In his research trajectory, MJ follows three goals. First, he conducts simulation modeling and informatics research for various population-based health policies, focusing on health outcomes and cost-effectiveness. In his modeling research—drawn on theories of optimization and strategy—he analyzes the impacts of large-scale policies for prevention, screening, and treatment. MJ has developed models for obesity, post-traumatic stress disorder, and depression. He is currently working with the FDA to develop an opioid systems model, informing opioid policies at the FDA and other government agencies. Other areas of his modeling research include drug-shortages in pharmaceutical supply chains, organizational cybersecurity in health care, and the diffusion of medical technologies.
Second, MJ focuses his research on mechanisms that connect human decision-making to health care systems, because that is where many important policy-resistant problems lie. In particular, he aims to understand how and why many health policies fail to produce lasting results or worse, create results counter to their goals.
Third, he wants his research to rigorously connect models with quantitative data. The growing complexity of health care issues, combined with the ubiquity of large amounts of data, requires increasingly sophisticated analytical methods. MJ complements his phenomenological research with methodological contributions that build bridges across methodological and application domains. For example, he has contributed to adapting various simulation-optimization approaches for model calibration and parameter estimation in dynamic models (e.g., the method of simulated moments and indirect inference), improving systematic review techniques, and developing a novel method for aggregation of prior stochastic and heterogeneous statistical findings.