Comparing different representativeness measures in clinical trial design

‍

‍

Motivation

To ensure that new treatments are safe and effective for everyone (generalizable), it’s crucial to consider the diversity of participants in clinical trials, and to do so at the beginning of the study design process. One major influence on representativeness is the trial’s eligibility criteria. However, each criterion’s impact on representativeness is unique to the study, and surprisingly complex: depending on other existing criteria, the indication population, and the representativeness dimension being assessed (such as sex, socioeconomic status, ethnicity, and other protected and historically underrepresented demographic groups).

‍

Solution

To help understand this complex variation, Plinth developed an analytic pipeline that simulates changes to eligibility criteria for clinical trial designs and calculates a suite of representativeness metrics between the resulting eligible population and the target population. To facilitate efficient insight generation, Plinth utilized it’s interactive research portal (Folio) to allow scientists to visualize and compare representativeness metrics across a wide array of clinical trial designs using real-world data.

‍

Impact

Plinth’s solution empowers investigators studying the impact of eligibility criteria on clinical trial inclusiveness to efficiently draw conclusions across representativeness metrics, studies, historically underrepresented subgroups, and eligibility criteria. These conclusions offer valuable guidance for clinical trial designers seeking to gain a fuller picture of treatment effects within diverse populations and ensure compliance with regulatory requirements on representativeness. Future work will focus on how inclusion criteria impact outcome variance and recruitment rates.

‍