Multi-agent system for recruiting patients for clinical trials

Abstract

Clinical trials are widely adopted for the purpose of evaluating medical research. In particular, they are used to study various aspects of medical science, as well as being a vital stage in the deployment of new drug treatments. However, a review of the UK Medical Research Council found that only 31% of trials actually recruited to their planned target, with 30–40% of costs arising during the recruitment phase alone. This is mainly due to the challenges that are involved in designing the clinical trial, and recruiting the required number of patients for this trial within a certain time-frame. Both tasks create significant overhead as they are slow and costly. In response, we propose a multi-agent architecture that helps ease the process of recruiting patients for clinical trials. This paper presents a results from a deployment of the architecture, showing that it succeeds in recruiting a sufficient number of patients for multiple clinical trials. The results also show that recruitment is better for some trials than for others, due to the differing trial requirements and recruitment processes