Introduction: This pilot project proposes a novel integration of two existing simulation technologies in order to model, analyze and measure treatment for diabetic retinopathy (DR) in the population served by the St. Louis VA Medical Center (St. Louis VAMC). Specifically, to use agent-based modeling to simulate the population, and discrete event simulation to model the eye clinic. The purpose of this hybrid simulation is to enable a new type of measurement of clinical services. Changes to the Medical Center capacity, flow, size or policy will be measurable at the level of the population health, on the rate of DR progression in the population. Conversely, changes in the population epidemiology or demography may be measured at the level of demand for access to clinical services.
Aims : This project will have two fundamental aims: (1) to build a simulated cohort of diabetic patients which mirrors in all relevant respects the population of diabetic patients seeking treatment or surveillance for DR at St. Louis VAMC;and (2) to implement this population in combination with a discrete event simulation representing the St. Louis VAMC eye clinic in order to project how demand for service, and DR progression in the population, will evolve over the next ten years. Methods: The cohort will be mined for relevant covariates for DR, which will be implemented in an accelerated failure time model for DR progression, and integrated into a statechart model of DR for the agent based population. This population will be married to the simulated clinic, where individual agents seek treatment and care for DR according to standard processes extant in the St. Louis VAMC eye clinic. Projections of demand for service over the next ten year based on population wide improvements in Body Mass Index and Hemoglobin HbA1c will be measured, as well as projections of access to care based on potential reductions in screening for background DR based on recommendations from the literature on DR. Implementation: The prospective PI has established ties with the Acting Chief of Ophthalmology at St. Louis VAMC, and the prospective Co-Investigator is a practicing ophthalmologist in the eye clinic. Results from the pilot study will be readily adoptable at the local level, and may be disseminated throughout VA in part through the VA Innovations Program, and through traditional channels for reporting successful research. Additionally, it is the intention of the investigators to advance this project through further Merit Review application upon completion of this pilot.

Public Health Relevance

This project proposes to create a novel computer simulation which is capable of modeling the interaction between a population of veterans and the Medical Center which provides them with access to medical care. Specifically, this pilot study will examine the projected consequences of changes made to screening practices for diabetic retinopathy (DR), as well as changes to the health of the population of veterans at risk for DR. By designing a simulated population which mimics in all relevant respects the actual population of veterans seeking treatment for DR at St. Louis VAMC, along with a simulation of the medical center's eye clinic services, we will be able to predict the changing epidemiology and projected demand for medical care over the next ten years. This investigation will allow us to pinpoint which proposed measures will be most beneficial to implement for optimal care.

Agency
National Institute of Health (NIH)
Institute
Veterans Affairs (VA)
Type
Non-HHS Research Projects (I01)
Project #
1I01HX000622-01
Application #
8195669
Study Section
Blank (HSR7)
Project Start
2011-10-01
Project End
2012-09-30
Budget Start
2011-10-01
Budget End
2012-09-30
Support Year
1
Fiscal Year
2013
Total Cost
Indirect Cost
Name
St. Louis VA Medical Center
Department
Type
DUNS #
033986766
City
St. Louis
State
MO
Country
United States
Zip Code
63106
Day, T Eugene; Ravi, Nathan; Xian, Hong et al. (2014) Sensitivity of diabetic retinopathy associated vision loss to screening interval in an agent-based/discrete event simulation model. Comput Biol Med 47:7-12
Day, Theodore Eugene; Ravi, Nathan; Xian, Hong et al. (2013) An Agent-Based Modeling Template for a Cohort of Veterans with Diabetic Retinopathy. PLoS One 8:e66812