Optimizing the Implementation of Clinical Predictive Models to Minimize National Costs: A Sepsis Case Study (Preprint)

Abstract

BACKGROUND Recent advancements in machine learning (ML) and the proliferation of healthcare data have led to widespread excitement about using these technologies to improve care. Predictive analytic models in domains such as sepsis, acute kidney injury, respiratory failure, and general deterioration have been proposed to improve the timely administration of life-saving treatments and mitigate expensive downstream complications. It has been argued that a more tailored approach that accounts for implementation constraints that may differ across care settings can further enhance the adoption of such systems. OBJECTIVE To optimize the parameters of a sepsis prediction model within distinct patient groups to minimize the excess cost of sepsis care and analyze the potential effect of factors contributing to end-user response to sepsis alerts on overall model utility. METHODS We calculated the excess costs of sepsis to the Center for Medicare and Medicaid Services (CMS) by comparing patients with and without a secondary sepsis diagnosis but with the same primary diagnosis and baseline comorbidities. We optimized the parameters of a sepsis prediction algorithm across different diagnostic categories to minimize these excess costs. At the optima, we evaluated diagnostic odds ratios and analyzed the impact of compliance factors—like non-compliance, treatment efficacy, and tolerance for false alarms—on the net benefit of triggering sepsis alerts. RESULTS Compliance factors significantly contributed to the net benefit of triggering a sepsis alert. However, a customized deployment policy can achieve a significantly higher diagnostic odds ratio and reduced costs of sepsis care. Implementing our optimization routine with powerful predictive models could result in $4.6 billion in excess cost savings for CMS. CONCLUSIONS We provide a framework that CMS policymakers could use to recommend minimum adherence rates to the early recognition and appropriate care of sepsis that is sensitive to hospital department-level incidence rates and national excess costs. Customizing the implementation of clinical predictive models by accounting for various behavioral and economic factors may improve the practical benefit of predictive models.