Portal de Periódicos da CAPES

New enhanced clustering algorithms for patient referrals in medical consortia

2022; Elsevier BV; Volume: 169; Linguagem: Inglês

10.1016/j.cie.2022.108257

1879-0550

Yuchen Hao, Jianghua Zhang, Weibo Liu, Mark Goh,

Healthcare Policy and Management

Medical consortia is a form of healthcare capacity sharing and an effective means to maximizing healthcare resource utilization. To fully satiate patient demand and derive maximum efficiency for the medical consortium, patient referrals among the consortium members is a key mechanism. Patient referrals require two main steps: allocating patients to hospitals (i.e., patient assignment) and assigning patients to each time slot (i.e., patient scheduling). Patient treatment cost and tardiness are two concerns for the patients and are treated simultaneously. A clustering algorithm based on Fuzzy-C Means (FCM) is proposed to solve the problem, with patients clustered for each hospital based on their characteristics and preferences. Solution-enhancing strategies are developed to improve the schedule quality, including a virtual patient moving strategy (VPMS) and local search algorithms (LSA). Three spatial distributions of patients, central clumped, random and uniform distributions, are used to validate the performance of the algorithms. From the experimental results, the FCM-VPMS algorithm yields patient referral solutions at the lowest treatment cost, while the FCM-LSA performs better on minimizing total patient tardiness. On the cost efficiency of the algorithms, VPMS performs well with a central clumped distribution, while the LSAs show higher efficiency with the random and uniform distributions respectively.