Changes in cardiovascular disease monitoring in English primary care during the COVID-19 pandemic: an observational cohort study
Bankhead C., Lay-Flurrie S., Nicholson B., Sheppard J., Gale C., Liyanage H., McGagh D., Minchin M., Perera R., Sherlock J., Smith M., Thomas NPB., Drakesmith CW., de Lusignan S., Hobbs FDR.
Objective To quantify the impact and recovery in cardiovascular disease monitoring in primary care associated with the first COVID-19 lockdown. Design Retrospective nationwide primary care cohort study, utilising data from 1st January 2018 to 27 th September 2020. Setting We extracted primary care electronic health records data from 514 primary care practices in England contributing to the Oxford Royal College of General Practitioners Clinical Informatics Digital Hub (ORCHID). These practices were representative of English primary care across urban and non-urban practices. Participants The ORCHID database included 6,157,327 active patients during the study period, and 13,938,390 patient years of observation (final date of follow-up 27 th September 2020). The mean (SD) age was 38±24 years, 49.4% were male and the majority were of white ethnicity (65% [21.9% had unknown ethnicity]) Exposure The primary exposure was the first national lockdown in the UK, starting on 23 rd March 2020. Main outcome measures Records of cholesterol, blood pressure, HbA1c and International Normalised Ratio (INR) measurement derived from coded entries in the primary care electronic health record. Results Rates of cholesterol, blood pressure, HbA1c and INR recording dropped by 23-87% in the week following the first UK national lockdown, compared with the previous week. The largest decline was seen in cholesterol (IRR 0.13, 95% CI 0.11 to 0.15) and smallest for INR (IRR 0.77, 95% CI 0.72 to 0.81). Following the immediate drop, rates of recorded tests increased on average by 5-9% per week until 27 th September 2020. However, the number of recorded measures remained below that expected for the time of year, reaching 51.8% (95% CI 51.8 to 51.9%) for blood pressure, 63.7%, (95% CI 63.7% to 63.8%) for cholesterol measurement and 70.3% (95% CI 70.2% to 70.4%) for HbA1c. Rates of INR recording declined throughout the previous two years, a trend that continued after lockdown. There were no differences in the times series trends based on sex, age, ethnicity or deprivation. Conclusions Cardiovascular disease monitoring in English primary care declined substantially from the time of the first UK lockdown. Despite a consistent recovery in activity, there is still a substantial shortfall in the numbers of recorded measurements to those expected. Strategies are required to ensure cardiovascular disease monitoring is maintained during the COVID-19 pandemic.