Background: Global efforts are ongoing to identify effective therapeutics to combat the coronavirus disease 2019 (COVID-19); for example, the UK RECOVERY trial demonstrated a mortality reduction among hospitalized patients with severe COVID-19 randomized to dexamethasone (DEX). Real-world data allow observation of changes in prescribing trends in response to new research amidst uncertainty and limited treatment options.

Objectives: Describe treatment trends for hospitalized COVID-19 patients in the US, overall and stratified by COVID-19 severity.

Methods: Using HealthVerity claims (March-December 2020), we identified hospitalized patients with a COVID-19 diagnosis or positive SARS-CoV-2 laboratory result. We report the weekly period prevalence (%) of patients newly treated (90-day washout) during the first week of the hospital stay with therapies of interest, including hydroxychloroquine (HCQ), azithromycin (AZ), remdesivir (RDV), and DEX. We examined trends for each treatment by calendar week of admission overall, by COVID-19 severity (requiring neither invasive mechanical ventilation [IMV] nor supplementary oxygen [O2] vs requiring IMV or O2), and, for DEX, before and after publication of RECOVERY trial results.

Results: Among the 85,970 hospitalized COVID-19 patients included, the most commonly used treatments were AZ (42.2%) and HCQ (39.2%) early in the pandemic (March/April 2020), followed by DEX (59.6%) and RDV (42.1%) later on (November/December 2020). HCQ treatment declined sharply after highest use in March (51.8%), with <1% use from July to December. AZ use declined from March (52.8%) to May (25.6%) before stabilizing at around 35% weekly use in August. Following the RECOVERY publication on June 16, DEX use increased sharply from June (5.8%) to July (48.1%), a trend consistent for both levels of COVID-19 severity. RDV use increased in early July among patients requiring IMV/O2, with use exceeding 10% by mid-July, while uptake in patients requiring neither IMV nor O2 lagged, exceeding 10% by October.

Conclusions: A considerable temporal shift in real-world prescribing trends in US hospitalized COVID-19 patients was observed, with RDV and DEX surpassing HCQ following publication of emerging scientific data and regulatory-related actions (e.g., Emergency Use Authorizations). Publication of the UK RECOVERY trial was followed by an immediate and substantial increase in DEX use in the US, among hospitalized COVID-19 patients in both severity levels. These trends may also have been influenced by factors such as physician experience with disease management and therapies, access/cost, guidelines, and perceived efficacy/safety relative to other treatment options.