Of Indian heritage? Your GP’s own guidelines already rate you high risk for diabetes

You do not have to be overweight. You do not need a parent with the condition. Under the guidelines Australian GPs work from, South Asian background is enough on its own to put an adult in the high-risk group for type 2 diabetes, and to earn a blood test every three years. That is South Asian diabetes risk as the guidelines define it.

Australian figures point the same way. An analysis of the 2021 Census, published in July 2025, found that people aged 30 and over who were born in Southern and Central Asia reported diabetes at more than twice the rate of the Australian-born: 12.5 per cent against 5.7, once the figures were adjusted for age. In their seventies, it was three in ten.

Where the South Asian diabetes risk rule comes from

The Royal Australian College of General Practitioners publishes its screening advice in a volume known as the Red Book, now in its tenth edition. The diabetes section is dated 28 June 2024. For the general population the advice is light-touch: a risk questionnaire every three years for those over 40.

Then comes the list of people considered high risk, and one item on it is enough. Being overweight counts from 40, or from 18 if high blood pressure or signs of insulin resistance come with it. A parent, brother, sister or child with diabetes counts. And so does background. The Red Book names Aboriginal and Torres Strait Islander, South Asian, South-east Asian, North African, Latin American, Middle Eastern, Māori and Pacific Islander people, and adds that people of mixed ethnicity are included. Notice what that last item lacks. The weight rule has an age attached to it. The background rule has none. A 28-year-old of Indian heritage with no weight problem and no family history is on the list, and if they have no symptoms the Red Book recommends a fasting glucose or HbA1c test every three years.

The college’s separate handbook on managing type 2 diabetes, in its 2024 edition, spells it out. The same groups are considered high risk “regardless of AUSDRISK score”, it says, and doctors are told to keep their suspicion high because the disease often has no symptoms and “is increasingly developing in younger people”.

The scales and the tape measure

The Red Book draws the weight lines differently too. Overweight normally begins at a body mass index of 25. For South Asian and South-east Asian people the guideline sets it above 23, and obesity above 27.5. A BMI of 24, below the general cut-off, is overweight here. Those cut-offs sit in the guideline’s very-high-risk tier, which takes in overweight people whose BMI is climbing: from 40, or from 18 with another risk factor. Screening there is every three years, or sooner if BMI keeps rising.

AUSDRISK, the questionnaire itself, was developed on behalf of Australian governments and estimates the chance of developing type 2 diabetes within five years. It has ten questions, and background shows up in two of them. Being born in Asia, the Indian subcontinent included, scores two points. The waist question runs on two scales: for men of Asian, Aboriginal or Torres Strait Islander descent, points start at 90 centimetres, against 102 for everyone else; for women it is 80 centimetres against 88. So for a man of Indian descent, born in Chennai or Melbourne, the tape measure is 12 centimetres stricter than for a colleague of European descent; for a woman, 8 centimetres. Twelve points or more is high risk, and the form’s advice at that point is direct: “See your doctor about having a blood test.”

What the Census can and cannot tell you

The study behind the headline figures was done by Davoud Pourmarzi at the Australian National University, using the whole 2021 Census rather than a sample. Only one group recorded a higher age-adjusted rate than the Southern and Central Asia-born: people born in Oceania outside Australia and New Zealand, at 12.6 per cent. Among people aged 70 to 79, 30.1 per cent of the Southern and Central Asia-born reported diabetes. For the Australian-born of the same age, it was 14.7 per cent. Unadjusted, the overall figures were 8.4 per cent against 7.1; the paper adjusts because migrant groups and the Australian-born differ in age structure.

Two caveats, both acknowledged in the paper. First, the region is a wide net. The Bureau of Statistics category puts India in with Bangladesh, Bhutan, the Maldives, Nepal, Pakistan and Sri Lanka, plus nine Central Asian countries, Afghanistan among them. The paper gives no figure for India on its own; the word does not appear in it. Second, the Census simply asked whether a doctor or nurse had ever told the person they had diabetes. The author accepts that self-reporting can skew the picture, and the Bureau says Census data on long-term health conditions are “not intended to provide prevalence estimates”.

What the blood test is looking for

The Red Book’s test is a fasting glucose or an HbA1c. An HbA1c between 6.0 and 6.4 per cent means higher risk, and the test is repeated in a year. At 6.5 per cent or above it is repeated to confirm a diagnosis. A fasting glucose between 5.5 and 6.9 mmol/L is an in-between result, and the guideline recommends an oral glucose tolerance test.

The handbook adds a safeguard. In someone with no symptoms, no single result settles it. A diagnosis needs an HbA1c of 6.5 per cent or more on two separate occasions, a fasting glucose of 7.0 mmol/L or more (or a random one of 11.1 or more) confirmed by a second abnormal fasting result on another day, or a tolerance test. “A second concordant laboratory result is required”, it says.

The guideline does not wait for a person to feel unwell. It asks where their family comes from, and if the answer is South Asia it asks for a blood test every three years, at ages when the general population is not asked at all.