Assessing the impact of healthcare service risks on healthcare demand under evolving economic and social structures: An improved GLDS decision making method considering risk attitudes

Abstract

Faced with backward and uneven economic conditions and an irrational and increasingly inequitable social structure, healthcare disparities are widening and the healthcare inequalities are expanding. This can lead to an increasing public demand for healthcare services, posing unprecedented risk, such as resource shortages, quality deterioration and rising healthcare costs. By identifying, managing, and optimizing the aforementioned risks within the healthcare system, it helps the healthcare system better address the challenges of expanding healthcare demands under evolving economic and social structure. To achieve this, this paper employs failure mode and effects analysis (FMEA) to identify severe failure modes in healthcare services. However, traditional FMEA methods possess limitations in assessing the dynamics of trust relationships and interaction effects among experts within social networks. Therefore, this paper proposes an improved gained and lost dominance score (GLDS) based social network decision-making method considering risk attitudes for FMEA to prioritize failure modes in healthcare services. First, transform linguistic risk assessment information from an interdisciplinary team of experts based on probabilistic linguistic term sets handle uncertain efficiently. Meanwhile, experts’ influence is reflected by an extended PageRank algorithm with trust degrees based on social networks analysis. Next, an improved GLDS method considering the differentiated risk attitudes of individuals and groups is developed to prioritize failure modes. Especially, to solve the problems of soft preference and incomparability relation among failure modes, which are ignored by the traditional GLDS method, a conflict analysis is established to determine the preference, indifference, and incomparability relations. Finally, the practicality of the proposed method is illustrated by the healthcare service risk case in the emergency department and the effectiveness of the developed method is validated by comparison and sensitivity experiments. The results suggest that failure to update labels showing patient severity promptly is the most dangerous. In addition, it would be more economically efficient to analyze the non-comparable failure modes facing different risk factors.