Meta-Analysis Based on ‘GRADE’ Method: Low-to-Moderate Evidence Ivermectin Can Help Treat COVID-19

Meta-Analysis Based on ‘GRADE’ Method Low-to-Moderate Evidence Ivermectin Can Help Treat COVID-19

The ivermectin debate intensifies around the globe as an overwhelming majority of the 40+ studies now worldwide demonstrate positive evidence, yet one recent study yielding no statistically significant results in Colombia certainly got the attention of CNN and the New York Times, both seemingly pouncing on the declaration that the economical, generic drug just doesn’t work according to the study results. Several experts have pointed out the study’s flaws. But why didn’t these news agencies at least follow the National Institute of Health (NIH) lead and mention some of the other studies as they have here? Bias could be a factor, as generally a confluence of societal elements evidence similar behavior. Recently, a group of researchers out of the United Kingdom also may have experienced a form of bias, as they were on track for acceptance with Lancet Respiratory Medicine having only to be told last minute they didn’t have sufficient data. Led by corresponding authors Andrew Bryant and Theresa A. Lawrie, the group of authors assessed whether the repurposed medicine could possibly have a role in helping treat patients infected with SARS-CoV-2, the virus behind COVID-19.  While ivermectin is known to have anti-viral and anti-inflammatory properties, the group sought to conduct a meta-analysis of a number of clinical trials to better understand the level of evidence. Utilizing the “GRADE” approach, they evaluated a total of 21 randomized controlled trials involving 2,741 participants which met the team’s review inclusion criteria. They also noted 13 trials that led to the conclusion that ivermectin may reduce the risk of mortality as compared to no ivermectin. They opined that there is low-to-moderate evidence favoring ivermectin as a treatment for COVID-19, meaning the probabilities are sufficient to consider for emergency use authorization, from one point of view. Critics of the underlying studies point out they are not necessarily appropriately designed, making the quality of the data suspect. The majority of the underlying clinical trials were sponsored by academic medical centers in low-to middle-income countries (LMICs).

What is Certainty of Evidence?

Those research scientists involved with the pursuit of evidence-based medicine make their findings most often on established measures of uncertainty. This is important in the ivermectin debate, as the number of studies now is significant with seemingly polar oppositional views as to the evidentiary weight inherent in the findings. 

Ivermectin is approved by the FDA for certain indications involving parasite-born conditions, but not for COVID-19. In regards to evidentiary weight of ivermectin as possible treatment for at least mild to moderate early onset COVID-19, on the one side are groups such as the Front Line COVID-19 Critical Care Alliance (FLCCC) that support emergency use authorization for ivermectin now, while on the other side are most apex research agencies such as the National Institutes of Health (NIH) which declares the need for more data, but doesn’t recommend for or against, at least according to their last declaration. 

But what is “certainty” of evidence anyway? A critically important concept, this is how therapies can be accepted for use. Often the result of recommendations from a clinical practice guideline or based on some methodical literature review, medical practitioners need to understand the trustworthiness of evidence based on a balance of benefits and risks or harms of a particular therapy.  

Can the process of deriving evidence become politicized by pecuniary interests? Possibly. As parties look at ivermectin for example, they must derive the certainty or quality of evidence. A standard method to accomplish this, the GRADE (Grading of Recommendation, Assessment, Development and Evaluation) approach represents a modern framework for rating the certainty of evidence. 

In the GRADE process, investigators use both randomized controlled trials and observational studies to identify, analyze, and assess high and low certainty evidence. This is where it gets really tricky (and potentially where the influence of subjective perspective can surface), as these grades are based on study design involving any number of critical parameters such as methodical limitations of a particular study, indirectness of evidence, imprecision of estimates, inconsistency of evidence, and likelihood of publication bias. 

Looking at ivermectin, overall evidence can be either downgraded or upgraded to the high, moderate, low, or very low rating depending on these factors. In the case of “high certainty” situations, investigators become quite confident that they have found a treatment that works—that is, the effect they have detected across studies represents in  fact true effect while “very low” equates to very little investigator confidence in such evidence. 

The Study

This UK-based team utilized the GRADE approach to perform a meta-analysis to assess the evidence associated with ivermectin as a treatment for COVID-19. Based on 21 randomized controlled trials involving 2,741 participants, the team also performed meta-analysis of 13 trials and found that ivermectin reduced the risk of death compared with no ivermectin (average Risk Ratio 0.32, 95% confidence interval (CI) 0.14 to 0.72; n=1892; I2=57%; low to moderate-certainty evidence. Low certainty evidence found ivermectin prophylaxis reduced covid-19 infection by an average 86% (95% CI 79% to 91%). They reported that for secondary outcomes they found very-low or low certainty evidence, suggesting that there may be no benefit with ivermectin for “need for mechanical ventilation,” although effect estimates for “improvement” and “deterioration” indicated potential benefits from the drug. The team declared that severe adverse events were rare and evidence of no difference was deemed low to very low certainty.  


The UK-based team of authors concluded that based on the meta-analysis of the underlying clinical trials the evidence for ivermectin using the GRADE method equals “low to moderate—certainty” evidence pointing to the conclusion, at least from the vantage of these authors, that ivermectin likely reduces COVID-19 mortality and that utilizing this drug early on may reduce the number COVID-19 patients progressing to a more severe stage. They conclude that based on their assessment of safety and the drug’s low cost, it could have considerable impact on the pandemic triggered by SARS-CoV-2.

Lead Research/Investigators

Andrew Bryant, Population Health Sciences Institute, Newcastle University, Newcastle upon Tyne, UK (Corresponding Author)

Theresa A. Lawrie, Evidence based Medicine Consultancy, Bath, UK.

Therese Dowswell,  Evidence-based Medicine Consultancy, Bath, UK

Edmund Fordham, Evidence-based Medicine Consultancy, Bath, UK

Scott Mitchell, Emergency Department, Princess Elizabeth Hospital, Guernsey, UK

Sarah R. Hill, Population Health Sciences Institute, Newcastle University, Newcastle upon Tyne, UK

Tony C. Tham, Ulster Hospital, Dundonald, Belfast, Northern Ireland, UK

Call to Action: Follow the link to the source above to read the study.


  1. Dear Trial Site News,
    Thank you for your timely review of our recent meta-analysis. We thought your readers may be interested to know that, whilst our paper is available on the pre-print server OSF, it has been fully peer reviewed. The peer review was conducted by four independent, expert peer-reviewers appointed by a high impact-factor journal to which we submitted the review on 11th February 2021. We addressed the peer review comments to the satisfaction of all four peer reviewers and the preprint now available is the outcome of this peer-review process. Despite having satisfied the peer reviewers, we received a rejection letter from the Senior Editor of the journal, who also wrote: “we don’t doubt that this is an important paper”.
    Yours sincerely, Andrew Bryant and Tess Lawrie (joint lead authors)