Data omitted from Soto Becerra et al; the one study showing worse outcomes with ivermectin

Missing data in Soto Becerra et al; the one study showing worse outcomes with ivermectin. PeterJYim1

The NIH COVID-19 Treatment Guidelines provide the following recommendation on the use of ivermectin:

“There are insufficient data for the COVID-19 Treatment Guidelines Panel (the Panel) to recommend either for or against the use of ivermectin for the treatment of COVID-19…”

There are 11 clinical studies of ivermectin that are cited as the basis for this recommendation. One of those studies showed worse outcomes for patients receiving treatment with ivermectin: Soto Becera et al released onMedrxiv.

After Soto Becerra et al was released, an administrator at Instituto de Evaluación de Tecnologías en Salud e Investigación, EsSalud (IETSI) the organization that employed the authors of that article wrote a memo entitled “Memorando №1843 IETSI-ESSALUD 2020”. In that memorandum, the administrator provided a table (page 4) showing the results of the study. Those results were not included in Soto Becera et al. The results in this table showed a significant improvement in survival without ICU admission in patients who received ivermectin. The table itself is in English. The caption to the table is in Spanish. Translated, the caption to the table is:

“Sensitivity Analysis 2 . Adjusted for oxygen saturation at admission, respiratory rate at admission, heart-rate at admission, systolic blood pressure at admission, temperature at admission, (variables with 40% to 60% missing data)- Multiple imputation by GBM was used. Assuming proportional hazards.”

I have contacted three of the four authors of the study (PSB through email, CC through email and by Tweet, and YHR through Twitter direct message) to ask if the results that were omitted from their publication are accurate. None of those authors has responded. I contacted the corresponding author earlier (RVAC through email) to ask about the integrity of the data. He has not responded.

Without further input from the authors, it is difficult to know if the results shown in the table in the Memorandum №1843 are valid. However, the table of results in Memorandum №1843 raises serious questions about Soto Becerra et al:

  1. The table of results from Memorandum №1843 showed a 30-day follow-up that was required by the clinical protocol. The results from Soto Becerra et al were based on post hoc analysis; patient follow-up was only until oxygen prescription, ICU transfer or death.
  2. The table of results in Memorandum №1843 showed that data on clinical presentation was collected and included in the analysis. This included oxygen saturation, respiration rate at presentation, heart-rate, blood-pressure, and temperature at admission. Soto Becerra et al omitted all of that data.

The deviation from the clinical protocol (item 1) and the omission of clinical data (item 2) are not mentioned in the Soto Becerra et al. Until these issues are addressed, the results from this article should be viewed with skepticism. These shortcomings suggest bias or worse. The results in Soto Becerra et al should not have been considered by the NIH COVID-19 Treatment Guidelines Panel as it formed its recommendation on ivermectin.

In an earlier article, the process followed by the COVID-19 Treatment Guidelines Panel to reach its recommendation decision on ivermectin was shown to be deceptive. The Panel was unable to confirm that a vote was held to endorse its recommendation on ivermectin. There are now serious questions about the recommendation itself.