RCT 490 late stage (>65% lung change chest radiography at baseline) hospitalized patients in Malaysia, showing no significant differences.
Mortality was 1.2% for ivermectin vs. 4% for control. If the same event rates continue, the trial would need to add ~13% more patients to reach statistical significance.
i.e., by continuing the trial for ~2 weeks, there is a reasonable chance of the result being a statistically significant ~69% reduction in mortality, which would equate to ~4 million lives saved if adopted at the start of the pandemic.
The mortality reduction is consistent with the results from all trials to date. While not reaching the significance threshold with the specified test, Bayesian analysis shows a 97% probability that ivermectin reduces mortality [normanfenton.com]
Authors describe the mortality results as "similar" and they are not mentioned in the visual abstract or the conclusion, suggesting substantial investigator bias with a preference for a null result.
The primary outcome is based on SpO2
<95%, however baseline SpO2
is not provided. This outcome is of limited use in evaluating treatment because it occurred before the end of treatment for > ~80% of patients. The trial was open label and the primary outcome is subject to investigator bias - clinicians could easily bias the results by altering how they monitor SpO2
, how precisely they enforced the threshold, or other aspects of SOC such as propensity to use prone positioning. Authors indicate the 95% value is from clinical stage 4, however the Malaysian government defines 94% as the threshold for stage 4 [covid-19.moh.gov.my]
, as per the NIH definition [covid19treatmentguidelines.nih.gov]
. Using death/IMV/NIV/high flow for severe (as per WHO) also shows more favorable results 
The mortality rate among all patients is too low to detect a 69% benefit with statistical significance, however the primary outcome gives us a subset of patients with severe cases that had progressed to SpO2
<95% shortly after randomization (and mostly before treatment ended). This result is statistically significant. For more discussion see: [twitter.com (B), ]
The trial started May 31, 2021 and outcomes were changed in the trial record on June 16, 2021 [clinicaltrials.gov]
. Previously the only clinical outcomes listed (under secondary outcomes) were mortality and clinical response, both at 28 days. Clinical response at 28 days would be more informative than complete recovery at day 5 as reported.
Contact information was deleted in the trial record on November 3, 2021 [clinicaltrials.gov (B)]
Data sharing: authors report that the data is available, send requests to: email@example.com. NCT04920942 (history)
Lim et al., 3 Nov 2021, Randomized Controlled Trial, Malaysia, peer-reviewed, 26 authors, study period 31 May, 2021 - 9 October, 2021, average treatment delay 5.1 days, dosage 400μg/kg days 1-5, trial NCT04920942 (history)