ABSTRACT
OBJECTIVES: Describe the epidemiology of sepsis across the transition from the International Classification of Diseases, 9th Edition, and International Classification of Diseases, 10th Edition, coding systems, evaluating estimates of two previously published International Classification of Diseases, 10th Edition, coding strategies. DESIGN: Serial cross-sectional analysis. SETTING: Healthcare Utilization Project's annual Nationwide Inpatient Sample of U.S. hospital discharges, 2012-2017. PATIENTS: Discharges greater than or equal to 18 years old, which met one of the three case definitions for sepsis. For the records using International Classification of Diseases, 9th Edition, we used previously published modified Angus criteria, and for records using International Classification of Diseases, 10th Edition, we deployed a case definition used by the Centers for Medicare & Medicaid Services and a case definition developed by the Institute for Health Metrics and Evaluation. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: During the study period, there were discontinuities in the sepsis incidence estimates using the modified Angus International Classification of Diseases, 9th Edition, criteria in 2014 and either Centers for Medicare & Medicaid Services or Institute for Health Metrics and Evaluation International Classification of Diseases, 10th Edition, criteria in 2016. In 2014, there were an estimated 1,009 cases (95% CI, 989-1,030) of modified Angus sepsis per 100,000 persons, whereas in 2016, there were 709 cases (95% CI, 694-724) of Centers for Medicare & Medicaid Services sepsis and 1,498 cases (95% CI, 1,471-1,092) of Institute for Health Metrics and Evaluation sepsis per 100,000 persons. Furthermore, the Institute for Health Metrics and Evaluation definition identified a sepsis cohort with similar hospital characteristics but a younger age distribution, higher proportion of women, lower severity of illness, and lower hospital mortality. CONCLUSIONS: The Institute for Health Metrics and Evaluation International Classification of Diseases, 10th Edition, coding strategy for identifying sepsis may capture a larger patient population within administrative datasets that are different from those identified with previously deployed International Classification of Diseases-based methods. Further work is required to determine the optimal International Classification of Diseases, 10th Edition, coding strategy for use in hospital discharge data.