SHM Converge Scientific Abstract ePoster Gallery-Development of a Novel Patient Distribution Tool

Pdf Summary

This project developed an algorithm-driven, semi-automated patient distribution tool to improve geographic co-location compliance and reduce physician time spent assigning newly admitted patients to hospitalist rounding teams.

<h3>Background</h3>
Hospitalist teams at the University of Chicago manage patient placement across 12 rounding teams with different capacities. Because admissions and discharges fluctuate, manual patient distribution can be time-consuming, inconsistent, and difficult to keep aligned with co-location rules. The process may also contribute to perceptions of unfairness among providers.

<h3>Purpose</h3>
The goal was to create a daily patient sorting and assignment tool for the Night Triage hospitalist that would:
- reduce inconsistency in patient assignments
- improve compliance with co-location and team assignment rules
- decrease physician workload
- improve perceived fairness in distribution

<h3>Description</h3>
The tool used a tiered algorithm to assign patients to primary, secondary, and tertiary teams based on service, unit, and team capacity. It was programmed in Stata and launched in August 2023 as an optional tool. It generated an Excel output file after scanning patients, calculating priority scores, and assigning them in order of service, location, and available capacity.

<h3>Results</h3>
Before rollout, manual patient distribution took 30 to 180 minutes. With the tool, the tested process including manual input took about 20 minutes, suggesting a potential time savings of 10 to 150 minutes. At the 3-month review, 7 of 17 hospitalists used the tool. Among those users, 5 preferred it, 1 had no preference, and 1 preferred manual distribution.

<h3>Conclusion</h3>
The study shows that a non-clinical workflow like patient distribution can be partially automated to save physician time and improve process consistency. However, broader adoption was limited by usability barriers, access issues, and the need for manual data entry. Future work will focus on better software integration, reducing manual steps, and measuring actual co-location effectiveness.

Asset Subtitle

Khanh T. Nguyen

Meta Tag

Author List Hui Zhang, Khanh T. Nguyen

Category Innovations

Concept Patient Assignment

Concept Algorithm

Concept Hospitalist Service

Concept Co-location Rules

Concept Semi-automated Tool

Distinguished Non-Finalist

Presenter Organization University of Chicago

Presenting Author Khanh T. Nguyen

Track Technology in Hospital Medicine

Keywords

patient distribution

hospitalist rounding teams

geographic co-location

algorithm-driven tool

semi-automated workflow

physician time savings

team assignment

patient assignment algorithm

workflow automation

hospital capacity management

Patient Assignment

Algorithm

Hospitalist Service

Co-location Rules

Semi-automated Tool