View commentaries on this research

This is a plain English summary of an original research article

Type 2 diabetes is not a single disease. This was shown in 2018 when doctors in Scandinavia identified several sub-groups of type 2 diabetes in their population. Each sub-group has distinct characteristics and may respond to different treatments.

Now, researchers have discovered different sub-groups among people in India. Their study included more than 19,000 people with type 2 diabetes. The researchers identified four sub-groups, two of which were linked to an especially high risk of kidney and eye disease.

People of Indian heritage are at a higher risk of developing type 2 diabetes, and at a younger age, than people from other backgrounds. This research could allow people with the highest risk of complications to be offered more intense treatment. It will prompt research to discover which treatments work best for people in each group. The findings are relevant to people of Indian heritage in the UK.

What’s the issue?

Worldwide, the number of people with diabetes has been rising for decades and now stands at 463 million. Nine in ten have type 2 diabetes and are unable to process glucose (sugar) in the blood.

Insulin is the hormone which controls the amount of glucose in the blood. In type 2 diabetes, glucose levels are high either because the body can no longer use insulin (insulin resistance) or is unable to produce enough (insulin deficiency).

People of Indian heritage are prone to getting diabetes at younger ages and lower body weights. This could be because of differences in how the disease develops in this population, compared to others. The risks of complications, and responses to treatments, might also differ.

This research was intended to look for sub-groups of diabetes patients in the Indian population and compare them to the sub-groups found recently in Scandinavian populations.

What’s new?

Researchers gathered data about 19,084 patients from diabetes centres in nine states across India. They found they could group patients into four clusters, or sub-groups, by considering eight factors: age at diagnosis, body mass index (BMI), waist measurement, HbA1c (a measure of average glucose concentration in the blood over time), fasting blood glucose, triglycerides (fats), HDL (‘good’) cholesterol levels in the blood, and C-peptide levels (a measure of how much insulin is produced).

In addition, researchers looked to see if people had signs of damage to the retina of the eye (retinopathy) or signs of kidney disease caused by diabetes.

In each of the four sub-groups, people with similar factors had similar outcomes. The subgroups were:

    1. Severe insulin deficient diabetes. People in this group had a lower BMI and waist measurement, but they produced least insulin and had highest average glucose concentration. One in four (26.2%) people were in this cluster, which had one of the highest risks of complications.
    2. Insulin resistant obese diabetes. People in this group had the highest BMI and waist measurement, and they produced most insulin. One in four (25.9%) people were in this cluster.
    3. Combined insulin resistant and deficient diabetes. People in this group were youngest when diagnosed, had BMI and waist measurements that were between groups 1 and 2, and produced insulin at levels between groups 1 and 2. One in eight (12.1%) people were in this cluster, which had the highest risk of complications from kidney disease and second highest risk for eye disease.
    4. Mild age-related diabetes. People in this group were older when diagnosed, had higher HDL cholesterol and produced reasonable levels of insulin. One in three (35.8%) people were in this cluster.

The second and third groups had not previously been identified.

Why is this important?

The results show that some people in India have diabetes that is different from the same condition in Scandinavian people, and potentially in other populations. Two new groups have not been seen in populations outside India.

One of the new groups, the combined insulin resistant and deficient diabetes group, is at particularly high risk of complications. Usual treatments may be less effective, and these patients may need a combination of agents targeting different aspects of diabetes. They may also need more frequent screening for eye and kidney complications.

The results are likely to be relevant in the UK, where there is a large Asian Indian community.

What’s next?

The next stage is to look at how different drugs vary in the way they work between people in different sub-groups, the researchers say. They want to work out which drug is best suited to which sub-group of patients, both as a starting drug and in response to changing factors as the disease develops. They say randomised clinical trials will be needed to see how different treatments work in different groups.

Current diabetes guidelines, published by NICE in 2015, do not use the sub-groups and may need to be revised in future, if it is shown that these sub-groups require need different treatments.

The researchers also want to develop an online tool or mobile app to allow doctors to quickly classify a patient into one of the four sub-groups.

You may be interested to read

The full paper: Anjana RM, and others. Novel subgroups of type 2 diabetes and their association with microvascular outcomes in an Asian Indian population: a data-driven cluster analysis: the INSPIRED study. BMJ Open Diab Res Care. 2020;8:e001506

More information on the INSPIRED (INdia-Scotland Partnership for pRecision mEdicine in Diabetes) project

Information from the American Medical Association about different types of diabetes: Precision Medicine in Diabetes Initiative

NICE guidance: Type 2 diabetes in adults: management [NG28] (2015, last updated 2020)


Funding: This research was an INSPIRED (INdia-Scotland Partnership for pRecision mEdicine in Diabetes) project, part of the NIHR Global Health Research Unit on Diabetes Outcomes.

Conflicts of Interest: The study authors declare no conflicts of interest.

Disclaimer: NIHR Alerts are not a substitute for professional medical advice. They provide information about research which is funded or supported by the NIHR. Please note that views expressed in NIHR Alerts are those of the author(s) and reviewer(s) and not necessarily those of the NHS, the NIHR or the Department of Health and Social Care.

  • Share via:
  • Print article


Study authors

The study is important because, with the establishment of such clusters, we will think twice before starting every person with type 2 diabetes on the same standard treatment. This is the approach in current guidelines. But we now know that diabetes in Asian Indians differs from the condition studied in White populations. This is relevant in the clinic not only in India but in countries like UK and US where there is a large Asian Indian community. The clusters tell us which sub-groups have a higher risk of complications and need more aggressive treatment and more frequent screening.We were partly surprised by the results, because they were so different from results in the Scandinavian population. At the same time, we already knew through earlier studies that insulin deficiency is a prominent feature of type 2 diabetes in India. So when we found two clusters with insulin deficiency, we were not completely surprised either.Ranjit Mohan Anjana, Vice President &  Viswanathan Mohan, President, Madras Diabetes Research Foundation, Chennai, Tamil Nadu, India

Lived experience

This research should influence practice in the UK. It should allow British Indians with type 2 diabetes to receive a more person-focused approach to their treatment. Once it is established which of the four sub-groups a patient most accurately sits in, they can be offered the most appropriate treatment at the earliest possible stage.If it can result in better screening of people with type 2 diabetes from a younger age, ideally at the pre-diabetes stage, this can only be positive. This could reduce the risks of complications from full-blown diabetes while at an early stage. And it will allow those with the greatest risks of morbidity and mortality to receive greater attention.Manoj Mistry, Public Contributor, Greater Manchester


This research improves our understanding of the patient population. It shows that the time taken to reach targets for blood glucose levels differs between the sub-groups and it will allow more personalised interventions to control blood glucose levels. Currently, some of the tests used to identify the sub-groups, such as C peptide levels, are not routinely carried out in type 2 diabetes, so this may need to change.Abd Tahrani, Senior Lecturer in Metabolic Endocrinology and Obesity Medicine, University of Birmingham, and Honorary Consultant in Endocrinology, Diabetes and Weight Management, University Hospitals Birmingham NHS Foundation Trust
Back to top