Deduce

Data drivEn Decision sUpport for Crohn’s disEase

Objective: To develop a decision support system (DSS) for patients with Crohn’s disease (CD) to help them decide between treatment options considering (i) predicted health outcomes based on Real World Data and patient’s characteristics; and, (ii) patient’s preferences. We will focus on two critical points in the treatment pathway of CD: (i) after failing thiopurines and (ii) patients who have failed Tumour Necrosis Factor (TNF) inhibitors.

Background: Crohn's disease is a chronic inflammatory bowel disease with a heterogeneous presentation and treatment response diagnosed mostly in young people. CD has an important impact on people’s quality of life, due to debilitating symptoms, complications, surgery and hospitalization. None of the available treatments (medical or surgery) are effective for all patients and the available drugs are associated with potential severe side effects.
Clinical DSSs provide clinicians and patients with evidence based personalized information and recommendations based on the patients’ characteristics and preferences. They are valuable tools and can now be derived from available Real World Data of this heterogeneous disease. However, currently there are no DSSs to assist CD patients and health care professionals with treatment decisions.

Methods: Semi-structured interviews will be conducted with patients, in order to identify preferences and informational needs for decision-making.
We will develop and validate clinical outcome prediction models through collaboration of expert clinicians and data-scientists by applying machine learning techniques to existing longitudinal Real World Datasets including clinical data and patient-reported outcomes (PROs). Different statistical and machine learning algorithms will be applied in order to pick the one with the best performance. A decision aid will be developed to inform patients about treatment options, while providing them with personalized health outcome predictions based on the new models. The DSS will be integrated in the telemedicine tool myIBDcoach and prospectively validated during a pilot programme in 14 clinical institutions in The Netherlands in in order to assess its usability, acceptance and clinical impact.

Anticipated results: A successful DSS will result in better-informed treatment decisions, increased patient satisfaction and improved health outcomes.

This is a grant from Janssen-Cilag B.V.

The Project Lead

The Team

Partners

  • Janssen-Cilag B.V., Breda, The Netherlands
  • Maastricht University, Maastricht, The Netherlands
  • Stichting Maastricht Radiation Oncology (Maastro), Maastricht, The Netherlands

Publications