View commentaries on this research

This is a plain English summary of an original research article

A screening pathway using blood and urine tests followed by two genetic (DNA) tests identified all people with a rare subtype of diabetes called monogenic diabetes. The screening pathway performed better than current practice based on age at diagnosis and family history which misses 63%. It is, therefore, a useful approach for ruling out this form of diabetes and probably cheaper overall than offering every young person with diabetes DNA testing.

Monogenic diabetes, caused by a mutation in a single gene can be misdiagnosed as type 1 diabetes, leading to unnecessary insulin treatment. Those recruited for this study were 98% white and tested at an average of 14 years after diagnosis.

Genetic tests definitively diagnose monogenic diabetes, but are costly, and not an appropriate test for everyone with the early onset diagnosis. The researchers wanted to offer simpler tests to patients diagnosed when young to see if they could catch a higher proportion of people likely to benefit from the genetic tests.

This NIHR-funded study shows that the sequence of tests can help identify more cases of monogenic diabetes than clinical features alone. It would be cheaper than DNA testing of all young adults at diagnosis. More data is needed on costs and performance when applied to wider populations, including other ethnic groups and closer to diagnosis.

Why was this study needed?

Monogenic diabetes is a rare type of diabetes caused by a mutation in single genes. It can be inherited or occur due to a spontaneous mutation. An estimated 20,000-40,000 people have monogenic diabetes in the UK, around 2% of all people with diabetes.

Monogenic diabetes tends to appear before age 25 and is often misdiagnosed as type 1 diabetes. Misdiagnosis can lead to inappropriate or unnecessary treatment. This is because type 1 diabetes is an autoimmune disease where antibodies destroy the cells that produce insulin in the pancreas. Injections are needed to replace insulin. However, in monogenic diabetes, treatment is possible with diabetic tablets or not at all, depending on the type of mutation.

The DNA tests to confirm monogenic diabetes are expensive, so this UK-based study looked at whether a three-step screening process could better identify who should be referred for DNA testing.

What did this study do?

The UNITED study included 1,418 people in two UK regions diagnosed with diabetes before the age of 30 and currently aged under 50. There were 42 known cases of monogenic diabetes in the sample (previously diagnosed by DNA test). The researchers tested everyone eligible using a three-step diagnostic pathway which included DNA tests for everyone.

Firstly, the participants taking insulin performed a urine test called UCPCR. If levels were 0.2nmol/mmol or more, then this suggests insulin is being produced, and type 1 diabetes is less likely. Participants progressed to the second step. Here, people including those not injecting insulin were given blood tests looking for autoantibodies GAD and IA2, a sensitive test for type 1 diabetes. If these were negative, participants were referred for genetic tests (see definitions tab).

The participants were mainly white (98%) and had been diagnosed for 14 years on average.

What did it find?

  • Prevalence of monogenic diabetes in this sample was 3.6% (slightly higher than the 2% quoted for the UK).
  • The new three-step screening pathway identified 17 new cases of monogenic diabetes in addition to the “already known” cases.
  • The usual pathway, where people aged less than 25 at diabetes diagnosis, not on insulin and with a parent who has diabetes are referred for DNA tests, would miss an estimated 63% of cases (positive predictive value 57.6%).
  • Following the UNITED screening pathway, five people would need to have DNA tests to pick up one case (positive predictive value 20%), but negative predictive value was high at 99.9%. This means that only 0.1% would be missed by the strategy.
  • The 17 new cases of monogenic diabetes identified in the study were less likely to have the three most common mutations than the known cases (47% versus 74%). The new cases were also less likely to have a parent with known diabetes than the cases already identified (47% versus 85%).

What does current guidance say on this issue?

NICE 2016 guidelines recommend assuming that children and young people with diabetes have type 1 diabetes unless there are strong indications otherwise. It currently recommends against routinely measuring UCPCR or autoantibodies.

Genetic or DNA testing is currently recommended only if the person’s diabetes does not respond to treatment, in the presence of suggestive clinical characteristics and where there is a family history of diabetes. The following signs may suggest monogenic diabetes: developing diabetes in the first year of life, absence of or occasional ketone bodies in the blood during an episode of high blood sugar (hyperglycaemia), systemic illnesses or syndromes that include deafness, optic atrophy (damage to the optic nerve) or retinitis pigmentosa (a rare genetic disorder causing breakdown of the cells in the retina).

What are the implications?

These findings suggest that using a sequence of tests in young people with diabetes can better identify who should be referred for genetic testing and ultimately lead to diagnosis of more people with monogenic diabetes, who might be able to reduce or stop insulin. More people are referred for DNA testing than with the clinical criteria currently used or with a family history.

This study did not consider overall costs of the new pathway compared to current practice across the population. The costs of the screening tests in the UK are £10.80 for UCPCR and £20 for autoantibodies. Genetic tests are much more expensive, for example, £650 in Exeter. So it is not yet clear where the balance of cost versus benefits lies.

This study does highlight the need to maintain awareness of this rare type of diabetes among healthcare professionals to ensure that people receive appropriate treatment. More data on relative costs and how the testing pathway performs closer to diagnosis is probably required before the strategy is adopted widely.

Citation and Funding

Shields BM, Shepherd M, Hudson M, et al. Population-Based Assessment of a Biomarker-Based Screening Pathway to Aid Diagnosis of Monogenic Diabetes in Young-Onset Patients. Diabetes Care. 2017;40(8):1017-25.

This study was supported by the Department of Health and Wellcome Trust Health Innovation Challenge Award (HICF-1009-041 and WT-091985).



Diabetes UK. Maturity-onset diabetes of the young (MODY). London: Diabetes UK.

NICE. Diabetes (type 1 and type 2) in children and young people: diagnosis and management. NG18. London: National Institute for Health and Care Excellence; 2016.

WHO. Definition and diagnosis of diabetes mellitus and intermediate hyperglycaemia. Geneva: World Health Organization; 2006.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre


  • Share via:
  • Print article


The testing process consisted of three rounds of tests, with people who tested positive progressing to the next test.
  • The first test – urinary C-peptide/creatinine ratio – was a urine test that people could do at home and send off to the lab. This test detected whether or not people produced insulin. This would exclude people who had type 1 diabetes for long enough for it to have destroyed all of the islet cells that produce insulin.
  • The second test looked for autoantibodies in the blood that indicate someone has type 1 diabetes rather than monogenic diabetes - Glutamic Acid Decarboxylase and Islet Antigen 2.
  • The third genetic tests first looked for the three genes most commonly associated with monogenic diabetes – the HNF1A, HNF4A and GCK genes. If these were negative, further genetic analysis was performed to look for 35 genes known to cause monogenic diabetes.


Expert commentary

The diagnosis of diabetes mellitus in people under 30 years often leads to the initiation of lifelong insulin therapy and the challenges associated with this treatment.This study uses a simple three-step algorithm. Measurements of accessible biomarkers are the first two steps. This allows the identification of people that require insulin therapy and those where further genetic testing may be required to identify monogenic diabetes, which may not require insulin therapy.This approach tailors optimum individual glycaemic therapy to people with an onset of diabetes that are 30 years and below.Jeffrey W Stephens, Clinical Professor of Diabetes, School of Medicine, Swansea University; Consultant Physician, Morriston Hospital, ABMU Health Board, AMD Medical Education
Back to top