Last week GPs were contacted by women asking why they had been told to shield when they were not at high risk from COVID-19 - prompting NHS England diabetes advisor Dr Partha Kar to call on officials to check the relevant data.
The Royal College of Obstetricians and Gynaecologists also launched an investigation, saying that there was ‘no previous suggestion’ that a history of gestational diabetes alone could be linked to severe COVID-19 illness.
NHS Digital has now confirmed that the algorithm used in the QCovid tool categorised some women as having diabetes when they in fact had gestational diabetes. It said that ‘incomplete data’ may also have led to the risk tool defaulting some women to a higher level of risk on a ‘precautionary basis’.
Last week an additional 1.7m patients were identified as being at increased risk from COVID-19 and added to the shielding list based on factors including age, ethnicity, BMI and medical conditions or treatments, in a process based on QCovid.
GPs were asked to prioritise 820,000 people for COVID vaccination. But GPs questioned why women with a history of gestational diabetes had been identified as at risk.
NHS Digital confirmed in an update this had been a mistake it said: ‘Some women with previous gestational diabetes have been identified by the QCovid model as being at high risk. This will be appropriate for many as the model performs an individual assessment based on a wide range of risk factors, and also considers an individual’s risk in comparison to others of the same age and sex.
‘However, because the risk assessment is based on routinely coded data from multiple systems some women may have been identified as having diabetes when in fact they had gestational diabetes. Others may have incomplete data in which case the risk tool may have defaulted, on a precautionary basis, to a higher level of risk for that category and this may influence the overall assessment results.'
NHS Digital added that QCovid was a ‘living model’, and would continue to be ‘refined and updated’ when new data was available.
Dr Kar confirmed to GPonline that a woman would only now be advised to shield if she had a previous diagnosis of gestational diabetes, subsequent tests showed that she had developed type 2 diabetes since, and their weight was high.
York GP Dr Abbie Brooks was one of a handful of GPs to highlight the issue last week, asking whether GPs would be asked to review the hundreds of women who had been wrongly asked to shield.
Anyone else noted the code for previous gestational diabetes leads to shielding in new NHS digital searches? @parthaskar @Richard_GP— Abbie (@AbbieSBrooks) February 19, 2021
Is this what they meant when they said “all diabetes for cohort 6”?
Bristol GP Dr Kyle Roys said: ‘We’ve been inundated with requests about this and BMI in the same context. Lots being directed from other healthcare practitioners. We’re trying to get a plan on how to deal with them all but it’s been a sharp crime for our staff organising the vaccinations.’
Dr Kar had asked NHS Digital to check the tool was correctly identifying patients, saying that the data provided to QCovid by the National Diabetes Programme did not include information on gestational diabetes.