Risk factors for diabetes

Risk factors for diabetes are worth separating by type, because the two main types have almost nothing in common here and conflating them produces the two commonest mistakes about the condition: that type 1 could have been prevented, and that type 2 is simply a consequence of eating badly.

Type 1 diabetes

Type 1 is an autoimmune condition, and it cannot currently be prevented. There is no diet, weight, activity level or parenting decision that causes or avoids it.

What is known is that a genetic susceptibility is involved, and that something in the environment appears to trigger the immune response in susceptible people. What that trigger is has been looked for intensively for decades and has not been found.

Family history raises the risk, and it is worth being precise about how much, because the figures circulating are often inflated. Having a parent or sibling with type 1 diabetes increases the risk considerably relative to the general population — but the great majority of people diagnosed with type 1 have no family history of it at all, which is why it so often arrives without warning. It is also more common in white European populations than in most others, the reverse of the pattern in type 2.

Parents of newly diagnosed children routinely search for what they did wrong. The answer is nothing, and that is worth stating rather than implying.

Type 2 diabetes: what cannot be changed

Type 2 is a different matter, and a substantial part of its risk is fixed.

Age. Risk rises with age, which is why routine risk assessment begins at forty for most people.

Family history. A parent, brother or sister with type 2 diabetes raises the risk substantially — more so than most people expect, and more than diet alone accounts for.

Ethnic background. This is the factor most often stated vaguely and it has concrete consequences. People of South Asian, Black African and African Caribbean background develop type 2 diabetes at younger ages and at lower body weights than white European populations. The reasons are not fully understood, but the practical effect is that the usual thresholds are the wrong ones, and UK guidance sets different numbers accordingly — set out below.

A previous pregnancy affected by gestational diabetes, which leaves a substantially raised risk of type 2 diabetes afterwards. That is the reason a glucose test is recommended after the birth and then annually, and it is one of the most frequently skipped follow-ups in the whole of diabetes care.

Polycystic ovary syndrome, which involves insulin resistance directly.

Some medicines and conditions. Long-term steroids raise glucose; some antipsychotic medicines carry a metabolic risk; and people with severe mental illness have markedly higher rates of type 2 diabetes, which is why physical health checks are built into that care.

Type 2 diabetes: what can

Weight is the largest modifiable factor, and where it is carried matters as much as how much there is. Fat around the middle, and in and around the liver and pancreas, is more closely associated with insulin resistance than fat elsewhere, which is why a waist measurement is sometimes more informative than a weight.

Alongside it: physical inactivity, independent of weight; what is eaten, with sugary drinks and heavily processed food the most consistently implicated and vegetables, pulses and wholegrains the most consistently protective; smoking, which raises the risk of type 2 diabetes as well as everything else; and high blood pressure, which travels with it.

Two corrections to the usual account. Eating sugar does not directly cause type 2 diabetes — the association runs through weight and through the overall pattern of eating, not through a direct toxic effect, and the "sugar causes diabetes" shorthand misleads people into thinking the rest of the diet is irrelevant. And plenty of people who develop type 2 diabetes are not overweight; a normal body weight is not a guarantee, particularly for anyone with a strong family history or from one of the higher-risk ethnic groups.

It is also worth saying what the risk factors are not. Type 2 diabetes is not a moral failing and it is not a punishment for indulgence. The genetics are strong, the environment people live in shapes a great deal of the rest, and treating the diagnosis as a verdict on someone's character is both wrong and a reliable way of stopping them attending appointments.

The numbers UK guidance actually uses

This is the part most writing on the subject leaves out, and it is the part that tells someone whether they should be doing anything.

Who should have a risk assessment:

• All adults over 40, apart from during pregnancy.
• Adults from the age of 25 who are of South Asian, Chinese, African Caribbean or Black African background, or from other higher-risk groups.
• Anyone at any age with other risk factors from the lists above.

Body weight thresholds. A healthy BMI is given as 18.5 to 24.9 — but 18.5 to 22.9 for people of South Asian or Chinese background. For people in the higher-risk ethnic groups, a BMI above 23 can justify a blood test even where a risk questionnaire does not score highly. For structured weight-loss support the threshold is a BMI of 30, or 27.5 for South Asian and Chinese people. Someone comparing themselves against the familiar figure of 30 may be well past the point at which they should have been assessed.

What the blood test shows. HbA1c, reported in the UK in mmol/mol:

• 42 to 47 mmol/mol (6.0 to 6.4%) — the high-risk range.
• 48 mmol/mol (6.5%) or above — diabetes.

Finding out, in practice

Three routes, none of which requires waiting for symptoms.

The NHS Health Check, offered in England to people aged 40 to 74 every five years, which includes a diabetes risk assessment along with blood pressure and cholesterol. Take-up is well below the number invited.

Diabetes UK's Know Your Risk tool, a short online questionnaire using age, sex, ethnicity, family history, waist measurement and BMI, which gives a risk score and says whether to see a GP. It is free and takes a few minutes.

And simply asking a GP for a test, which is a reasonable request for anyone who recognises themselves in the lists above.

The high-risk range is not a sentence

Being told that HbA1c is in the 42 to 47 range is the single most useful warning available, because at that point the trajectory can still be changed. It does not inevitably become diabetes.

England runs the NHS Diabetes Prevention Programme for people in that range — a structured programme of sessions on food, activity and weight over several months, with equivalents in the other UK nations. It exists because structured programmes of this kind substantially reduce the proportion of people who go on to develop type 2 diabetes, which informal advice to eat better does not.

Weight loss is the strongest single lever, and the amount needed is less dramatic than people assume — a modest percentage of body weight, sustained, makes a measurable difference to risk. Activity helps independently of whether weight falls.

Anyone already in that range is also due the rest of it: blood pressure and cholesterol checked and treated, and a repeat HbA1c at least annually so that progression is spotted early rather than years late.

See also diabetes, which covers what the condition is, the symptoms and how a diagnosis is made, and living with diabetes.