Table 1

Methodological considerations for future observational studies assessing the association between incretin-based drugs and adverse pancreatic events

• Study population should be limited to patients with type 2 diabetes
• Study population should be limited to new users of antidiabetes drugs
• Incretin-based drugs should be compared with active comparators for which there is clinical equipoise (i.e., other second- or third-line therapies)
• For pancreatic cancer, exposure should be lagged for latency considerations and to minimize protopathic bias
• Analyses should control for diabetes-related variables, such as duration of diabetes, hemoglobin A1c, and diabetes-related complications
• Appropriate study design and analytic approaches should be used to avoid time-related biases (18)
• Given the rarity of the outcomes, studies need large sample sizes and sufficient durations of follow-up