Defining Provider Prescribing Preference as an Instrumental Variable

This case study evaluates the use of provider prescribing preference (PPP) as an instrumental variable (IV) in assessing the effectiveness of non-vitamin K antagonist oral anticoagulants (NOACs) for stroke prevention in patients with non-valvular atrial fibrillation (NVAF). The study aims to determine the strength and validity of the IV by analyzing different definitions of PPP for rivaroxaban compared to apixaban. Data were sourced from the OM1 Real World Data Cloud, encompassing a cohort of NVAF patients linked to provider datasets. The analysis reveals that the definition of PPP significantly impacts instrument strength, with varying patient counts leading to different associations with treatment choices. The study identifies a total of 182,071 patients and 47,758 providers, demonstrating that instrument strength improves with a larger patient count in defining PPP. The findings contribute to understanding how PPP definitions affect the validity of IVs in health research.