Uptake of the Post COVID-19 Condition ICD-10-CM Diagnosis Code By Social Factors

ABSTRACT

Objectives: Post-COVID conditions (PCC) are increasingly reported in people who had COVID. Certain racial or socioeconomic groups may be at greater risk for PCC and less likely to seek care. We examined the uptake of the new ICD-10-CM diagnosis code for PCC in routine clinical practice in the United States and how it varied by race and payer group. Method(s) Using the Optum de-identified Electronic Health Record (EHR) dataset, we identified patients with an ICD-10-CM code for PCC (U09.9) between October 1, 2021, through March 31, 2022, with 6 months of prior EHR activity. The earliest diagnosis defined the index date. All concurrent diagnoses were measured on the index date. Prior COVID diagnosis was assessed using all available data before the index date. Result(s) There were 23,647 patients 9.9% were African American, 12.1% had Medicaid, and 2.4% were uninsured. There was an overrepresentation of white patients among those with PCC (78.6% compared with 69.6% of the overall EHR in 2021). More African American (24.1%), Medicaid (23.1%), and uninsured (27.5%) patients were diagnosed in the inpatient setting or emergency department than whites (14.0%) and commercially insured patients (10.0%). Among racial groups, African Americans had the highest percentage of documented prior COVID diagnosis at 63.6%. Of concurrent diagnoses, shortness of breath and acute respiratory failure with hypoxia were higher among African Americans (13.9% and 6.1%, respectively) than whites (11.5% and 4.3%, respectively). The same pattern was seen when comparing Medicaid and uninsured to commercial payors. Conclusion(s) The PCC code was used differently across racial groups and payor types and captures varying manifestations of PCC. The differences in diagnosis locations underscore the importance of using data capturing all care settings when conducting studies using this code. Subgroup analyses are important for future studies using U09.9 due to variability in code application.Copyright © 2023