Scientists are questioning the reliability of algorithms used to trawl through patients’ health records and flag those who should be asked to shield and prioritised for vaccination, according to The Guardian.
GPs have reported being contacted by young, healthy patients confused as to why they have been told they are at high risk, or have been invited for a Covid-19 jab.
The Joint Committee on Vaccination and Immunisation (JCVI) has defined nine priority groups for vaccination, including everyone aged over 50, plus frontline health and social care workers and people aged 16-64 with serious health conditions.
The QCovid risk prediction algorithm, introduced last month, combines various characteristics, including age, sex, ethnicity and body mass index (BMI), to estimate the risk of catching, being admitted to hospital, or dying from Covid-19.
However, the algorithm appears to throw up certain inconsistencies. For example, if a patient’s weight or ethnicity are not recorded on their health records, QCovid automatically ascribes them a BMI of 31 (obese) and the highest risk ethnicity (black African), meaning they are more likely to be invited for a vaccine. One York-based GP, Dr Abbie Brooks, has identified 110 seemingly healthy individuals who were added to the shielding list and invited to book a vaccine last month.
About a hundred of them were women who had experienced gestational diabetes during pregnancy, but were now healthy. NHS Digital later clarified that such women may have been identified as diabetic by QCovid, and could be removed from the shielding list, if requested, and considered no longer at risk. They would still be called for an earlier vaccine.
However, Brooks said, “there were about 10 or 12 patients for whom I couldn’t see explanation why they’d been added to the shielding list and therefore invited for a vaccine. This is causing anxiety for a lot of people. They’re wondering: ‘Am I high risk without knowing about it?’”
A Swindon-based GP, Dr Gavin Jamie, said that, in addition to women who experienced gestational diabetes, his practice had been contacted by a patient with a fatty liver who had been told to shield, presumably because the algorithm incorrectly coded him as having severe liver disease.
NHS Digital said: “Clinicians can apply their clinical judgment and remove any of these patients from the shielded patient list, and people with concerns can speak to their GP.”
Many of the patients Brooks identified were young men, some of whom took up the offer of early vaccination.
Young, healthy people are less likely to have measurements such as body weight recorded in their health records, said Irene Petersen, professor of epidemiology and health informatics at University College London. “For young men, only about 10% of them have their weight recorded, because they barely go to see their GP. But if you have a high BMI, you’re far more likely to have this recorded, because you may have other health conditions such as type 2 diabetes,” she said.
Although it will not harm younger, healthier individuals to receive a jab before their peers, it may mean more vulnerable people have to wait longer for their doses.
“I can see that they don’t want to miss out people, which may be why they’re taking this conservative approach, but the problem is that when you make that prioritisation, you are pushing others further down the line,” Petersen said.
Other young and healthy patients may be being flagged as having serious underlying medical conditions by separate pieces of code, used to trawl through records and identify patients with serious health conditions, such as chronic heart, kidney or liver disease, a suppressed immune system or severe mental illness.
At the same time, there are those who meet the criteria to be vaccinated now who have not been invited, or mistakenly told they do not qualify – including some people who have previously been hospitalised with asthma.
Image Source: The Guardian
