Glucose-lowering medications and the risk of cancer: A methodological review of studies based on real-world data.

The prevalence of well-known design and analysis flaws that may lead to biased results remains high among observational studies on glucose-lowering medications and cancer, limiting the conclusions that can be drawn from these studies. Avoiding known pitfalls could substantially improve the quality and validity of real-world evidence in this field.