Abstract
Purpose :
With the growing adoption of teleretinal imaging (TRI), it is important to design a TRI-based screening program that is easily accessible over a large geographic area. This study develops a modeling framework that identifies locations to place future TRI screening centers that maximize TRI visits for in-need census tracts in a large, county-wide safety-net system.
Methods :
A location-based Monte Carlo discrete event simulation model was developed including all census tracts in Harris County to evaluate patient visits to current TRI screening centers. A total of 500,000 hypothetical patients were generated and tracked over their lifetime under annual TRI screening recommendations. Patient compliance rates were varied based on factors such as race, poverty level, and travel time. The top 20% of census tracts with the greatest need was identified by factors such as total number of visits and minimum travel time to the nearest center. A new potential center was then identified based on a weighted centroid of the top 20% census tracts where the weights are assigned proportional to census tract-specific needs. The simulation was then run again with the new center added to analyze the benefits for patients within the in-need census tracts.
Results :
For the top 20% of census tracts, the model identifies a new potential TRI screening center in northwest Harris County in an area that is currently not covered by existing centers (See Fig 1&2). For census tracts in the top 20% that are impacted by the placement of the new center, yearly TRI visits increase by 32.81% (95% CI; 32.21-33.48%) with a reduction in average travel time per impacted patient of 7.96 minutes (95% CI; 7.93-7.99 minutes). Furthermore, average cost ($)/QALY per impacted patient decreased by $19.06 (95% CI; $15.42-$22.70).
Conclusions :
The location model can be a useful tool in designing a TRI-based screening program that maximizes TRI visits for a heterogeneous population over a large geographic area. TRI screening locations can be identified that account for patient-specific compliance data and geodemographic factors. Positive benefits in cost/QALY can be seen on a county-level when new locations are placed strategically.
This is a 2021 ARVO Annual Meeting abstract.