Setting sites for faster results: Optimizing locations and capacities of new tuberculosis testing laboratories

Abstract

A reduced turnaround time to obtain initial test results for detecting the bacterial agent that causes tuberculosis (TB) can significantly improve the management of this communicable disease. We propose a new facility location modeling approach for designing a network of TB testing laboratories so as to reduce transportation times and thereby decrease overall test turnaround time. Our mathematical formulation chooses new laboratory facility locations, the capacities of those laboratories, as well as the assignment of TB samples to regional laboratories so as to minimize total sample transportation time, while considering constraints on the laboratory capacities, budget, and maximum transportation times between any origin and the testing laboratory. We ran the model using the TB testing demand and laboratory data for the TB reference laboratory in the Canadian province of British Columbia (BC), and demonstrate its potential to improve patient care and population health by optimizing the locations of additional testing laboratories. Furthermore, we explored data limitations and assumptions through several sensitivity analyses.