In an accompanying editorial, authors Pascal Meyre, from the Cardiovascular Research Institute at the Basel University Hospital in Switzerland, and David Conen, from the Population Health Research Institute, McMaster University, Canada, offer a critical look at the findings.
They agree that the strengths of the study "are the large sample size, with over 160,000 individuals included in the study, the large number of outcome events, and the long follow-up duration." "This allowed the investigators to carry out meaningful analyses and adjust the multivariable models for many covariates, such that some of the confounding could be controlled," they add.
However, the retrospective design of the study "may have introduced selection bias," say the authors of the editorial. Furthermore, the participants' "level of education, marital status, and information on inflammatory biomarkers, such as C-reactive protein, were not available."
Read More...