Scottish study casts doubt on claims of Covid diabetes link

It comes after a controversial study published in January by the US Centres of Disease Control reported that under-18s who caught Covid were up to 2.5 times more likely to be diagnosed with diabetes within 30 days of a positive test.

Critics stressed that the data was observational, did not account for potential confounders, and could not prove causation.

READ MORE: Scotland Covid expert backs 'targeted measures' to keep vulnerable out of hospital 

Professor Helen Colhoun, an expert in epidemiology who led the Scottish study, said the findings had caused "huge consternation" for charities and parents, particularly at a time when the Omicron wave was leading record numbers of schoolchildren to become infected.

She said: "To be honest, it was very obvious that there was something seriously wrong with their methodology and analysis.

"Even in people who'd never been exposed to Covid, their incidence rate of type 1 diabetes was at least 10-fold higher than had ever been reported in any population in the US before - so that was implausible."

Professor Helen Colhoun

In Scotland, scientists identified 1,074 cases of type 1 diabetes diagnosed between March 1 2020 and November 22 2021 among individuals under 35 who had also had at least one confirmed Covid infection.

Notably, they found that individuals with no known history of Covid were actually more likely to have been diagnosed with diabetes over the study period than someone who had had tested positive, but who was more than 30 days on from their Covid infection at the point when the diabetes was diagnosed.

But something else appeared, at face value, to mirror the US results: diabetes diagnoses were around 2.6 times higher among participants within 30 days of a Covid infection compared to those with no history of Covid.

READ MORE: New fast-spreading 'Centaurus' Covid variant detected in Scotland

However, further analysis indicates that this "strong association" is down to correlation, rather than causation.

Prof Colhoun said: "Apart from anything else, there was a policy of testing everyone who came into hospital so by definition you were massively more likely to get tested for Covid simply because you were being admitted to hospital with symptoms of type 1 diabetes.

"So clustering of Covid close to type 1 diabetes within a 30-day interval is really not evidence of causation.

"We already know that if you dig deep into the diagnosis data for type 1 diabetes most people have been suffering symptoms for several weeks before they actually get diagnosed, so even those 30-day cases would have been much more likely to have begun before Covid than after Covid."

Previous studies have shown that the median time from symptom onset to diagnosis for type 1 diabetes in under-16s is 25 days.

Anyone showing signs of the disease for the first time is also prioritised urgent hospital admission.

"The increased frequency of recent negative, as well as positive, [Covid] tests around the time of presentation with type 1 diabetes indicates that the association is partly attributable to higher detection of infection through increased testing around the time of presentation with type 1 diabetes", 'Relation of Incident Type 1 Diabetes to Recent COVID-19 Infection', Diabetes Care

Analysis of the Scottish data shows Covid testing concentrated in the two days before or after diabetes diagnosis, but the vast majority of test results were negative.

The researchers also note that since people with type 1 diabetes are more likely to become sick from Covid, the infection probably led undiagnosed diabetes to be discovered earlier.

Separate analysis of the longer term diabetes trends by age groups over time, from January 2015 to January 2022, also found "no evidence of an increase" during 2020 and 2021 among those aged 15 to 34 beyond what was already being tracked pre-pandemic.

In the 0-14 age group, incidence of type 1 diabetes in 2020 and 2021 was 20% higher than the seven-year average, but this increase largely predated exposure to Covid in this age group.

"The time course of the increase in incidence of diabetes in those aged 0–14 years predated most of the cumulative incidence of infection in this age group", Diabetes Care

Recent estimates for England suggest that only 8% of children aged 14 and under had had Covid by the end of

August 2020, rising to 25% by the end of June 2021.

Prof Colhoun, an honourary consultant in public health at NHS Fife and chair of Medical Informatics and Life Course Epidemiology at Edinburgh University who co-chaired Public Health Scotland's Covid-19 Modelling & Research group, said the epidemic was likely to have followed a very similar course in Scotland.

She said: "What we found was that in the younger people - 15 years and under - there was a 20% increase in type 1 diabetes between 2020 and late 2021 but most of that was in the first year of the pandemic, whereas most children who got Covid got it later than that.

"What's it's most likely to be explained by is disruption to these other viruses that are linked to type 1 diabetes, by disrupting social mixing patterns.

"The general message is that, when we do these social interventions of drastic alterations in behaviour, there can be unintended consequences in the epidemiology of other viruses."

READ MORE: Why lockdown may have exposed a previously unrecognised viral cause of hepatitis in children 

Diagnoses of type 1 diabetes peak annually in February and September, in line with theories that seasonal viruses can trigger the disease by damaging the cells which produce insulin.

Studies of gut microbes suggest a "ying and yang interplay" where some viruses appear to have a protective effect against type 1 diabetes while others increase the risk.

Prof Colhoun said this normal balance was likely disrupted during lockdowns when many common childhood viruses all but vanished.

She said:"We don't know that for certain - but the general issue of what's happened more broadly to viral epidemiology over the course of the pandemic and changes in human behaviour is very much an area for investigation."

However, she added that parents should feel "reassured" by the findings, which are published in the journal Diabetes Care.

"We found that there was no evidence that Covid-19 infection itself led to diabetes," said Prof Colhoun.

"Instead it is clear that there was simply more testing for, and incidental detection of, Covid-19 around the time of onset of type 1 diabetes."