Perceived Nursing Errors and Sleep Quality among Shift Working Nursing Officers in Two Selected Government Hospitals, Sri Lanka

Abstract

Background: Nurses play key roles to carry out core components of routine, delegated patient care in a clinical setting, where shift work is crucial. Regardless shift work being important, the disturbances to sleep and therein forth the sleep quality of a nurse, compromises rapidly. Level of performance accuracy, with an increased risk of nursing errors in patient care, is affected by the alterations to daily routine/shiftwork among nurses. Objectives: To assess sleep quality and its association to perceived nursing errors among shift working nursing officers. Methods: Utilizing pre-tested, standardized, Pittsburgh Sleep Quality Index (PSQI) to assess sleep quality, three visual analogue scales to self-report categorized nursing errors, and a demographic profile assessment was carried out in a descriptive cross-sectional study, using convenient sampling method. Ethical clearance from the Faculty of Medical Sciences, University of Sri Jayewardenepura, Colombo South Teaching Hospital, and permission to study conduct from Homagama Base Hospital were obtained. One-way analysis of variance and descriptive statistics were used using SPSS version 25.0 software. Results: A total number of 279 individual data were collected. Global PSQI scores reported 60.9% proportion having perceived poor sleep quality, with a minimum zero hours sleep, though having 10.5 maximum hours stayed in bed. Minor, moderate, and severe nursing error mean values were 9±16, 6±13 and 1±4 respectively. Shifts of longer hours and working in more than one consecutive shift, perceived higher error rates of minor (p<0.01) and moderate (p<0.01) nursing errors. Sleep quality by PSQI was only associated with moderate nursing errors in significance (p=0.03). Conclusions: Moderate nursing errors perceived by nurses in shift work was only statistically significant to sleep quality by PSQI. Error rates being unreported and not being directly observed may have limited the study's outcomes of interest.