Designing a rural network of dialysis facilities

Kidney specialists treat chronic kidney failure with dialysis until transplant or death. Patients travel to in-centre or satellite hemodialysis facilities for each four-hour treatment, three times per week or participate in home peritoneal dialysis or home hemodialysis. We observe through a patient preference survey that regardless of the travel burden, many patients will always opt to go to an in-centre or satellite facility, while some will always opt for home dialysis. For others, the choice will vary depending on the location of available facilities. We propose a mathematical model for the Dialysis Facility Network Design Problem, where long travel times impact access to care, especially for patients who live in rural areas. The proposed model can help identify the best network of dialysis facilities with consideration for budget and capacity management constraints as well as patient preferences for facility or home dialysis. In the application of our proposed model through a case study, we find that with the same budget as the existing facility network, it is possible to achieve considerable reductions in maximum and mean patient travel times with less variability. Due to the potential cost savings from home dialysis, we compare further improvements from increasing the proportion of patients who will consider home dialysis with similar improvements from increases in budget. For larger size problem instances, we provide a tabu search heuristic that can be used to find optimal or near-optimal solutions in a fraction of the computation time required by a commercial solver.