Missing Data in Non-Inferiority Clinical Trials

Abstract

In drug, device and behavioral clinical trials, patient withdrawal, loss-to-follow-up, and non-compliance with treatment protocols complicate analysis. When the data planned for collection are compromised or incomplete, estimates for treatment effect may be biased and trial conclusions may not be generalizable. Non-inferiority trials aim to show that an experimental treatment is therapeutically no worse than existing treatments. If a new treatment may be preferred for reasons such as lower cost, convenience, or improved safety profile, a non-inferiority design may be ideal for investigating whether the treatment is as efficacious as an active control within some pre-defined margin. Non-inferiority trials are by nature less conservative than superiority and placebo-controlled studies, and many of the challenges in their analysis and interpretation are exacerbated by missing data. Although missing data problems have been extensively studied, there has been little research on best practices for their handling in non-inferiority hypothesis testing to ensure control of type I error. I present a systematic review of non-inferiority trials to demonstrate the prevalence of missing data and understand current practices. Next, I conduct simulation studies to characterize the effects of missing data in non-inferiority trials. Using information from my systematic review of non-inferiority trials, I select methods that are commonly used as well as state-of-the-art, statistically-based methods under various missing data mechanisms. Finally, I develop a sensitivity analysis using a pattern-mixture model approach for multiple imputation adapted for the non-inferiority design. These are necessary for examining how sensitive the trial conclusions are to assumptions about missing data. Given the increasing popularity of the non-inferiority design, the persistent challenge of missing data and patient compliance, and the reliance of regulators and clinicians on trial results, there is a critical need to improve rigor and reproducibility of non-inferiority analyses. Better practices have potential for patients' easier access to new treatments and for minimizing risk of exposure to treatments that are ineffective.