Implementing Community-Based Model-Driven Decision Support to Reduce Overdose Deaths: Lessons Learned at Study Midpoint

The overdose crisis continues to be a critical public health challenge in the United States and elsewhere. While funding has been provided at the federal, state, and municipal levels, implementation of strategies to address overdose is accomplished at the county level in most states. model-driven decision support, using tools like agent-based modeling, has emerged as a promising approach for predicting outcomes, tailoring interventions, and allocating implementation resources. While there are a number of examples of model development, there has been little study of the implementation of model-driven decision support to reduce overdose. This paper describes the lessons learned at the midpoint in an implementation study on the adoption of model-driven decision support to optimize overdose mitigation strategies by three counties. The Stages of Implementation Completion ® is used within the Reach, Effectiveness, Adoption, Implementation, and Maintenance framework to describe the process of incorporating modeling into the decision-making process in three counties. Key lessons are that the process of developing, running, and discussing models with the community is of value regardless of model output, that useful models can be built with incomplete and less recent data, and that community decision makers are interested in using predictive models when they understand the inputs and assumptions that drive the output. This work highlights the potential for model-driven decision support’s incorporation into evidence-based planning in local communities and provides insight for future implementation.