Active‐comparator restricted disproportionality analysis for pharmacovigilance signal detection studies of chronic disease medications: An example using sodium/glucose cotransporter 2 inhibitors

Aims Disproportionality analysis is a common pharmacovigilance tool to detect safety signals of type 2 diabetes medications from spontaneous drug reporting databases. The aim was to demonstrate the impact of using active‐comparator restricted disproportionality analysis (ACR‐DA), wherein the reference group is restricted to reports with a clinically appropriate active comparator. Methods Using reports from the Food and Drug Administration Adverse Event Reporting System, we assessed if sodium/glucose cotransporter 2 (SGLT2) inhibitors are associated with higher reporting of 5 potential adverse events: acute kidney injury, genitourinary tract infections, diabetic ketoacidosis, fractures, and amputations. For each adverse event, we calculated the proportional reporting ratio (PRR) and adjusted reporting odds ratio (aROR [95% confidence interval, CI]) using 3 types of reference groups: no SGLT2 inhibitor (background risk reference), other diabetes drugs (therapeutic class reference), and dipeptidyl peptidase 4 inhibitors (active comparator reference). Results Based on ACR‐DA, we did not detect a safety signal for acute kidney injury (PRR 0.92 [0.81–1.04]; aROR 0.78 [95% CI 0.72–0.85]) or fractures (PRR 0.44[95% CI 0.17–1.15]; aROR 0.74 [95% CI 0.61–0.91]) associated with SGLT2 inhibitors compared to dipeptidyl peptidase 4 inhibitors. However, we detected safety signals for genitourinary tract infections (PRR 2.75[2.02–3.76]; aROR 2.54[2.26–2.86], diabetic ketoacidosis (PRR 63.85[39.37–103.53; aROR 91.49[70.66–118.48]), and amputations (PRR 52.60 [19.66–140.75]; aROR 22.64 [15.32–33.42]. Conclusion The use of the proposed ACR‐DA to detect safety signals of type 2 diabetes medications may reduce false positive safety signals through careful selection of the comparator which is expected to reduce channelling bias.