Objective:

To characterize temporal, demographic, and geographic patterns of documented Long COVID using the Texas All-Payer Claims Database (TX-APCD).

Approach:

• Population-level assessment: Analyzed weekly COVID-19 emergency department (ED) encounters and Long COVID diagnoses from October 2021 through October 2023 across SARS-CoV-2 variant-dominated periods.
• Individual-level survival analysis: Compared time to first documented Long COVID diagnosis following ED and non-ED COVID-19 encounters, adjusting for age, sex, payer, and Texas Health and Human Services region.
• County-level spatial assessment: Examined geographic, urban–rural, and coastal–inland variation. Long COVID was identified using ICD-10-CM code U09.9, counting only each person’s first diagnosis.

Key Findings:

• Weekly COVID-19 ED activity preceded Long COVID claims, with the strongest association at a 2-week lag during the Omicron BA.1 period and a 3-week lag during later variant periods.
• The median documented interval was 22 days after an ED encounter versus 26 days after a non-ED encounter.
• Crude incidence was highest among Medicare Fee-for-Service beneficiaries and individuals older than 70 years, with higher rates among females than males.
• Documented diagnoses clustered in High Plains and Northwest Texas.

Interpretation:

Claims data can identify populations and regions with greater documented Long COVID burden. The short recorded intervals reflect diagnosis and documentation pathways rather than established clinical latency.

Limitations:

• The database covers approximately 60% of insured Texans and excludes populations such as veterans and uninsured individuals.
• Claims do not capture infections identified only through home testing or managed without healthcare contact.
• Diagnosis-code reliance may underestimate burden and does not characterize symptoms, severity, or functional impairment.

Conclusion:

TX-APCD supports surveillance of documented Long COVID and may help guide expansion of postacute care capacity, particularly for older adults, Medicare beneficiaries, and regions with higher recorded incidence.