{"doi":"10.1002/ajh.26553","title":"Clinical predictors of <scp>COVID</scp>‐19 severity and bleeding in the <scp>ACTIV‐4B COVID</scp>‐19 outpatient thrombosis prevention trial","abstract":"The ACTIV-4B trial randomized 657 symptomatic outpatients recently infected with SARS-CoV-2 to 45 days of treatment with aspirin 81 mg qd, apixaban 2.5 mg bid, apixaban 5 mg bid, or placebo. The primary analysis of the 558 participants who initiated treatment demonstrated that the majority do not require antithrombotic therapy-- only 3 experienced an adjudicated endpoint of the primary composite outcome of all-cause mortality, symptomatic venous or arterial thromboembolism, myocardial infarction, stroke, and hospitalization for cardiovascular or pulmonary indication.1 Although the trial answered the question of whether symptomatic stable COVID-19 outpatients should be given antithrombotic therapy, closing earlier than expected due to a low number of events, the predictive clinical characteristics of all randomized participants experiencing a positively adjudicated event encompassed in the definition of the primary composite outcome (3.6%) or a hemorrhagic event (5.3%) have not been described. This analysis identifies the relevant predictors of these events and bleeding events, with implications for the management of newly diagnosed outpatients. Primary results of the ACTIV-4B trial and the study protocol, eligibility criteria, treatment allocation, methods for outcome assessment, and statistical analysis plan have been reported.1 The trial primary endpoint was defined as the composite of symptomatic deep venous thrombosis, pulmonary embolism, arterial thromboembolism, myocardial infarction, ischemic stroke, hospitalization for cardiovascular or pulmonary events, and all-cause mortality at 45 days. Although the primary analysis population was limited to randomized participants who took at least one dose of medication, all suspected primary endpoints occurring after randomization were adjudicated; this analysis includes all adjudicated primary endpoint events occurring in the complete set of randomized participants. The principal safety endpoints were ISTH defined major bleeding and clinically relevant non-major bleeding (CRNMB).2 Bleeding events not meeting these definitions were classified as minor. Only suspected major or CRNMB bleeding events were adjudicated, however, all bleeding events were reviewed by investigators. The cumulative incidence of each endpoint was estimated using Kaplan–Meier methods; differences by assigned treatment were tested with log rank statistics. Logistic regression was used for the occurrence of the primary endpoint, and Cox regression for time to the bleeding endpoint. Candidate explanatory variables included assigned treatment, all collected demographic and clinical history variables (age, sex, race/ethnicity, body mass index, history of smoking, past history of hypertension, diabetes, or venous thromboembolism), key laboratory variables (D-dimer, CRP, and creatinine clearance) and the number of days between SARS-CoV-2 test and randomization. Multivariable logistic and Cox regression models were created by including all variables individually associated with the outcome (p < 0.10) and applying backward stepwise regression to sequentially remove each factor with a p-value ≥0.10. The c-statistic and Harrell's concordance index quantified the discrimination of the final logistic and Cox regression models, respectively. Among all 657 randomized participants, 26 suspected primary endpoints occurred during the 45-day period after randomization with 24 confirmed by adjudication (3.6%). The majority (21/24) occurred following randomization but before participants initiated assigned treatment, a median interval of 3 days [first quartile (Q1) - third quartile (Q3): 2–5 days].1 All 24 confirmed events were hospitalizations for cardiovascular or pulmonary indications, with 23 admitted for progressive COVID-19 pneumonia, and 1 admitted for suspected myocardial infarction which was adjudicated as negative. Among the 23 patients hospitalized with progressive COVID-19 pneumonia, one non-fatal deep venous thromb","journal":"American Journal of Hematology","year":2022,"id":289903,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":4,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9613,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2022-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":309260,"name":"Maria M. Brooks","orcid":"0000-0002-2030-7873","position":1,"is_corresponding":false},{"id":25014,"name":"Frank C. Sciurba","orcid":"0000-0001-7468-2225","position":2,"is_corresponding":false},{"id":549895,"name":"Zhuxuan Fu","orcid":"0000-0002-9190-195X","position":3,"is_corresponding":false},{"id":21979,"name":"Paul M. Ridker","orcid":"0000-0003-1249-4522","position":4,"is_corresponding":false},{"id":287954,"name":"Jean M. Connors","orcid":"0000-0001-6445-582X","position":0,"is_corresponding":true}],"reference_count":9,"raw_metadata":null,"created_at":"2026-07-19T00:30:26.667578Z","pmid":"35358342","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}