Understanding urban hospital bypass behaviour based on big trace data

The debates about patients' decisions regarding which hospital to visit have been shown in many health research literature for the past few years. Researchers have developed various methods to understand hospital bypass behaviour of patients; however, previous studies ignore the impact of spatial heterogeneity and awareness of patients about surrounding hospitals such as travel distance and hospital distributions in the process of bypass behaviour definitions and evaluations. To address these limitations, this study puts forward a Hospital Bypass Index (HBI) to understand urban hospital bypass behaviour by using big trace data collected by urban taxis. To evaluate the bypass behaviour, we defines three evaluation indicators for HBI including a potential bypass rate, an overall distance decay parameter, and a diurnal variation in distance decay parameter by mining large-scale patient-hospital trips from a spatiotemporal perspective. Experiments are conducted with 30 general hospitals and 13 specialty hospitals in Wuhan city, China, by using one month of taxi traces. The results of bypass behaviour evaluation and comparisons indicate that the proposed method is effective and feasible, which is promising for health departments to optimize the medical services and rationally allocate the medical facilities.