Practitioner perspectives on informing decisions in One Health sectors with predictive models

The continued emergence of challenges in human, animal, and environmental health (One Health sectors) requires public servants to make management and policy decisions about system-level ecological and sociological processes that are complex, poorly understood, and change over time. Relying on intuition, evidence, and experience for robust decision-making is challenging without a formal assimilation of these elements (a model), especially when the decision needs to consider potential impacts if an action is or is not taken. Models can provide assistance to this challenge, but effective development and use of model-based evidence in decision-making (‘model-to-decision workflow’) can be challenging. To address this gap, we examined conditions that maximize the value of model-based evidence in decision-making in One Health sectors by conducting 41 semi-structured interviews of researchers, science advisors, operational managers, and policy decision-makers with direct experience in model-to-decision workflows (‘Practitioners’) in One Health sectors. Broadly, our interview guide was structured to understand practitioner perspectives about the utility of models in health policy or management decision-making, challenges and risks with using models in this capacity, experience with using models, factors that affect trust in model-based evidence, and perspectives about conditions that lead to the most effective model-to-decision workflow. We used inductive qualitative analysis of the interview data with iterative coding to identify key themes for maximizing the value of model-based evidence in One Health applications. Our analysis describes practitioner perspectives for improved collaboration among modelers and decision-makers in public service, and priorities for increasing accessibility and value of model-based evidence in One Health decision-making. Two emergent priorities include establishing different standards for development of model-based evidence before or after decisions are made, or in real-time versus preparedness phases of emergency response, and investment in knowledge brokers with modeling expertise working in teams with decision-makers.