Um estudo para identificar fatores que conduzem ao atraso no processo de reembolso de contas hospitalares via mineração de processos e mineração de dados

Abstract

The healthcare reimbursement process impact has been enormous for the healthcare providers and the economy. The arising of new clinical procedures, changes in regulations and policies have been increasing the complexity of the reimbursement process and consequently its duration and costs. Therefore, methods of process analysis have been used as a basic strategy to improve the organizational effectiveness of healthcare institutions. In this context, the present study investigates factors that cause delays in the reimbursement process. The proposed approach aims to combine Process Mining and Data Mining techniques to identify factors that can explain the reimbursement process delay. Process Mining techniques allow exploring in detail how activities can impact the process execution, as well as the occurrence of bottlenecks that may indicate the need for a systematic investigation to detect its root causes. Considering this, Data Mining can be employed through techniques, such as associative rules that can be used to identify unknown relationships. Hence, this study demonstrates through a real case the benefits that the combination of Process Mining and Data Mining techniques to support the audit and billing activities of the reimbursement process. A subjective evaluation of the mined rules showed that almost 45% of them were considered relevant or very relevant for the identification of factors that can lead to delay in the reimbursement process.