Completing the score so the result is comparable to the research
1. Say first whether you have already been diagnosed
This question comes first because it decides whether the rest of the page means anything. FINDRISC estimates the chance of developing type 2 diabetes over ten years. If you already have it, that question has been answered and a risk score is not the tool you need. The previous version never asked.
2. Enter height and weight rather than choosing a BMI band
The old build asked you to select which BMI band you were in. Most people do not know their BMI, so they guessed — and a guess moves the total by up to three points, which is enough to shift you a whole risk category. The band is now derived from measurements you actually have.
3. Measure your waist properly, in whichever units you use
Midway between the lowest rib and the top of the hip bone, at the end of a normal breath out, tape snug but not compressing. Centimetres or inches — the conversion is done for you rather than left as mental arithmetic before choosing from a dropdown, which is what the previous version required.
4. Answer the activity question honestly
It asks whether you do at least 30 minutes of physical activity on most days, counting activity at work and in ordinary daily life, not just deliberate exercise. A physically demanding job counts. Walking to and from a station counts.
5. Include the two items people most often skip
Whether you have ever been told you had high blood glucose — in pregnancy, during an illness, or at a routine check — is worth five points on its own, the largest single item in the score. Family history is worth up to five. Both are frequently forgotten and both matter more than most people expect.
6. Read the category, not the decimal
The output is a band with a published ten-year estimate attached, not a personal prediction. The useful reading is which of the five bands you fall in and what the score says about which of your own factors are driving it.
What FINDRISC is, and why we did not modify it
The score used here is FINDRISC, the Finnish Diabetes Risk Score, developed by Jaana Lindström and Jaakko Tuomilehto and published in Diabetes Care in March 2003. Its purpose was specific: to identify people at raised risk of type 2 diabetes without laboratory tests, using only questions anyone can answer and two measurements anyone can take.
It is one of the most widely used and best validated risk instruments in the world. It underpins national screening programmes across Europe, and adapted versions are used well beyond it.
Eight items, twenty-six points maximum:
- Age — 0 under 45, 2 for 45–54, 3 for 55–64, 4 for over 64
- BMI — 0 under 25, 1 for 25–30, 3 for over 30
- Waist circumference — 0, 3 or 4 points, on thresholds that differ for men and women
- Physical activity — 0 if at least 30 minutes most days, 2 if not
- Vegetables, fruit or berries — 0 if daily, 1 if not
- Blood pressure medication — 0 if never, 2 if ever taken regularly
- High blood glucose ever found — 0 no, 5 yes
- Family history of diabetes — 0 none, 3 for grandparent, aunt, uncle or first cousin, 5 for parent, sibling or child
The five bands and their published ten-year estimates of developing type 2 diabetes are: under 7, low, about 1 in 100; 7 to 11, slightly elevated, about 1 in 25; 12 to 14, moderate, about 1 in 6; 15 to 20, high, about 1 in 3; over 20, very high, about 1 in 2.
We changed none of it. A validated instrument derives its meaning from having been scored the same way in the studies that established what its output means. Adjusting the weights or the bands to seem more sophisticated would produce a number that could not be compared with anything, and that is not a calculator’s decision to make.
The waist thresholds are European, and this site is not
FINDRISC was derived and validated in a Finnish population, and its waist cut-offs reflect that: for men, under 94 cm scores 0, 94 to 102 scores 3, over 102 scores 4; for women, under 80 cm scores 0, 80 to 88 scores 3, over 88 scores 4.
Those thresholds are not universal. The International Diabetes Federation publishes lower ethnic-specific values, because abdominal obesity carries metabolic risk at a smaller waist in several populations — for South Asian, Chinese and South-East Asian men the threshold is 90 cm rather than 94, and for Japanese men 85 cm. Women’s thresholds are 80 cm across most of these groups, with Japan at 90.
For a site with a large Indian and South-East Asian readership this is not an academic footnote. A South Asian man with a 92 cm waist scores zero points on FINDRISC while sitting above the IDF threshold for his background — a real risk factor, invisible in the total.
The tool handles this the only honest way available. The score itself continues to use the validated European cut-offs, because changing them would break the instrument. Alongside it, the page now tells you when your waist falls under the FINDRISC threshold but over the IDF threshold for your background, so the information is in front of you instead of being silently discarded.
The claim we had to correct
The previous version of this page stated that structured prevention programmes “and, for some, medication like metformin” can cut progression to diabetes “by more than half”. That sentence is half right, which in a health tool is a defect rather than an approximation.
The Diabetes Prevention Program, published by Knowler and colleagues for the DPP Research Group in the New England Journal of Medicine on 7 February 2002, is the trial those numbers come from. It randomised 3,234 people with impaired glucose tolerance to intensive lifestyle intervention, metformin, or placebo. Over an average of 2.8 years, the lifestyle arm reduced the incidence of diabetes by 58 per cent compared with placebo. The metformin arm reduced it by 31 per cent.
“More than half” describes the lifestyle result. It does not describe metformin, and attaching it to both overstated the effect of a prescription medicine by a wide margin on a page read by people trying to decide what to do. Both figures are now given separately and attributed to the trial they come from.
The lifestyle intervention itself was specific rather than vague: a target of 7 per cent weight loss and at least 150 minutes of physical activity per week, with structured support. It also outperformed metformin in participants over 60, where the lifestyle effect was larger still.
Two further results are worth having alongside it. The Finnish Diabetes Prevention Study, published by Tuomilehto and colleagues in the New England Journal of Medicine on 3 May 2001 — by the same group that produced the score used on this page — independently found a 58 per cent reduction from lifestyle intervention, which is an unusually clean replication. And the Diabetes Prevention Program Outcomes Study, published in The Lancet on 14 November 2009, followed the original participants for ten years and found the benefit persisted rather than evaporating: diabetes incidence remained reduced by 34 per cent in the lifestyle group and 18 per cent in the metformin group. Smaller than the headline figures, and still there a decade later, which is the more useful thing to know.
What a questionnaire can and cannot tell you
FINDRISC is a screening instrument. It identifies people who would benefit from a blood test. It does not diagnose anything, and it cannot: type 2 diabetes is diagnosed on blood glucose or HbA1c, and nothing you can answer in a form substitutes for that.
A high score means your probability is raised relative to the population, not that you will develop the condition. A low score does not rule it out — the score contains no measurement of your actual glucose, and a substantial proportion of people with type 2 diabetes have no obvious risk factors at the point of diagnosis.
It was also derived in one population, in the 1990s, and validated principally in European cohorts. Its calibration in South Asian, African and East Asian populations is less well established, and in general the risk of type 2 diabetes rises at a lower BMI and a smaller waist in South Asian populations than the score’s thresholds assume. If you are of South Asian descent, a moderate score deserves more attention than the same score in a Finnish cohort would have warranted.
The score also says nothing about type 1 diabetes, which is an autoimmune condition with entirely different causes, or about gestational diabetes.
Which of your own points are worth acting on
The value of a scored questionnaire is that it separates what you can change from what you cannot.
Fixed: age and family history, together worth up to nine points. Nothing to be done, and they are still worth knowing, because they raise the value of watching everything else.
Modifiable: BMI, waist, physical activity and diet, worth up to ten points between them — and the two body-shape items carry the most weight. A previously recorded high glucose reading is worth five and cannot be undone, but it is the strongest single signal in the score that a blood test is overdue.
The evidence base for acting on the modifiable items is unusually strong. Modest weight loss — the DPP target was 7 per cent, not a transformation — combined with 150 minutes of activity a week produced the 58 per cent reduction described above. That is one of the better-supported findings in preventive medicine, and it applies most to the people whose scores are highest.
What this page is not
It is a validated questionnaire, scored correctly, with its source named. It is not a diagnosis, not a blood test and not medical advice. Nobody can tell from these eight answers whether you have diabetes now.
If your score falls in the moderate band or above, the useful next step is a fasting glucose or HbA1c test through your GP, not a decision made here. See a doctor sooner regardless of score if you have symptoms — unusual thirst, passing urine much more often especially at night, unexplained weight loss, persistent fatigue, blurred vision, slow-healing cuts or recurrent infections. Those warrant a test now rather than a questionnaire.
Your answers stay in your browser
Every answer, measurement and the resulting score are processed locally. Nothing is transmitted, stored or logged; there is no account and no upload, and the page keeps working with the connection off once loaded. For a health questionnaire that is the only arrangement worth having.
Questions people ask about diabetes risk scores
Is a high score a diagnosis?
No. It means a blood test is warranted. Type 2 diabetes is diagnosed on fasting glucose, an oral glucose tolerance test or HbA1c — never on a questionnaire.
Can I have diabetes with a low score?
Yes. The score contains no measurement of your glucose, and a meaningful proportion of people are diagnosed without obvious risk factors. A low score is reassuring, not conclusive, and symptoms override it entirely.
Why does waist matter as much as BMI here?
Because where fat is stored matters as much as how much there is. Fat around the abdomen sits alongside the organs and is far more strongly linked to insulin resistance than fat on the hips and thighs. Waist and BMI together are worth up to seven of the twenty-six points.
The waist thresholds seem high for me. Is that right?
Possibly. FINDRISC uses European cut-offs because that is where it was validated. The IDF publishes lower thresholds for South Asian, Chinese, South-East Asian and Japanese populations. The tool keeps the validated cut-offs for the score and tells you separately if you are over the IDF threshold for your background.
How much can lifestyle change actually reduce the risk?
In the Diabetes Prevention Program, intensive lifestyle intervention reduced progression to diabetes by 58 per cent over an average of 2.8 years; metformin reduced it by 31 per cent. The Finnish Diabetes Prevention Study found the same 58 per cent independently in 2001. At ten-year follow-up the reductions were 34 and 18 per cent — smaller, and still present. The lifestyle target was 7 per cent weight loss and 150 minutes of activity a week.
I was told I had high blood sugar once in pregnancy. Does that count?
Yes, and it is worth five points — the largest single item. Gestational diabetes is a well-established predictor of later type 2 diabetes and is exactly the kind of history the question is designed to capture.
Does the score cover type 1 diabetes?
No. Type 1 is an autoimmune condition with different causes and no relationship to these risk factors. FINDRISC is specific to type 2.
How often should I redo this?
Annually is reasonable if your score is elevated, or whenever something in it changes — a significant weight change, starting blood pressure medication, or a new family diagnosis.
Do you store my answers?
No. Nothing is saved or transmitted.
Sources behind these numbers
- Lindström J, Tuomilehto J, “The diabetes risk score: a practical tool to predict type 2 diabetes risk”, Diabetes Care, March 2003, 26(3):725–731 (PMID 12610029) — the FINDRISC instrument, its items, weights and risk bands.
- Knowler WC et al, Diabetes Prevention Program Research Group, “Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin”, New England Journal of Medicine, 7 February 2002, 346(6):393–403 (PMID 11832527) — the 58% and 31% figures, 3,234 participants.
- Tuomilehto J, Lindström J et al, Finnish Diabetes Prevention Study Group, “Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance”, New England Journal of Medicine, 3 May 2001, 344(18):1343–1350 (PMID 11333990).
- Diabetes Prevention Program Research Group, “10-year follow-up of diabetes incidence and weight loss in the Diabetes Prevention Program Outcomes Study”, The Lancet, 14 November 2009, 374(9702):1677–1686 (PMID 19878986) — 34% and 18% sustained at ten years.
- International Diabetes Federation — the ethnic-specific waist thresholds.
- NIDDK — Diabetes Prevention Program.
- NHS — type 2 diabetes, for symptoms and when to seek testing.
- CDC — diabetes.
- American Diabetes Association.
Tools people open next to this one
- Calories Burned Calculator — for planning the 150 minutes a week the DPP used.
- Sleep Calculator — short sleep duration is an established independent risk factor.