What is it about?
This editorial critically analyzes an early study sponsored by the maker of a diabetes drug that found the drug reduced cardiovascular events compared to a placebo. While this finding is encouraging, all research, especially early research, must be viewed skeptically. The most important issue in this situation was a lack of correction for blood sugar levels and selective reporting of standard error over standard deviations. Based on this early study alone, in 2017 there was inconclusive evidence for cardiovascular benefits beyond blood sugar reduction.
Featured Image
Photo by jesse orrico on Unsplash
Why is it important?
This editorial provides an important critical analysis of a major pharmaceutical industry-funded study on diabetes drugs. It highlights the need to interpret industry-sponsored research carefully and transparently, considering limitations like those raised here. The authors provide a thoughtful model for scrutinizing potential biases in clinical research. By early removal of biases, medical research can progress at a faster pace.
Perspectives
This editorial raises essential points about examining the methodology and potential conflicts of interest in clinical studies. While industry funding is common, it can introduce biases, as can selective statistical reporting. The reason for not performing a statistical correction on blood sugar levels is unknown. This editorial is from 2017 and subsequent research suggests that there may be a true benefit of these medications beyond its effects on blood sugar levels. However, coming to this conclusion would have been faster if initial research into this took into account important covariates like blood sugar levels.
Thomas F Heston MD
University of Washington
Read the Original
This page is a summary of: Canagliflozin lowers blood sugar, but does it also lower cardiovascular risk? Maybe not, Annals of Translational Medicine, December 2017, AME Publishing Company,
DOI: 10.21037/atm.2017.09.28.
You can read the full text:
Resources
Contributors
The following have contributed to this page
