Assessment of channeling bias among initiators of glucose-lowering drugs: A UK cohort study

Abstract

BackgroundChanneling bias may occur when a newly marketed drug and an established drug, despite similar indications, are prescribed to patients with different prognostic characteristics (ie, confounding).AimTo investigate channeling bias and its impact on relative effectiveness of glucagon-like peptide-1 (GLP-1) analogs versus basal insulin and dipeptidyl peptidase-4 inhibitors (DPP-4i) versus sulfonylurea.MethodsIn the UK Clinical Practice Research Datalink, patients with type 2 diabetes initiating treatment between 2006 and 2015 were included. Analyses were stratified by years since first prescription of GLP-1 and DPP-4i, respectively. The characteristics of GLP-1 versus insulin and DPP-4i versus sulfonylurea initiators were compared over time. After propensity score matching, the relative effectiveness regarding 6-month changes in glycated hemoglobin (HbA1c) and body weight was estimated.ResultsIn total, 8,398 GLP-1, 14,807 insulin, 24,481 DPP-4i, and 33,505 sulfonylurea initiators were identified. No major channeling was observed. Considerable overlap in distributions of characteristics allowed for propensity score-matched analyses. Relative effectiveness was similar across time. The overall relative effect of GLP-1 versus insulin showed no difference for HbA1c and relative increase in body weight (3.57 kg [95% confidence interval {CI}: 3.21, 3.92]) for insulin. The overall relative effect of DPP-4i versus sulfonylurea showed relative decrease in HbA1c (−0.34% [95% CI: −0.38, −0.30]) and increase in body weight (1.58 kg [95% CI: 1.38, 1.78]) for sulfonylurea.ConclusionNo major channeling was identified in the investigated glucose-lowering drugs. Relative effectiveness could be estimated already in the first year after launch and was consistent in the years thereafter.

A pragmatic randomized clinical trial of insulin glargine 300 U/mL vs first‐generation basal insulin analogues in insulin‐naïve adults with type 2 diabetes: 6‐month outcomes of the ACHIEVE Control study

“…SOC‐BI‐treated adults could be assigned to Gla‐100 or IDet at the investigator's discretion. While this reflects clinical practice, it could have introduced channelling bias 28 . Furthermore, grouping of SOC‐BI and the lack of a titration algorithm might have introduced additional bias.…”

Use of GLP‐1 receptor agonists and subsequent risk of alcohol‐related events. A nationwide register‐based cohort and self‐controlled case series study

“…In Denmark, the first GLP‐1 receptor agonists and DPP4 inhibitors were approved for antidiabetic treatment in 2006 and 2007, respectively, but the number of users was not stable until 2009 (supporting information Table S1). Consequently, we did not include those who had initiated treatment before 2009 as they might represent a selected group of patients 22 . The date of first prescription of GLP‐1 receptor agonists or DPP‐4 inhibitors was used as the index date and we excluded the following: (1) individuals with prescriptions of both GLP‐1 receptor agonists and DPP‐4 inhibitors on the index date ( n = 14), (2) individuals with prescriptions for GLP‐1 receptor agonists or DPP‐4 inhibitors within 365 days before the index date ( n = 3525) and (3) individuals who immigrated or died before the index date or age< 15 years ( n = 3556).…”

Duration of treatment among patients prescribed afatinib or erlotinib as first-line therapy for EGFR mutation-positive non-small-cell lung cancer in the USA

“…Patients identified in 2013 (from July to December) were excluded from the primary analysis to reduce channelling bias arising from selecting patients in the same year that afatinib was approved (July 2013) by the US FDA. This approach to minimizing channelling bias has often been used previously [39][40][41].…”