In this episode, I’ll discuss assessing the risk of bias in a review of nonrandomized studies.
According to training by Cochrane, bias in a meta-analysis is a systematic error or deviation from the truth. A risk of bias assessment of studies included in a meta-analysis is essential because the meta-analysis is only as good as the studies it includes, and incorrect studies could lead to a misleading conclusion for the review.
Most reviews and meta-analyses are of randomized trials and the risk of bias assessment involves evaluating the included trials on things like random sequence generation and allocation concealment. However when the review is of non randomized studies, different metrics must be used to assess the risk of bias.
A recent example of a review that had to deal with this issue is one published in Pharmacotherapy looking at Andexanet alfa versus PCC products for factor Xa inhibitor bleeding.
This review concluded that andexanet alfa had better hemostatic efficacy and 30-day mortality risk than PCC products when reversing factor Xa inhibitors. The review included 18 comparative evaluations of andexanet alfa versus PCC, none of which were randomized.
To assess the risk of bias the authors relied on the MINORS criteria which stands for Methodological index for non-randomized studies. This criteria uses 12 items.
- Is there a clearly stated aim
- Are patients enrolled consecutively
- Is data collection prospective
- Are the endpoints appropriate
- Is the investigators assessment of the endpoint blinded
- Was the follow-up period appropriate
- Is the percentage of patients lost to follow up low
- Was study size prospectively calculated
- Was the control group adequate
- Did the control and treatment groups occur concurrently
- Are baselines between groups similar
- Were the statistical methods appropriate
The Pharmacotherapy authors included a thirteenth criteria: Are the criteria for patient inclusion into the study clear and valid?
In addition to evaluating systematic reviews, pharmacists can use these criteria when evaluating individual non randomized studies to identify study strengths and weaknesses.
Members of the Hospital Pharmacy Academy have access to trainings on biostatistics for journal club and those likely to appear on a BPS exam, in addition to hundreds of other trainings to help you in your practice. To get immediate access go to pharmacyjoe.com/academy.
If you like this post, check out my book – A Pharmacist’s Guide to Inpatient Medical Emergencies: How to respond to code blue, rapid response calls, and other medical emergencies.
Leave a Reply