Los Alamos National Laboratory
 
 
NISAC  National Infrastructure Simulation and Analysis Center




NISAC Tools: HCSim

HCSim: Agent-Based Healthcare Surge Capacity Model

Structure of an agent representing an influenza case

Structure of an agent representing an influenza case (Click graphic to enlarge)

Overview

The HCSim model is used to gauge the impact of mass-casualty incidents on hospital capacity.

Model Characteristics

  • Agent-based representation of hospitals and casualties
  • Stochastic disease progression model can be adapted to any disease process
  • Simple hospital allocation model assigns cases to hospital facilities based on patient condition and distance to available hospital resources
  • Can be configured to model urban and national scale problems
State transition mechanism for disease progression

State transition mechanism for disease progression (Click graphic to enlarge)

Disease cases or casualties carry information about their disease and treatment states. These states are transformed by a stochastic disease progression model. Patients are allocated to the nearest available resources depending on their condition.

Where Applied

HCSim was originally developed to model scenarios involving an improvised nuclear device (IND). A version of HCSim was recently adapted to predict the impact on hospital surge capacity from a national influenza pandemic.

Conceptual view of an IND incident

Conceptual view of an IND incident showing blast and fallout areas along with the incident population and local hospitals (Click graphic to enlarge)

Application to an Improvised Nuclear Device Incident

Hospital agents admit and process patients based on their condition until they either recover or die.

To model IND incidents, information on blast effects and fallout is introduced from standard models. This information manifests as casualties exposed to radiation at various levels. The model then follows these cases to assess their impact on local hospital resources.

Application to Pandemic Influenza

Submodels and their relationships in the application of HCSim to IND incidents

Submodels and their relationships in the application of HCSim to IND incidents (Click graphic to enlarge)

For pandemic influenza, hospital resources are aggregated to the county level.

Case incidence data is provided by a national-scale epidemic model. Patients are transported to the nearest county with available resources, and the impact on national hospital capacity, especially critical care facilities, is followed.

Confidence in the Model

HCSim is a relatively new capability still in development.

Acute radiation syndrome progression model

Acute radiation syndrome progression model (Click graphic to enlarge)

Results for the two types of applications shown here have been presented at several high-level workshops and forums, where they have been subjected to expert scrutiny. As a result, NISAC’s confidence in the model for these applications is relatively high.

Documentation/References

  • J. Ambrosiano, R. Bent, B. Edwards, HCSim: Developing an Agent Model for Urban-scale Healthcare Facility Impact, Risk Symposium 2009, Santa Fe, NM, April 7-9

Percentage of occupancy of hospital beds following an IND incident

Comparing the percentage of occupancy of hospital beds in the days following an IND incident in a major metropolitan area under three different hospital admission policies: permissive (top), moderate (middle) and highly restrictive (bottom) (Click graphic to enlarge)

Normalized influenza case

Normalized influenza case incidence for several locations (Click to enlarge)

Percentage of occupancy of hospital beds following an IND incident

Demand for critical care beds as a result of seasonal flu for several counties in the United States. (Click graphic to enlarge)

Operated by Los Alamos National Security, LLC for the U.S. Department of Energy's NNSA
Inside | © Copyright 2014 Los Alamos National Security, LLC All rights reserved | Disclaimer/Privacy | Web Contact