Derivation of a Unique, Algorithm-Based Approach to Cancer Patient Navigator Workload Management.

Area of Science:

• Oncology
• Healthcare Management
• Health Informatics

Background:

• Cancer patient navigators (CPNs) are crucial for reducing diagnosis-to-treatment times.
• Current random patient distribution leads to variable workloads, potentially causing CPN burnout and suboptimal navigation.
• No automated algorithms for patient distribution to CPNs have been previously reported.

Purpose of the Study:

• To develop an automated algorithm for equitable patient distribution among CPNs specializing in the same cancer types.
• To assess the performance of this algorithm using simulation on retrospective data.

Main Methods:

• A 3-year dataset was used to identify a workload proxy and develop predictive models.
• An XGBoost-based model was selected for its superior performance in predicting weekly workload.
• A distribution model was created to assign new patients based on predicted workload, comparing predictor-informed to random distribution.

Main Results:

• The predictor-informed distribution model significantly improved workload equalization among CPNs within specialties compared to random distribution.
• Workload unfairness was demonstrably reduced using the automated algorithm.

Conclusions:

• An automated model for patient distribution is feasible and more equitable than random assignment.
• Improved workload management can potentially mitigate CPN burnout and enhance navigation services for cancer patients.