Thousands of Undiagnosed T2D Cases Found in UK Biobank Cohort

Liam Davenport

September 25, 2020

One percent of individuals in the UK Biobank cohort apparently without diabetes have undiagnosed type 2 diabetes, suggests an analysis that found it can take more than 5 years for a diagnosis to be made.

The research was presented at the European Association for the Study of Diabetes (EASD) Virtual Meeting 2020 on September 24, held online due to the COVID-19 pandemic.

Dr Katherine Young, from the College of Medicine and Health at the University of Exeter, and colleagues, looked at data on more than 200,000 individuals from the UK Biobank who had had glycated haemoglobin (HbA1c) measured at baseline.

They found 1% with levels indicative of type 2 diabetes, yet comparison with primary care records showed it took a median of more than 2 years for them to be diagnosed, with nearly a quarter still undiagnosed after 5 years.

Dr Young said in a press release that their study "shows that population-level screening could identify cases of type 2 diabetes far earlier and potentially reduce complications".

She added that, while a preliminary analysis suggested that the diagnostic delays "did not significantly impact diabetes-related complications in this group of people, further research is required to ascertain whether screening for diabetes in this age group would reduce diabetes-related complications".

Undetected Type 2 Diabetes

Dr Faye Riley, senior research communications officer at Diabetes UK, which was not involved in the study, said: "Type 2 diabetes can go undetected for years and this research flags that all too often people can have the condition but not know it.

"The first signs of type 2 diabetes can be hard to pick up but getting diagnosed early gives people the best chance of avoiding devastating complications and living a long and healthy life with type 2 diabetes."

She continued: "12.3 million people in the UK are at an increased risk of developing the condition, but often people are unaware of their risk."

Presenting the study, Dr Young said that "approximately 4 million people in the UK have diabetes", of whom more than 90% have type 2 disease.

She added that it is thought that an additional 500,000 people may be living with undiagnosed diabetes, "where the longer the delay before these people receive a diagnosis, the higher the risk of complications from higher blood glucose levels".

Dr Young noted that "there have been discussions in the UK and elsewhere recently regarding whether routine screening for diabetes should be carried out, particularly in older adults".

However, she pointed out: "We actually have very little data on the impact of these proposed screening programmes.

"We don’t know how many people would be identified, how much earlier they would be diagnosed, and what the benefits would be in terms of any possible reductions in terms of complications."

Biobank Search

To investigate further, the team selected individuals aged 40–70 years from the UK Biobank database who had an HbA1c measurement at baseline; primary care data available, and no evidence of a diabetes diagnosis prior to recruitment.

Undiagnosed diabetes was defined as an HbA1c level ≥48 mmol/mol at the baseline assessment, with measurements recalibrated after finding that they were systematically lower than those in primary care records for the same cohort in cases where readings were taken within 100 days of each other.

Diabetes diagnosis dates were defined as the earliest date in the primary care records for a diagnosis code for diabetes; a prescription for glucose-lowering medication, or an HbA1c measurement ≥48 mmol/mol, whichever came first.

From 466,488 people in the UK Biobank cohort with an HbA1c reading measured at baseline, 216,644 had primary care data available, and 201,465 had no evidence of a diabetes diagnosis prior to recruitment.

Of these, 2022 (1%) had a baseline HbA1c level ≥48 mmol/l, indicating undiagnosed diabetes.

Primary care data covering the period immediately after the baseline assessment was available for 1648 patients.

The team determined that individuals with undiagnosed diabetes were older than those without, at a median age of 61 versus 58 years (p<0.001), and were more likely to be male, at 60% versus 45% (p<0.001).

Individuals with undiagnosed diabetes also had a significantly higher body mass index than those without, at 31.0 kg/m2 versus 26.6 kg/m2 (p<0.001).

Analysis revealed that the median time to a clinical diagnosis of diabetes among those with undiagnosed disease at baseline was 2.3 years, at an interquartile range of 0.8–4.7 years.

It was calculated that 23% of patients with undiagnosed diabetes at baseline still had not received a diagnosis for the condition 5 years later.

Dr Young said that further work in progress "will include examining the impact of delays in diagnosis on clinical care, including diabetes related complications".

The research is funded by Professor Andrew Hattersley's Novo Nordisk Foundation Diabetes Prize for Excellence.

No relevant financial relationships declared.

EASD Virtual Meeting 2020: Abstract 331. Presented September 24.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.