Electronic health records to capture primary outcome measures: two case studies in HIV prevention research

• Published in: Current controlled trials in cardiovascular medicine 2023-03, Vol.24 (1), p.244-244, Article 244
• Main Authors: Dunn, David, McCabe, Leanne, White, Ellen, Delpech, Valerie, Kirwan, Peter D, Khawam, Jameel, Croxford, Sara, Ward, Denise, Brodnicki, Elizabeth, Rodger, Alison, McCormack, Sheena
• Format: Article
• Language: English
• Subjects: Clinical outcomes; Clinical trials; Confidence intervals; Diagnosis; Disease prevention; Electronic Health Records; HIV; HIV infection; HIV Infections - diagnosis; HIV Infections - epidemiology; HIV Infections - prevention & control; Human immunodeficiency virus; Humans; Infections; Information management; Internet; Medical records; Medical tests; Methods; Outcome and process assessment (Health Care); Performance evaluation; Pre-Exposure Prophylaxis; Prevention; Primary outcome; Public health; Randomized Controlled Trials as Topic; Research Design; Routinely collected health-care data; Sexual behavior
• ISSN: 1745-6215; 1745-6215
• DOI: 10.1186/s13063-023-07264-6

There is increasing interest in the use of electronic health records (EHRs) to improve the efficiency and cost-effectiveness of clinical trials, including the capture of outcome measures. We describe our experience of using EHRs to capture the primary outcome measure - HIV infection or the diagnosis of HIV infection - in two randomised HIV prevention trials conducted in the UK. PROUD was a clinic-based trial evaluating pre-exposure prophylaxis (PrEP), and SELPHI was an internet-based trial evaluating HIV self-testing kits. The EHR was the national database of HIV diagnoses in the UK, curated by the UK Health Security Agency (UKHSA). In PROUD, linkage to the UKHSA database was performed at the end of the trial and identified five primary outcomes in addition to the 30 outcomes diagnosed by the participating clinics. Linkage also produced an additional 345 person-years follow-up, an increase of 27% over clinic-based follow-up. In SELPHI, new HIV diagnoses were primarily identified via UKHSA linkage, complemented by participant self-report through internet surveys. Rates of survey completion were low, and only 14 of the 33 new diagnoses recorded in the UKHSA database were also self-reported. Thus UKHSA linkage was essential for capturing HIV diagnoses and the successful conduct of the trial. Our experience of using the UKHSA database of HIV diagnoses as a source of primary outcomes in two randomised trials in the field of HIV prevention was highly favourable and encourages the use of a similar approach in future trials in this disease area.