Rothnie, Kieran J; Müllerová, Hana; Hurst, John R; Smeeth, Liam; Davis, Kourtney; Thomas, Sara L; Quint, Jennifer K; (2016) Validation of the Recording of Acute Exacerbations of COPD in UK Primary Care Electronic Healthcare Records. PloS one, 11 (3). e0151357-. ISSN 1932-6203 DOI: https://doi.org/10.1371/journal.pone.0151357
Permanent Identifier
Use this Digital Object Identifier when citing or linking to this resource.
Abstract
BACKGROUND: Acute Exacerbations of COPD (AECOPD) identified from electronic healthcare records (EHR) are important for research, public health and to inform healthcare utilisation and service provision. However, there is no standardised method of identifying AECOPD in UK EHR. We aimed to validate the recording of AECOPD in UK EHR. METHODS: We randomly selected 1385 patients with COPD from the Clinical Practice Research Datalink. We selected dates of possible AECOPD based on 15 different algorithms between January 2004 and August 2013. Questionnaires were sent to GPs asking for confirmation of their patients' AECOPD on the dates identified and for any additional relevant information. Responses were reviewed independently by two respiratory physicians. Positive predictive value (PPV) and sensitivity were calculated. RESULTS: The response rate was 71.3%. AECOPD diagnostic codes, lower respiratory tract infection (LRTI) codes, and prescriptions of antibiotics and oral corticosteroids (OCS) together for 5-14 days had a high PPV (>75%) for identifying AECOPD. Symptom-based algorithms and prescription of antibiotics or OCS alone had lower PPVs (60-75%). A combined strategy of antibiotic and OCS prescriptions for 5-14 days, or LRTI or AECOPD code resulted in a PPV of 85.5% (95% CI, 82.7-88.3%) and a sensitivity of 62.9% (55.4-70.4%). CONCLUSION: Using a combination of diagnostic and therapy codes, the validity of AECOPD identified from EHR can be high. These strategies are useful for understanding health-care utilisation for AECOPD, informing service provision and for researchers. These results highlight the need for common coding strategies to be adopted in primary care to allow easy and accurate identification of events.
Item Type | Article |
---|---|
Faculty and Department |
Faculty of Epidemiology and Population Health > Dept of Non-Communicable Disease Epidemiology Academic Services & Administration > Directorate Faculty of Epidemiology and Population Health > Dept of Infectious Disease Epidemiology (-2023) |
Research Centre | EHR Research Group |
PubMed ID | 26959820 |
ISI | 371992300118 |
Related URLs |