You are here

Pathways to Diagnosis (P2D): A Comparative Effectiveness Trial

The objective of this project is to identify diagnostic pathways (algorithms) in large healthcare administrative databases that predict specific or preliminary diagnoses that would lead to targeted screening and referral, and the prospectively apply these algorithms within health systems to accelerate accurate diagnoses. The overarching goal is to change the healthcare delivery model to lead to earlier diagnoses and intervention, improve clinical outcomes, demonstrate these methodologies within PCORnet, to then apply this approach to other diseases. 

Specific Aims are:

  1. Identify diagnostic algorithms for 2 autoimmune diseases (vasculitis and spondyloarthritis) in large healthcare databases using predictive analytis and machine learning techniques to calculate the risk of having the specific autoimmune disease
  2. Test the comparative effectiveness of applying the predictive analytics models developed in Aim 1 within learning healthcare systems to calculate patients' predicted probability of having the condition of interest, notify the appropriate physician, enact change in evaluation and care, and improve patients' diagnostic journeys.

 

Other Information: 

  • Principal Investigator: Mei Liu, PhD; University of Kansas Medical Center
  • Participating GPC Sites: 
    • University of Kansas Medical Center
    • University of Nebraska Medical Center
  • Project Period: May 1, 2020 - April 30, 2021

 


Back to list of initiatives