OBJECTIVES: Assess real-world evidence data on the prevalence and impact of long COVID (LC) to establish a baseline for the value of potential therapeutic interventions.

METHODS: This study was a retrospective, longitudinal analysis of administrative claims from multiple payer channels spanning 4/1/2020-6/30/2022. Inclusion criteria: 1) ICD-10-CM diagnosis code of COVID-19 (U07.1) on or after 4/1/2020 (COVID-19 diagnosis date=index date), 2) 18+ years of age on index, and 3) at least -365/+30 days of continuous plan enrollment surrounding index. Employing a conservative LC definition, patients were classified as LC if they presented at least 1 claim ≥ 28 days following the index date which included both a COVID-19 diagnosis and ≥1 of 8 LC-related symptoms. LC and non-LC patients were compared on demographics, COVID-19 symptoms, healthcare utilization, and medical costs. Descriptive statistics were presented for outcomes, and bivariate tests of significance were used to assess differences between cohorts.

RESULTS: Of 4,938,801 medically attended COVID-19 patients meeting inclusion criteria, 386,153 (7.8%) qualified as LC. The LC patients were older (Mean(SD) = 67.0(19.0) vs. 51.0(20.7)), were more likely to be female (65.1% vs. 60.4%), were in poorer health (Deyo-Charlson Comorbidity Index=3.51(3.24) vs. 1.47(2.45)), and presented greater baseline total medical expenditures ($39,769($60,401) vs. $15,275($35,640); p < 0.0001). On index, LC patients had a higher rate of LC-related symptoms, and in the 180-day post-index period, LC patients incurred increased total medical costs ($38,874($54,098) vs. $7,319($18,439); p < 0.001) and greater use of inpatient and outpatient medical services.

CONCLUSIONS: Patients with LC presented elevated rates of symptoms and incurred 5-fold greater medical costs post-index compared to non-LC patients. This study is one of the first to longitudinally quantify the cost and symptom burden of LC in a real-world setting and helps to establish a baseline for the value of potential therapeutic interventions.