JMIR Publications

ABSTRACT

Background:

Evaluation of artificial intelligence tools in clinical trials remains the gold standard for translation into clinical settings, yet design factors associated with successful trial completion and common reasons for trial failure are unknown.

Objective:

To compare trial design factors of complete and incomplete clinical trials testing artificial intelligence tools, we conducted a cross-sectional study of complete (n=485) and incomplete (n=51) clinical trials that evaluated AI as an intervention of ClinicalTrials.gov.

Methods:

Trial design factors, including area of clinical application, intended use population, and intended role of AI, were extracted. Trials that did not evaluate AI as an intervention and active trials were excluded. Assessed trial design factors included area of clinical application, intended use population, role of AI, study type, allocation, intervention model, masking, age, sex, funder, continent, length of time, sample size, number of enrollment sites, and study start year. Main outcome was completion of clinical trial. Odds ratios and 95% confidence intervals were calculated for all trial design factors using propensity-matched, multivariable logistic regression.

Results:

Our nested propensity-matched case-control results suggest that trials conducted in Europe were significantly associated with trial completion when compared to North American trials (OR, 2.848; 95% CI, 1.141-7.113; p=0.025) and trial sample size was positively associated with trial completion (OR, 1.001192; 95% CI, 1.000202-1.00218; p=0.0183).

Conclusions:

In this cross-sectional study, trial design factors positively associated with trial completion include trials conducted in Europe and sample size. Given the promising clinical utility of artificial intelligence tools in healthcare, future translational research can prioritize addressing the design factors of AI clinical trials associated with trial incompletion and common reasons for study failure.