A methodology for determining patients' eligibility for clinical trials

Area of Science:

• Medical Informatics
• Clinical Research Management
• Health Data Science

Background:

• Determining patient eligibility for clinical trials is complex and data-intensive.
• Current methods often rely on manual review of extensive patient data.
• Efficiently identifying eligible patients is crucial for clinical trial success.

Purpose of the Study:

• To present a computational model for automating patient eligibility determination for clinical trials.
• To explore how computer systems can aid clinical researchers in this process.
• To investigate the application of these methods to electronic patient records for guideline adherence.

Main Methods:

• Development of a model for eligibility determination.
• Comparison of qualitative and probabilistic approaches to assessing patient eligibility.
• Synthesis of approaches to leverage the strengths of each method.
• Application of the model to a database of HIV-positive patient cases.

Main Results:

• The proposed computer system can assist clinical researchers in eligibility determination.
• Qualitative and probabilistic methods were evaluated for computing eligibility status.
• A synthesized approach combining both methods was proposed.
• Application to HIV patient data demonstrated potential for increased trial accrual.

Conclusions:

• Computer-assisted eligibility determination can significantly increase patient accrual rates in clinical trials.
• The developed methods offer a scalable solution for complex data-intensive tasks.
• These techniques are adaptable for assessing practice guideline applicability in electronic health records.