Abstract
Background
The long-term duration of post COVID condition (PCC, “Long COVID”) remains unclear. In this study, we estimated the risk of healthcare encounters in Military Health System (MHS) beneficiaries for the 12 months post SARS-CoV-2 diagnosis, adjusting for prior healthcare use, and compared to those without known prior SARS-COV-2 diagnosis.
Methods
Follow up continues for the Epidemiology, Immunology and Clinical Characteristics of Emerging Infectious Diseases of Pandemic Potential (EPICC) COVID-19 cohort study MHS beneficiaries who were tested for SARS-COV-2 or vaccinated from March 2020 to April 2022. Participants with SARS-COV-2 diagnosis from 3/1/20 through 12/31/21 were matched 1:1 with participants in the same age group with no record of SARS-COV-2 diagnosis. We identified categories of ICD-10 diagnoses occurring from 3 months before through 12 months after first SARS-COV-2 diagnosis (or matched time point) from electronic medical records. Multivariable Poisson regression models were used to estimate the risk of ICD-10 diagnosis categories for those with past SARS-COV-2 diagnosis, compared to those with no SARS-COV-2 diagnosis, adjusting for age, sex, BMI, variant era, and prior healthcare use.
Table 1.
Characteristics of matched Epidemiology, Immunology and Clinical Characteristics of Emerging Infectious Diseases of Pandemic Potential (EPICC) participants included in analyses. Statistical comparisons are Pearson’s Chi squared tests, unless otherwise specified.
Results
Analyses included 1,819 matched pairs with a median age of 35 years. Participants were primarily male (63.0%) or white (57.3%) (Table 1) and severe acute COVID-19 was infrequent (9.6% hospitalized). Compared to those without a history of SARS-COV-2 diagnosis, medical encounters for all diagnosis groups (pulmonary, cardiovascular, diabetes, anxiety/depression, and neurology) were elevated in the first month after SARS-COV-2 diagnosis (Figures 1 and 2). Among the different diagnosis categories, only pulmonary diagnoses remained elevated at 9 months post-infection compared to those without a history of SARS-COV-2 diagnosis (risk ratio: 1.95 (95% CI 1.34, 2.83)).
Percent of EPICC participants with medical encounters / diagnoses (by organ system or other domain) in health records. Participants without a history of SARS-CoV-2 infection were assigned the infection date of their matched case.
Poisson regression analysis run using each category of healthcare encounters (pulmonary, cardiovascular, diabetes, neurology, and anxiety/depression) as the outcome. The models included time in 30-day periods around SARS-CoV-2 diagnosis date (or matched time point), SARS-CoV-2 diagnosis status, age, sex, BMI category, and variant/calendar period (calendar times with predominant Ancestral, Alpha, or Delta circulation), as well as a random effect for participant.
Conclusion
MHS beneficiaries with prior SARS-COV-2 diagnosis were at higher risk of pulmonary-associated healthcare encounters through 9 months post-infection compared to those without prior SARS-COV-2 diagnosis, even after adjusting for baseline characteristics and calendar time. Future work will assess the effect of vaccination and boosting on this relationship.
Disclosures
Michael Morris, MD, Janssen Pharmaceuticals: Paid speaker (unrelated to this project and COVID-19 in general) Ryan C. Maves, MD, Sound Pharmaceuticals: Grant/Research Support Mark Simons, PhD, AstraZeneca: TBD Timothy Burgess, MD, AstraZeneca: TBD Simon Pollett, MBBS, AstraZeneca: The IDCRP and the Henry M. Jackson Foundation (HJF) were funded to conduct an unrelated phase III COVID-19 monoclonal antibody immunoprophylaxis trial