Risk Modeling of Chronic Opioid Use for Physician Decision Support

Chronic opioid use (COU) has been associated with serious adverse outcomes and the social and economic impact of continuing opioid treatment is sizeable. The net effect of COU on a given patient's health – beneficial, adverse, or neutral – may be difficult to determine ex ante, and affected by many unobservable factors. Given the risks and adverse outcomes associated with COU, many consider COU a care choice of last resort. Thus, to the extent that clinicians need to make decisions regarding whether opioid use will begin or continue, decision support on the individual-patient probability of COU appears critical. Personalized guidelines, built on decision support systems (DSSs), have the potential to influence care at the point of service. Building models that can serve as the foundation of such systems can, therefore, contribute to changes in physician prescribing and program policies. As a result, we aimed to study personalized pain management-related decision support for opioid prescribing. Specifically, we applied state of the art machine learning algorithms, as well as more traditional models, to study the feasibility and potential impact of a COU DSS, using a large data set from the U.S. Army. We also investigated the economic impact of such decision support.

This work was supported by the National Institute for Health Care Management (NIHCM) through its Research Grants program.

Research Team:

Ritu Agarwal 1,2 (PI), David Anderson 1,5, Margret Bjarnadottir 1,2, Kenyon Crowley 1,2, D. Alan Nelson 1,4, Kislaya Prasad 1,2,3

1. Center for Health Information and Decision Systems (CHIDS) 2. Decision, Operations and Information Technology Department, Robert H. Smith School of Business 3. Brookings Institution 4. Innolytics LLC 5. Zicklin School of Business, Baruch College

Project Results:

Webinar (April 10, 2018)

Webinar Slides (April 10, 2018)

Webinar Video (April 10, 2018)

Research Brief (3-minute quick read)

NIHCM Research Grant Final Narrative Report:


  • Early results were presented by Dr. Margret Bjarnadottir at the IBM Behavioral Analytics Workshop on November 9, 2017.
    Early Detection and Prevention of Chronic Opioid Use

Blog/Social Media Posts: 

CHIDS Research Spotlight—Promoting Better Pain Management Outcomes:

  • Precision Decision Support for Opioid Prescription
Back to Top