Assessing channelling bias and effectiveness in newly marketed glucose-lowering medications - a retrospective cohort study

Study type
Protocol
Date of Approval
Study reference ID
15_164
Lay Summary

Patients who receive a newly marketed drug may differ from patients who receive established drugs in regards to e.g. demographic and baseline characteristics. This selective differential prescribing may change over time when the new drug becomes established. If so, this will impact the feasibility of assessing the effects of the drug. However, very few studies have examined this problem. Our study will examine anti-diabetic drugs to treat type 2 diabetics. This project will establish how patient groups who receive newly marketed anti-diabetic drugs differ from patient groups who receive commonly used anti-diabetic drugs, how the selective use changes over time, and how the estimated effect of the drugs changes over time.

Technical Summary

Channelling bias is a potential risk when relative effectiveness is investigated in a newly marketed drug compared to an established drug in observational data. We want to investigate the potential for channelling bias in glucose-lowering medication: as injectable treatments glucagon-like peptide-1 agonist (GLP-1) will be compared with insulin; and as oral treatments dipeptidyl peptidase-4 inhibitors (DPP-4) will be compared with sulfonylurea derivatives (SU). To investigate the potential for channelling bias the changes in patient characteristics over time in the users of the newly marketed drug will be compared to the users of the established drug. This will be done for patient characteristics separately as well as using a propensity score as a summary of patient characteristics. The magnitude of channelling bias will be investigated in both an intention-to-treat cohort and in a per-protocol cohort, where a comparison will be made between the new drugs and the established drugs, for oral and injectable treatments separately.

Collaborators

Olaf Klungel - Chief Investigator - Utrecht University
Mikkel Zollner Ankarfeldt - Corresponding Applicant - Novo Nordisk A/S
Brian Thorsted - Collaborator - Novo Nordisk A/S
Johan Erpur Adalsteinsson - Collaborator - Novo Nordisk A/S
Rolf H.H. Groenwold - Collaborator - University Medical Centre Utrecht
Sanni Ali - Collaborator - London School of Hygiene & Tropical Medicine ( LSHTM )
Vinay Mehta - Collaborator - GSK