Whitfield, Emma; (2024) Leveraging electronic health records to evaluate the potential for earlier diagnosis. Doctoral thesis (Ph.D), UCL (University College London).

Text Emma Whitfield - PhD Thesis Dec 24.pdf - Submitted Version Access restricted to UCL open access staff until 1 January 2027. Download (14MB)

Abstract

Improving diagnostic quality and safety is a priority for contemporary healthcare systems. Understanding if, where, when, how, and for whom earlier diagnosis may be possible across a spectrum of conditions can help guide research priorities and target interventions. Evidence from population-based electronic health records (EHRs) can support diagnostic quality and safety research, but such analyses have thus far been concentrated in diseases such as cancer or acute conditions (e.g., sepsis). This thesis uses the current evidence from diagnostic research in cancer to develop a proposed framework for evaluating the potential for earlier diagnosis of a range of conditions using EHR data. Following a literature review of pre-diagnostic healthcare use in non-neoplastic conditions, I analysed linked primary care, secondary care, and death records to carry out three quantitative studies exploring measures of the potential for earlier diagnosis of 18 conditions. The first study examines the length of diagnostic windows, showing that rates of GP appointments start to increase in some patients several months or years before diagnosis. The second study describes the frequency of emergency presentations and their prognostic implications, alongside the frequency of death certificate-only diagnoses. Emergency presentations account for notable, yet variable, proportions of diagnoses across all studied conditions and are associated with increased mortality and hospitalisation risk in the year post-diagnosis. Lastly, I assess gender differences in both diagnostic window length and emergency presentations, showing that a ‘gender health gap’ may be affecting diagnostic processes. Further research can enhance these enquiries, by extending analyses to additional conditions, and by using electronic health records to directly measure diagnostic quality and safety indicators, such as missed diagnostic opportunities. In combination with future research, these findings could guide the design and implementation of interventions to aid earlier diagnosis.

Download activity - last month

Export as CSVJSONXML

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as XMLJSONCSV

Archive Staff Only

View Item