Abstract
BACKGROUND: National case rates of chlamydia and gonorrhea (CT/GC) among American Indian and Alaska Native (AI/AN) persons are disproportionately high. The Indian Health Service (IHS), which provides health care to members of federally recognized tribes, does not currently have a dedicated CT/GC surveillance system. The purpose of this study was to validate the use of CT/GC diagnostic codes for estimating diagnosed CT/GC infections among AI/AN persons who use IHS services. METHODS: We conducted a retrospective study using IHS medical records from all persons 15 years and older from 2016 to 2021. We linked records with CT (A56, A74) and GC (A54, O98.2) International Classification of Diseases, 10th Revision, Clinical Modification diagnostic codes to laboratory results within 30 days for each person. We calculated the sensitivity, specificity, and positive and negative predictive values of CT/GC diagnostic codes using laboratory test results as the reference standard. RESULTS: We identified more than 1.6 million CT/GC laboratory tests, and 52,815 CT and 19,971 GC diagnostic codes. Diagnostic code sensitivity was slightly higher for CT (54%) than GC (50%). Specificity, positive predictive value, and negative predictive value were high for CT and GC (range, 83.3%-99.8%). About one-third of CT/GC diagnostic codes could not be linked to a test result. CONCLUSIONS: The validation indicates that diagnostic codes align well with linked laboratory test results. However, because of the relatively large number of diagnostic codes and positive test results that could not be linked, combining the 2 would inform more reliable estimates of diagnosed CT/GC infections among AI/AN persons who use IHS for health care.