Clinical coding of long COVID in English primary care: a federated analysis of 58 million patient records in situ using OpenSAFELY

doi:10.3399/bjgp21X717449

All letters are subject to editing and may be shortened. General letters can be sent to bjgpdisc{at}rcgp.org.uk (please include your postal address for publication), and letters responding directly to BJGP articles can be submitted online via eLetters. We regret we cannot notify authors regarding publication.

For submission instructions visit: bjgp.org/letters

Our paper in the BJGP¹ on diagnostic coding of long COVID described wide variation in use of the codes, and dramatically lower rates of use of the diagnostic coding when compared with long COVID as measured in self-reported survey data. As COVID is an unprecedented and evolving situation, we are providing updated analyses for key findings. A report on long COVID diagnostic code usage can be found on the OpenSAFELY reports website (reports. opensafely.org).²

As of this letter, data are current to 5 September 2021, providing 19 weeks’ additional follow-up time from the original paper. During this period, 33 827 additional people had a code for long COVID recorded, making 57 100 people in total. The rate at which new diagnoses are being recorded remains largely unchanged. The overall prevalence of coded long COVID diagnosis in the total population is now 99.6 per 100 000, compared with 40.1 per 100 000 in the original paper, due solely to greater followup time. As before, prevalence of coding in EMIS practices (126.0) remains higher than in TPP practices (63.1); however, this gap is diminishing over time, with a rate ratio of 2.0 now, and 2.6 at 19 weeks previously.

As we discussed in our article, it is critical for research and planning of services that GPs are able to appropriately code cases of long COVID. We will continue to update this report regularly to inform clinical coding of long COVID. Readers are encouraged to view the full updated report to see trends at the time of reading.²

REFERENCES

1.↵
1. Walker AJ,
2. MacKenna B,
3. Inglesby P,
4. et al.
(2021) Clinical coding of long COVID in English primary care: a federated analysis of 58 million patient records in situ using OpenSAFELY. Br J Gen Pract, DOI: https://doi.org/10.3399/BJGP.2021.0301.
2.↵
1. Walker AJ,
2. MacKenna B,
3. Inglesby P,
4. et al.
(2021) Clinical coding of long COVID in English primary care: a federated analysis of 58 million patient records in situ using OpenSAFELY. DOI: http://doi.org/10.53764/rpt.3917ab5ac5.