Approaches in Analyzing Predictors of Trial Failure: A Scoping Review and Meta-epidemiological study

Although there are numerous studies exploring predictors of clinical trial failure, there is a lack of structured knowledge of the methodological nuances of published studies in this field. We performed a scoping review with the aim of exploring the methodological approaches in analyzing predictors of clinical trial failure. The Ovid Medline and Embase databases were systematically searched from inception to December 13, 2024, for studies employing frequentist statistics or machine learning (ML) approaches to assess predictors of trial failure across multiple clinical trials. A generalized linear model (GLM) was employed to assess the impact of methodological variations on reported failure proportions. To estimate the effects of the predictors included in the model on failure proportions, odds ratios (OR) with 95% confidence interval (95% CI) were calculated from model coefficients. The literature search identified 17,961 records, 81 of which were included in the review. Most of the studies used Clinicaltrials.gov data (73 studies, 90.1%). Frequentist statistics were used to analyze predictors of trial failure in 73 studies (90.1%), and remaining 8 studies employed ML techniques (9.9%). The GLM demonstrated that methodological factors explain 27.5% of the observed variability in failure proportions. Studies including both completed and ongoing status when calculating failure proportion had lower odds of failure compared to those just including completed status (OR = 0.44, 95% CI: 0.29–0.67, p < 0.001). There has been a recent expansion of ML approaches, potentially signaling the beginning of a paradigm shift. Methodological variations account for a significant amount of variation in failure proportion, signaling the need for adoption of standardized definitions of failure and calculation approach. What are the methodological specificities of studies exploring predictors of clinical trial failure? The choice of denominator and of included study type significantly influenced failure proportions. The use of machine learning to assess predictors of clinical trial failure is an emerging approach. There is a need for adoption of standardized definitions of trial failure and non- failure to have meaningful comparisons.