A hybrid model to support decision making in the stroke clinical pathway

Health processes are highly complex and sensitive, requiring continued improvement. Several methodologies have been developed to support these improvement processes, many of which apply Process Mining (PM) to investigate the current states of processes. Some even use Discrete Event Simulation (DES) techniques to explore alternative scenarios. Among the ones using DES, some rely on Multicriteria Decision-Making Methods (MCDM) to guide them in choosing the best scenario, especially in complex decision-making environments. However, no author presents a single framework combining all three techniques mentioned above, to wit, PM, DES, and MCDM. This paper targets developing this combination and exploring its usefulness in guiding process improvement in the health services area. The clinical path for stroke patients, covering from onset of symptoms to hospital discharge, was used as a test case. The research demonstrated that this proposal provided an innovative diagnostic of the process, identifying the activities that should be improved to obtain the most significant results in the process management indicators. Identifying the activities ensures that the impacts of these improvements are indicated with statistical ac-curacy and eliminating the need to perform on-site testing to test for the best solutions. The research does not address root cause diagnosis of activities to be improved. Therefore, further research needs to be undertaken on an extension of the model to enhance root cause analysis capabilities with a view to validating the framework proposed in other environments and, in this way, assess its replicability.