Estimating the workload of a multi-disciplinary care team using patient-level encounter histories

Healthcare spending in the United States is concentrated on a small percentage of individuals, with 5% of the population accounting for 50% of annual spending. Many patients among the top 5% of spenders have complex health and social needs. Care coordination interventions, often led by a multidisciplinary team consisting of nurses, community health workers and social workers, are one strategy for addressing the challenges facing such patients. Care teams strive to improve health outcomes by forging strong relationships with clients, visiting them on a regular basis, reconciling medications, arranging primary and speciality care visits, and addressing social needs such as housing instability, unemployment and insurance. In this paper, we propose a simulation algorithm that samples longitudinal patient-level encounter histories to estimate the staffing needs for a multidisciplinary care team. Our numerical results illustrate multiple uses of the algorithm for staffing under stationary and non-stationary patient enrollment rates.