Back

Assessing the Impact of the Application of PMSI Data on Enrollment Efficiency, Retention, and Therapeutic Area Performance in Clinical Trials

Nama Ravi, S. K.; WALI, A.; Gordejuela, R.; Collins, B.; Llamas, E.; Lampure, A.; Sommerford, N.

2025-04-20 public and global health
10.1101/2025.04.18.25326069 medRxiv
Show abstract

ObjectiveThis study evaluates whether the use of Programme de Medicalisation des Systemes dInformation (PMSI) data in France positively impacts clinical trial performance or not. Specifically, we compared enrollment performance and dropout rates across therapy areas between PMSI-supported and non-PMSI supported clinical trials. MethodsA retrospective analysis was conducted using enrollment and dropout data for clinical trials conducted between Jan 2019 and Oct 2024, across multiple therapy areas. Comparative statistics and visualizations were used to assess the difference in enrollment performance and dropout rates between PMSI-supported and non-PMSI supported trials. ResultsPMSI supported trials demonstrated higher median enrollment rates and fewer outliers across several therapy areas, particularly Infectious Diseases, Gastrointestinal and Dermatology. For Cardiovascular and Oncology trials, the median enrollment rates for PMSI supported trials were slightly higher than the non-PMSI supported trials, indicating a smaller impact on trials in these therapy areas. Specifically, PMSI-supported studies showed enrollment rates that were between 5% and 238% higher than non-PMSI-supported studies, depending on the therapy area. Also, the dropout rates were also lower in PMSI supported trials. ConclusionThe application of PMSI data appears to be associated with improved trial performance in terms of enrollment efficiency, fewer outliers, and consistent trial execution particularly in select therapy areas, namely Infectious Diseases, Gastrointestinal and Dermatology. These findings support the strategic use of PMSI data to enhance trial success and patient recruitment in France.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.