Accuracy of administrative codes for hidradenitis suppurativa in the veterans affairs healthcare system

Summary: Abstract Body: The positive predictive value (PPV) and negative predictive value (NPV) of a single administrative code for hidradenitis suppurativa (HS) have been reported on in the literature using the general population as a control, but this resulted in no false negative cases. We aimed to assess the diagnostic accuracy of a single ICD code for HS in the Veterans Affairs Healthcare System (VAHCS) using patients with codes for abscess as a control group. Patients with an ICD code for HS from 2011 to 2022 were identified, and 140 were randomly selected for chart review. Additionally, 140 age- and sex-matched controls with a code for cutaneous abscess (without codes for HS) were identified. Two blinded reviewers evaluated charts for HS diagnosis, defined by three Dessau criteria at the date of ICD code entry, HS in the problem list, or evidence of HS on review of recent documentation going back to the date of the ICD code entry. A third blinded reviewer resolved discrepancies. The PPV of a single ICD code for HS was 0.864 (95% CI 0.794-0.914), and NPV was 0.971 (95% CI 0.924-0.991). Sensitivity and specificity were not calculated, as defining cohorts based on the outcomes of the diagnostic test being evaluated (ICD codes) introduces sampling error. Despite using a control cohort with a higher presumed prevalence of HS, our NPV remained high. Limitations include variability in physical exam documentation, which may underestimate HS cases. This method of cohort identification by ICD code entry will be used in planned VAHCS database studies evaluating risk factors for HS progression. John Meisenheimer^{2, 1}, Madeeha Rizvi³, Emily Grimshaw⁴, Zachary Wendland^{5, 6, 2}, Fatai Y. Agiri⁷, Julie Lynch⁷, Noah Goldfarb^{2, 1} 1. Department of Dermatology, University of Minnesota Twin Cities, Minneapolis, MN, United States. 2. Department of Dermatology, Minneapolis VA Medical Center, Minneapolis, MN, United States. 3. Des Moines University College of Osteopathic Medicine, Des Moines, IA, United States. 4. Department of Surgery, Minneapolis VA Medical Center, Minneapolis, MN, United States. 5. HCA Florida JFK Hospital, Atlantis, FL, United States. 6. University of Miami Miller School of Medicine, Miami, FL, United States. 7. VA Informatics and Computing Infrastructure (VINCI), VA Salt Lake City Health Care System, Salt Lake City, UT, United States. Clinical Research: Epidemiology and Observational Research