Decision support for personalized chronic pain care


Chronic pain patients constitute a large and heterogeneous patient group, and there is presently no clear best option for treating chronic pain. Thus, it is important to build tools and methods that can identify the effective and efficient treatment alternative for each individual patient. Current tools for supporting patients and clinicians in shared decision making do not allow personalization.

In this project, we aim to provide personalized decision support with realistic probabilities for outcomes, side effects, and adverse events, based on information obtained from previous similar cases.


The project has two main goals:

  • Processing health service data and patient-reported outcome of treatment (e.g., perceived intensity of pain) respecting the patient’s privacy.
  • Prepare a solution that provide the chronic pain patient and his/her physician personalized decision alternatives, with evidence-based probabilities for outcomes and side effects.


To achieve the first goal, we aim at using a mobile application for collecting the data from the patients. To ensure protection of the patient’s privacy, we will use the principal of secret sharing for storing of patient data.

To achieve the second goal, the solution will allow real-time construction of decision models using electronic health record (EHR) data from multiple health institutions. For analyzing the data, the solution will exploit privacy preserving secure multi-party computation techniques. Thus, EHR data at each institution will contribute to the final result, although no data about individuals (only aggregated data) will be shared and/or transferred outside each institution.

Chronic pain patients will be involved in all the stages of the project, from the design of the applications to the test and data collection phases.