Adverse events at the end of life of hospital patients with or without a condition relevant for palliative care: a nationwide retrospective record review study in the Netherlands.

Abstract

Background: Patient safety is crucial for quality of care. Preventable adverse events (AEs) occur in 1 of 20 patients in the hospital, but it is unknown whether this is different for patients with a condition relevant for palliative care. The majority of the limited available research on this topic is only focused on patients already receiving palliative care, and do not make comparisons with other patients at the end-of-life. We identified and compared the prevalence, preventability, nature and causes of AEs in patients with and without a condition relevant for palliative care.

Methods: A nationwide retrospective record review study was performed in 20 Dutch hospitals. A total of 2,998 records of patients who died in hospital in 2019 was included. Records were reviewed for AEs. We identified two subgroups: patients with (n = 2,370) or without (n = 248) a condition relevant for palliative care through the selection method of Etkind (2017). Descriptive analyses were performed to calculate prevalence, nature, causes and prevention strategies. T-tests were performed to calculate differences between subgroups.

Results: We found no significant differences between subgroups regarding AE prevalence, this was 15.3% in patients with a condition relevant for palliative care, versus 12.0% in patients without a condition relevant for palliative care (p = 0.148). Potentially preventable AE prevalence was 4.3% versus 4.4% (p = 0.975). Potentially preventable death prevalence in both groups was 3.2% (p = 0.938). There were differences in the nature of AEs: in patients with a condition relevant for palliative care this was mostly related to medication (33.1%), and in patients without a condition relevant for palliative care to surgery (50.8%). In both subgroups in the majority of AEs a patient related cause was identified. For the potentially preventable AEs in both subgroups the two most important prevention strategies as suggested by the medical reviewers were reflection and evaluation and quality assurance.

Discussion: Patient safety risks appeared to be equally prevalent in both subgroups. The nature of AEs does differ between subgroups: medication- versus surgery-related, indicating that tailored safety measures are needed. Recommendations for practice are to focus on reflecting on AEs, complemented with case evaluations.