Utilizing Clinical Metadata to Predict High-Cost Complications and Treatment Response in IBD: Development of Clinical Decision Support Tools

Abstract

IBD is a costly and debilitating disease, significantly affecting quality of life. Our research plans is to generate easy to use, internet based tools (similar to a calculator) to determine which patient will go on to have costly disease over the next several years, and/or is unlikely to respond to traditional biologic therapies with anti-TNF medications. We propose using an already available IBD patient registry database which has been developed by the P.I. and the research team at UPMC/University of Pittsburgh. The short term goal is to use accessible patient information and routinely collected prospective clinical data derived from the electronic medical record from over 3,000 IBD patients followed for >7years, to generate personalized prediction models and tools to assess response to biologic therapy and risk of high costs complications, including enteric infection and disability for the care of patients with IBD. We will generate a publically accessible computer based risk prediction calculator that allows for risk stratification after entering routinely collected patient information. The goal of this web-based technology will be to use routine clinical information to facilitate a personalized clinical approach for treatment and stratification of IBD patients based on severity and phenotype. Personalized approaches for IBD treatment will help to avoid unnecessary exposure to biologic therapies and their associated risks in patients likely to fail a standard biologic treatment (i.e. anti-TNF) approach. Similarly, identifying patients that are at risk for future high-cost complications will provide a window of opportunity for cost-saving outpatient care, proactive lifestyle modifications and dietary interventions to prevent hospitalization, surgery, infectious complications, or disability. This personalized approach to IBD treatment will positively impact patients and their experience with disease, avoiding risks and given the opportunity for early interventions

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 2018
Accession Number
AD1074293

Entities

People

  • Claudia R. Rivers
  • David G. Binion

Organizations

  • University of Pittsburgh

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Bayesian Networks
  • Biological Therapy
  • Biomedical Research
  • Colitis
  • Decision Support Systems
  • Dimensionality Reduction
  • Diseases And Disorders
  • Employment
  • Health Services
  • Institutional Review Board
  • Machine Learning
  • Medical Personnel
  • Patient Care
  • Probabilistic Models
  • Professional Development
  • Public Health
  • Quality Of Life
  • Standards
  • Students
  • Therapy
  • Universities
  • User Interface
  • User Interface Engineering

Fields of Study

  • Medicine

Readers

  • Database Systems and Applications
  • Gulf War Illness and Chronic Multisymptom Illness in Veterans.

Technology Areas

  • Microelectronics