Associating Healthcare Teamwork with Patient Outcomes for Predictive Analysis

TL;DR

This paper tackles how healthcare team collaboration, as captured in EHR usage traces, relates to cancer patient survival, addressing gaps where team factors are understudied. It proposes an end-to-end workflow that constructs directed bipartite collaboration networks from EHR data and leverages a GraphSAGE-based GNN with explainable AI to predict survival while guarding against data leakage. The study provides evidence that HCP collaboration patterns—especially GP involvement—offer predictive signals and clinically plausible explanations, validated by domain experts. The framework is transferrable to other complex care settings and supports data-informed interventions to improve team-based delivery.

Abstract

Cancer treatment outcomes are influenced not only by clinical and demographic factors but also by the collaboration of healthcare teams. However, prior work has largely overlooked the potential role of human collaboration in shaping patient survival. This paper presents an applied AI approach to uncovering the impact of healthcare professionals' (HCPs) collaboration-captured through electronic health record (EHR) systems-on cancer patient outcomes. We model EHR-mediated HCP interactions as networks and apply machine learning techniques to detect predictive signals of patient survival embedded in these collaborations. Our models are cross validated to ensure generalizability, and we explain the predictions by identifying key network traits associated with improved outcomes. Importantly, clinical experts and literature validate the relevance of the identified crucial collaboration traits, reinforcing their potential for real-world applications. This work contributes to a practical workflow for leveraging digital traces of collaboration and AI to assess and improve team-based healthcare. The approach is potentially transferable to other domains involving complex collaboration and offers actionable insights to support data-informed interventions in healthcare delivery.

Figures (5)

• Figure 1: The key milestones in the data collection and processing timeframe are as follows. We divide collected data at the nine-month post-diagnosis mark into an observation window for training and a gap window for preventing data leakage.
• Figure 2: The illustration of the attributed bipartite collaboration network.
• Figure 3: Our prediction model uses GraphSAGE layers to aggregate information from neighboring nodes and generate graph embeddings. A fully connected layer is then applied to these embeddings to predict patient survival.
• Figure 4: The illustration depicts the simplification of the collaboration network from a bipartite structure to an all-HCP collaboration network. The same edge-rerouting procedure is also applied when simplifying from the bipartite structure to the all-notes network.
• Figure 5: The comparison between two breast cancer patients with different outcomes.