes. 15

Full size image
Figure 4:
figure 4

Lifestyle health habits among individuals with self-reported diabetes.15

Full size image

The study also highlights that the poor achievement of glycemic targets despite widespread use of anti-diabetic drugs suggests a lack of timely escalation of treatment, which could be due to insufficient monitoring and follow-up. Regular follow-up with health-care providers can help to achieve better glycemic control and prevent complications. Strengthening public health-care facilities throughout the country will be an effective way to improve overall diabetes care in India.

These findings suggest that the achievement of treatment goals and adoption of healthy behaviors remains suboptimal in India. These findings underscore the need for better control of glycemia, blood pressure and lipid parameters, so as to reduce the risk of diabetes related complications in Indians. These results should drive Government policy to strengthen diabetes care at the primary, secondary and tertiary levels in India. These results can help in formulating Government policy regarding priorities for improving diabetes care delivery and surveillance in India.

4.5 Cardiometabolic Risk Factors in Asian Indians: ICMR-INDIAB Study

4.5.1 Obesity

Generalized obesity (GO), defined as a BMI of less than 25 kg/m2 and abdominal obesity (AO), which is defined as a waist circumference of less than 90 cm for men and less than 80 cm for women is increasing in both sexes. Combined obesity (CO) was defined as the presence of both GO and AO. In the ICMR-INDIAB study, the prevalence of GO, AO and CO were evaluated. Of the four states surveyed in Phase I of ICMR-INDIAB, Chandigarh (15.9%) had the highest prevalence of overweight, followed by Tamil Nadu (15.2%), Maharashtra (11.3%), and Jharkhand (7.8%).12 Residents of Tamil Nadu, Maharashtra, Jharkhand, and Chandigarh had GO prevalence rates of 24.6, 16.6, 11.8, and 31.3%, respectively. The prevalence of AO and CO increased along with the rate of overweight. In all four of the study's regions, urban inhabitants had much greater rates of GO, AO, and CO than rural residents did. Female gender, hypertension, diabetes, higher socioeconomic status, physical inactivity, and urban residency were among the risk variables associated to GO, AO, and CO. Age was strongly linked with AO and CO in all four of the regions analyzed, but not with GO.

4.5.2 Hypertension

The third most significant risk factor for the attributable burden of disease in south Asia is high blood pressure.20 In India, hypertension is becoming more common. According to the ICMR-INDIAB study, the overall age-standardized prevalence of hypertension, based on the Joint National Committee 7 Criteria,21 was 26.3% in the four regions surveyed (self-reported hypertension prevalence was 5.5% and newly diagnosed hypertension prevalence was 20.8%).11 Tamil Nadu had the highest age-standardized prevalence of hypertension (27.6%), followed by Chandigarh (25.8%), Maharashtra (25.0%), and Jharkhand (23.8%). In all three states and the UT, the prevalence of hypertension was higher in urban than in rural areas, and newly diagnosed hypertension was more common than self-reported hypertension. The ratio of newly diagnosed to self-reported hypertension in the four states/UT was 3.8:1. The prevalence of hypertension increased with increasing age. Even in the age group of 20–24 years, the prevalence of hypertension ranged from 5.4 to 13.9% in urban areas and from 9 to 10% in rural areas. Even after controlling for confounding factors, salt intake of 6.5 g per day was associated with a 1.4 times greater risk of hypertension.

4.5.3 Dyslipidemia

Asian Indians are recognized to have a distinct pattern of dyslipidemia, characterized by lower HDL cholesterol, higher triglyceride levels, and a higher proportion of small dense LDL cholesterol.22 Lipid levels were assessed in every fifth individual (n = 2042 participants) in the ICMR-INDIAB study to determine the pattern and prevalence of dyslipidemia in India.13 According to National Cholesterol Education Programme (NCEP) guidelines, the prevalence of dyslipidemia, defined as having at least one lipid abnormality, was 79% overall in the four regions examined, with Chandigarh having the highest rates (82.9%), followed by Jharkhand (80%), Maharashtra (77%) and Tamil Nadu (76.9%). 13.9% of people had hypercholesterolemia (serum cholesterol levels below 200 mg/dl), and 29.5% had hypertriglyceridemia (serum triglycerides levels below 150 mg/dl). In all four of the states/UT surveyed, low HDL-C (HDL cholesterol levels < 40 mg/dl for males and < 50 mg/dl for females) was the most prevalent lipid abnormality (72.3%). In about 45% of the individuals, low HDL-C was present as an isolated abnormality. In 11.8% of individuals, high LDL-C values was observed. In each of the four regions under study, no disparities between urban and rural areas were found. 7.7% of the population studied had all three abnormal lipid abnormalities (hypercholesterolemia, hypertriglyceridemia, and low HDL-C), while 4.8% of the population had all four lipid abnormalities (hypercholesterolemia, hypertriglyceridemia, low HDL-C, and high LDL-C). Only 21.1% of the population were found to have no lipid abnormalities. Obesity, diabetes, and dysglycemia were found to be the significantly associated with dyslipidemia.

4.6 Role of Migration in Diabetes

The majority of NCD determinants are social and environmental, and a growing body of evidence indicates that migration, globalization, and urbanization have a considerable impact as NCD risk factors.23 Research have shown that migrants may develop unhealthy lifestyle habits like poor diet and inactivity as part of the acculturation process when they relocate to new societies, which raises their risk for NCDs including diabetes and cardiovascular disease.24 The ICMR-INDIAB study is the first research from India to assess and evaluate the prevalence of diabetes and other metabolic disorders among urban and rural residents of 30 Indian states and union territories who are migrants and non-migrants.25 Individuals who moved to a different place from their place of birth and had resided in the new location for at least one year were considered as migrants. The 113,043 individuals who participated in study were categorized based on the migration details into non-migrant rural dwellers (66.4%), non-migrant urban dwellers (19.4%), rural–urban rural–urban migrants (8.4%), urban–rural migrants (2.0%) and multiple migrants, who had moved two or more times from rural to urban or vice versa, (3.8%).

This study reports that 10.8% were distinct migrants (either rural–urban or urban–rural migrants). It was observed that rural–urban migrants had higher prevalence rates of diabetes, hypertension and obesity than rural non-migrants (Fig. 5). Longer time since migration was observed to increase the prevalence and risk of diabetes and other metabolic parameters. The risk for diabetes was 1.9 times higher in rural–urban migrants compared to non-migrant rural dwellers. The partial population attributable risk (PARp) for diabetes among non-migrant urban dwellers and rural–urban migrants was assessed and five risk factors (hypertension, abdominal obesity, generalized obesity, physical inactivity and low fruit and vegetable intake) explained nearly 70% of diabetes among rural–urban migrants compared to 66.4% among non-migrant urban dwellers. The low levels of physical activity among rural–urban migrants may be attributed to an unfamiliar urban environment and loss of social networks and support. These findings have significant implications for health systems. Given the rapid urbanization taking place in India, prevention programs emphasizing healthy lifestyles for those living in urban settings, regardless of their place of origin, are urgently needed. Long-term non-migrant urban dwellers and rural–urban migrants have higher prevalence/risk of diabetes and other metabolic disorders than urban rural migrants and non-migrant rural dwellers.

Figure 5:
figure 5

Prevalence of various disorders in migrant and non-migrant urban and rural dwellers.

Full size image

4.7 Role of Diet in Diabetes

Using pooled data from 30 states in the ICMR-INDIAB study, macronutrient recommendations for remission and prevention of type 2 diabetes in Asian Indians were developed using a data-driven optimization method.26 The macronutrient recommendation was tested among three categories: newly diagnosed diabetes (NDD) (n = 1594), prediabetes (n = 7336), and normal glucose tolerance (NGT) (n = 9160). To determine the optimal macronutrient compositions that would lower the sum of the difference between the estimated HbA1c from the linear regression model and the targets for remission (6.4% for NDD and 5.6% for prediabetes). Table 2 presents the macronutrient recommendations to achieve remission in those with NDD and prediabetes. In addition, macronutrient recommendation for prevention of progression to type 2 diabetes in prediabetes and NGT groups, were also developed as a constrained quadratic programming problem (QPP).

Table 2: Macronutrient recommended to achieve remission in those with newly diagnosed diabetes (NDD) and prediabetes.26
Full size table

Four macronutrient recommendations (%E- Energy) developed as follows:

  1. (1)

    Diabetes remission in NDD: carbohydrate, 49–54%; protein, 19–20%; and fat, 21–26%

  2. (2)

    Prediabetes remission to NGT: carbohydrate, 50–56%; protein, 18–20%; and fat, 21–27%

  3. (3)

    Prevention of progression to T2D in PD: carbohydrate, 54–57%; protein, 16–20%, and fat 20–24%

  4. (4)

    Prevention of progression to T2D in NGT: carbohydrate, 56–60%; protein, 14–17%; and fat 20–24%

For both remission and prevention of T2D in all glycemic ranges, the study findings advise decreasing carbohydrate calories while increasing protein calories. Furthermore, a minimum protein consumption of 14–16%E was identified to prevent progression in the prediabetes and normal glucose tolerance (NGT) categories. These findings highlight the need for updated dietary recommendations that suggest suitable adjustments to the macronutrient composition in order to lessen the prevalence of diabetes among Asian Indians. These findings reflect the natural diversity in dietary practices and culture as it was conducted on data from national sample and hence can be used by the whole Indian population.

4.8 Role of Physical Inactivity in Diabetes

The synergistic impact of behavioral risk factors, such as physical inactivity, poor diets, and use of tobacco and alcohol, cause the majority of NCDs. Two of the most significant modifiable risk factors for NCDs, viz. increased levels of physical inactivity and poor diets, can be blamed, at least in part, for the rising prevalence of these diseases.27 Using the Global Physical Activity Questionnaire (GPAQ), the ICMR-INDIAB study evaluated the data on the levels of physical activity and inactivity in India.28 Overall, 54.4% of the population were inactive (males: 41.7%), while 31.9% (males: 58.3%) were active and 13.7% (males: 61.3%) were highly active.14 The physical inactivity rates in the urban and rural areas were examined in Phase I of the ICMR-INDIAB project. Female participants were much more inactive than their male counterparts, and individuals in urban areas were more inactive than those in the rural areas (65.0 vs. 50.0%). Chandigarh (66.8%), Tamil Nadu (60.0%), Maharashtra (55.2%), and Jharkhand (34.9%) had the highest rates of physical inactivity. In Chandigarh, Jharkhand, Maharashtra, and Tamil Nadu, respectively, 88.4, 94.8, 91.3, and 93.1% of the subjects reported not engaging in any leisure time physical activity. Most of the time spent in moderate to vigorous intensity activity was at the workplace. The average daily time spent engaging in moderate to high intensity activity was under 20 min, even among those who reported engaging in recreational physical activity (8.1%).

5 Summary

In summary, the results of ICMR-INDIAB shows that the overall prevalence of diabetes from 15 states/UT was 7.3%, ranging from 4.3% in Bihar to 10.0% in Punjab. The prevalence of diabetes is high, in both urban and rural areas and the explosion in diabetes is inevitable, with increasing prevalence of prediabetes. The prevalence of diabetes varies greatly amongst the states of India. With a higher prevalence of diabetes among low socioeconomic groups in the urban areas of the more economically developed states, our findings indicate evidence of epidemiological transition. Only 43% of the participants in the survey had ever heard of diabetes and the residents in urban areas were more aware of diabetes than those in rural areas. The achievement of treatment goals and adoption of healthy behaviors remains suboptimal in India with only a third of individuals with known diabetes having good control of diabetes, fewer than half having good control of blood pressure and LDL cholesterol, and only 7.7% meeting all three targets (A1c, BP, and cholesterol targets). The study shows that a large percentage of people in India are inactive with fewer than 10% engaging in recreational physical activity. Migration from rural to urban areas is linked to an increased risk of diabetes and other cardiometabolic disorders. For both remission and prevention of diabetes, the study findings advise decreasing carbohydrate calories while increasing protein calories. These findings highlight the urgent need for government level prevention and management plans and programs to curtail the epidemic of diabetes and related NCDs in India.