Community-acquired pneumonia identification from electronic health records in the absence of a gold standard: A Bayesian latent class analysis

Community-acquired pneumonia (CAP) is common and a significant cause of mortality. However, CAP surveillance commonly relies on diagnostic codes from electronic health records (EHRs), with imperfect accuracy. We used Bayesian latent class models with multiple imputation to assess the accuracy of CAP diagnostic codes in the absence of a gold standard and to explore the contribution of various EHR data sources in improving CAP identification. Using 491,681 hospital admissions in Oxfordshire, UK, from 2016 to 2023, we investigated four EHR-based algorithms for CAP detection based on 1) primary diagnostic codes, 2) clinician-documented indications for antibiotic prescriptions, 3) radiology free-text reports, and 4) vital signs and blood tests. The estimated prevalence of CAP as the reason for emergency hospital admission was 13.6% (95% credible interval 13.3-14.0%). Primary diagnostic codes had low sensitivity but a high specificity (best fitting model, 0.275 and 0.997 respectively), as did vital signs with blood tests (0.348 and 0.963). Antibiotic indication text had a higher sensitivity (0.590) but a lower specificity (0.982), with radiology reports intermediate (0.485 and 0.960). Defining CAP as present when detected by any algorithm produced sensitivity and specificity of 0.873 and 0.905 respectively. Results remained consistent using alternative priors and in sensitivity analyses. Relying solely on diagnostic codes for CAP surveillance leads to substantial under-detection; combining EHR data across multiple algorithms enhances identification accuracy. Bayesian latent class analysis-based approaches could improve CAP surveillance and epidemiological estimates by integrating multiple EHR sources, even without a gold standard for CAP diagnosis.Author summary: Community-acquired pneumonia (CAP) is a common and serious illness, but current surveillance usually relies on diagnostic codes from electronic health records (EHRs), which can be inaccurate. In this study, we analysed half a million hospital admissions in Oxfordshire, UK (2016–2023), to evaluate how well different EHR-based methods detect CAP. Since there is no perfect diagnostic test for CAP, we used Bayesian latent class models to estimate the performance of each method. We found that relying solely on primary diagnostic codes often misses CAP cases, while other data including antibiotic prescriptions, radiology reports, and patients’ vital signs and blood tests could improve detection. When combined all data sources together, the performance of identifying CAP increased significantly. Our findings suggest that integrating multiple types of EHR data improves CAP identification. This approach could support more accurate disease surveillance and a better understanding of CAP prevalence, even in the absence of a perfect diagnostic standard.