Exploring complaints by female and male patients at Swedish hospitals using a probabilistic graphical model
Journal article, 2018
Background: Patients’ complaints have been highlighted as important for constructively improving healthcare services. In so doing, it may be important to identify disparities in experiences based on patients’ demographics, such as sex. Aim: To explore hospital recorded complaints addressing potential sex differences and whether complaints were reported by the patient or a relative. Methods: Quantitative study of all 835 closed patient complaints during 2013 at three mid-sized hospitals in Sweden. The complaints were categorisation based on perceived quality theory and analysed using a probabilistic graphical model. The findings were validated through qualitative interviews. Findings: Female patients were more likely than male patients to report dissatisfaction with interpersonal issues, whereas male patients were more likely to report dissatisfaction with administration. If a complaint from a male patient had been reported by a relative, the matter was more likely to be interpersonal. Improvement suggestions were predominantly reported by staff. However, patients and relatives proved more likely than staff to report improvement suggestions when dissatisfied with interpersonal matters. Conclusion: Using a Bayesian network, this article suggests that complaints in health care should be more holistically understood and the factors should be viewed as interconnected. This article addresses complaints as an important source of identifying not only perceived healthcare deficiencies and sex disparities, but also improvement suggestions.
probabilistic graphical model