Patients who were hospitalized for COVID-19 and who were not critically ill were randomized in a response-adaptive manner to therapeutic-dose anticoagulation with heparin vs. usual-care pharmacologic thromboprophylaxis. The outcome reported here is the effect on organ support-free days (i.e. days without oxygen delivered by high-flow nasal cannula, noninvasive or invasive mechanical ventilation, or the use of vasopressors or inotropes). The statistical result was a median adjusted odds ratio 1.27 (95% credible interval 1.03 to 1.58), based on 1,740 events among 2,219 participants with known outcome out of 2,244 enrolled participants. The probability of superiority of therapeutic-dose anticoagulation with heparin was 98.6%. The risk of bias in this effect estimate is of extremely serious concern based on a serious concern for confounding covariate bias (confounding difference in calendar time), a very serious concern for performance bias (inadequate blinding of intervention deliverers who may determine the outcome based in part on exposure status), and very serious concern for analysis bias (bias related to selection of the analysis, and early trial termination).

[No data.]