23rd Conference of the International Federation of Operational Research Societies
Abstract Submission

744. Data Science for Medical Programming in a Network of Public Hospitals

Invited abstract in session TB-21: Healthcare Analytics, cluster Healthcare Management.

Tuesday, 10:30-12:00
Room: FENH201

Authors (first author is the speaker)

1. René Lagos
Digital Health, South East Metropolitan Health Service

Abstract

Introduction
Public hospitals in Chile face high demand and need to align medical resources to populations’ healthcare needs. Medical programming (or medical master scheduling) is a crucial process for this purpose, but data gathering and processing with spreadsheets makes it complex and time consuming for hospitals, reducing its utility. The South East Metropolitan Health Service (SEMHS) developed a system to optimize medical programming and to support waiting list reduction strategies.

Methodology
A redesign of the process was conducted together with hospitals and SEMHS managers. A web platform to support the process was designed and developed using an agile project management framework. Prototypes were developed to apply data science and operations research tools to explore strategies to reduce waiting lists.

Results
The Medical Programming Platform has five modules: staffing, clinical units programming, production forecasting, demand trends and physical resources. It was the first medical programming platform and was implemented in 5 hospitals and 10 mental health community centers; it programmed 4700 health professionals, including 1850 physicians, on a quarterly basis. The percentage of programmed contracts increased from 75% in 2019 to 96% in 2022. An Extract, Transform, Load process supports reporting and analytics. Prototypes were developed for forecasting medical consultation demand, optimizing box allocation, outpatient waiting lists simulation and stochastic optimization of medical schedules.

Discussion
The platform replaced spreadsheets, previously consuming valuable time from hospital managers, and enabled participation of clinical managers in the programming process. Medical programs are more reliable in terms of the staffing levels and clinical activities offered, turning into a fundamental tool for managers. The implementation of the platform constitutes a digitalization of public health as it provides value, based on a data centered process to understand populations’ healthcare needs and the alternatives to align medical services. Visual analytics and OR prototypes have demonstrated the potential of medical programming as a tool for public health data science, setting a new standard for public hospitals in Chile.

Conclusion
Operations Research and public health data science provide a solid framework to advance medical programming in Chilean public hospitals and set its technical and ethical standards.

Keywords

Status: accepted


Back to the list of papers