In a recent study posted to the medRxiv* preprint server, researchers in the United Kingdom described key measurements for monitoring clinical activity during the coronavirus disease 2019 (COVID-19) pandemic.
The COVID-19 pandemic has considerably impacted the capacity and delivery of primary and secondary care. After the first wave of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections in March 2020, the team discovered that not all clinical activities had returned to nearly normal levels by December 2020, as predicted by National Health Service (NHS) England's guidance published in July 2020. This required a tremendous amount of data analysis, which could be considered more than what the clinical and commissioning teams could reasonably monitor. Therefore, the team proposed essential primary care clinical activity metrics to facilitate routine monitoring, focused action and informed response to the COVID-19 pandemic in light of our work and in partnership with the clinical advisory group.
Study: Eleven key measures for monitoring general practice clinical activity during COVID-19 using federated analytics on 48million adults’ primary care records through OpenSAFELY. Image Credit: ratlos / Shutterstock
About the study
In this study, researchers developed important primary care activity metrics and outlined their trends during the COVID-19 pandemic.
The team performed a retrospective cohort analysis using electronic health record (EHR) data from a general practitioner (GP) primary care in England's GP practices. Within the OpenSAFELY platform, all data were securely linked, archived, and analyzed. Pseudonymized information, including coded diagnoses, drugs, and physiological measurements, and all adult patients in England who were alive and registered with either an EMIS or TPP general practice from January 2019 to December 2021 at the beginning of each month. The team also obtained demographic factors such as age, gender, geographic location of practice address, ethnicity, and index of multidimensional deprivation (IMD).
A clinical advisory panel composed of pathology leads, national clinical advisers; front-line general practitioners and pharmacists; and clinical and research professionals from the Bennett Institute was assembled to develop important markers of NHS clinical activity. Altogether, the team produced a significant number of charts showing the differences in clinical coding activity across various clinical domains using the CTV3 terminology coding hierarchy. The clinical advisory panel manually analyzed these charts for each clinical area in several online meetings in order to identify the clinical topics that could be enriched with ongoing observation and focused action.
The clinical advisory panel was requested to make recommendations for each therapeutic area, considering the following factors: high volume utilization, clinical relevance to front-line practice, and whether the factors are more generally indicative of other issues with NHS service delivery. Following these recommendations, the Bennett Institute team created custom lists of codes by hand. The clinical advisory committee was then supplied with charts related to the recently designed measures for a final evaluation and a paragraph explaining why each measure was crucial to monitor.
The team tabulated the total number of practices employing each codelist, the number of distinct patients who suffered events during the trial, and the number of events each patient experienced. The median and deciles of the monthly coding activity rates for all practices were subsequently determined.
The study cohort comprised 48,352,770 registered adult patients from 6,389 practices, representing over 98% of all practices in England. In partnership with a clinical advisory panel, 11 important clinical activity indicators were created to guide the restoration of NHS service in general practice. Regular blood tests, evaluations for ongoing conditions, risk assessments for cardiovascular disease (CVD), and blood pressure monitoring were some of these crucial procedures. The team found 447 million occurrences recorded between January 2019 and December 2021, corresponding to the 11 measures.
Between January 2019 and December 2021, the COPD review measures reported 1.16 million patients, and blood pressure monitoring was performed for 27.77 million, representing 2.6 million and 79.3 million coded events, respectively. The team also observed that the median practice level rate for 1000 registered patients was between 1.10 for COPD reviews and 65.03 for blood pressure monitoring.
All metrics' medians in April 2020 were much lower than in April 2019 and ranged between a 91.75% reduction in cholesterol testing and a 36.42% decrease in medication reviews. By April 2021, there was a decline in the median compared to April 2019 by 42.03% for blood pressure monitoring and 1.65% for thyroid testing. Based on the SRO classification system, activity in the six blood monitoring parameters had recovered within 15% of baseline by April 2021. All remaining measurements were deemed to have had a sustained decrease. In 2020, there were decreases in reviews for COPD and asthma of 39.89% and 72.73%, respectively. From April 2019 to April 2021, the rates for blood pressure monitoring and the evaluation of cardiovascular 10-year risk decreased by 42.03% and 37.74%, respectively.
Overall, the study findings demonstrated an open-source software framework that described the trends and variations in clinical activity across a large volume of primary care data. The researchers believe that the publicly accessible NHS OpenSAFELY SRO dashboards equipped with near real-time data will effectively monitor a more significant number of critical performance indicators regularly in the future.
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.
- Eleven key measures for monitoring general practice clinical activity during COVID-19 using federated analytics on 48 million adults' primary care records through OpenSAFELY, Louis Fisher, Helen J Curtis, Richard Croker, Milan Wiedemann, Victoria Speed, Christopher Wood, Andrew Brown, Lisa EM Hopcroft, Rose Higgins, et al, medRxiv 2022.10.17.22281058, DOI: https://doi.org/10.1101/2022.10.17.22281058, https://www.medrxiv.org/content/10.1101/2022.10.17.22281058v1