Bykov K, et al. - Researchers reviewed the methodology of observational studies that explored the connection between glucose-lowering therapies and cancer occurrence, identifying the most common challenges and sources of bias, and assessed whether key methodological choices changed over time. Of the 155 studies assessed, only 26% implemented a new user design, 41% used an active comparator, 33% used a lag or latency period, and 51% adjusted for the duration of diabetes. Apart from a decreasing trend in adjustment for variables measured during the follow-up, the authors did not observe any trends in methodological choices over time. Overall, the prevalence of well-known design and analysis flaws that may result in biased outcomes remains high among glucose-lowering drug and cancer observational studies, limiting the conclusions that can be drawn from these studies. Avoiding known pitfalls could improve the quality and validity of real-world evidence in this area substantially, according to the researchers.