Retrospective database study of 5683 patients, 692 received HC
Q+AZ, 200 received HC
Q, 203 received ivermectin, 1600 received AZ, 358 received ivermectin+AZ, and 2630 received standard of care.
Q+AZ was associated with 84% higher all-cause mortality compared to standard care, aHR = 1.84 [1.12-3.02]. Substantial confounding by indication is likely, with more serious cases more likely to receive treatment. Substantial increased treatment mortality is seen, independent of the actual treatment used. KM curves also agree with this, showing increased mortality early on, with very high mortality on day 2, but every treatment did better than standard of care at the latest available date.
Authors use a machine learning based propensity scoring system that appears over-parameterized and likely to result in significant overfitting and inaccurate results. Essentially they test for all interactions between two and three covariates. The nature and large number of covariates means many random correlations may be found. COVID-19 severity is not used.
See also: .
death, ↑84.0%, p=0.02