Richards‐Campbell sleep questionnaire: psychometric properties of Chinese critically ill patients.

Background: Sleep abnormalities occur frequently in critically ill patients. Nurses are strategically placed, specifically in intensive care units, to promote sleep in such patients. Currently, an effective sleep assessment tool in Chinese is not available for intensive care settings. Aim: This study aimed to assess the reliability and validity of the Chinese version of the Richards‐Campbell Sleep Questionnaire (RCSQ‐C). It also aimed to evaluate patient‐nurse reliability and agreement of the RCSQ‐C in the intensive care unit (ICU). Methods: We translated the original RCSQ into Chinese and then back‐translated it into English to ensure its accuracy of translation. Internal consistency, discrimination validity and construct validity of the RCSQ‐C were examined in 150 critically ill patients. The convergent validity of the RCSQ‐C was evaluated in 44 of 150 critically ill patients, and data from the RCSQ‐C were compared with those of the Chinese version of St Mary's Hospital Sleep Questionnaire (SMHSQ). Comparisons were also made between RCSQ‐C scores obtained from patients and their nurses. Results: Cronbach's α of the RCSQ‐C was 0.923; thus, it showed high reliability. The corrected item‐total correlation coefficient was in the range of 0·680∼0·805, which showed that the items were homogeneous for evaluating sleep. The content validity was 0·84. One factor was extracted with a cumulative contribution rate of 76·597%. Confirmatory factor analysis showed that the original single‐factor structure proposed by Richards adequately fit the data. The RCSQ‐C could discriminate poor and good sleepers, which supported discriminant validity. There was a close correlation between the scores obtained from the RCSQ patient's version and those from the SMHSQ. The intraclass correlation coefficients of the patient and nurse ranged from 0·315 to 0·609. Conclusions: The psychometric properties of the RCSQ‐C suggest its utility in critically ill patients. Patient‐nurse reliability on the RCSQ‐C was "fair" to "substantial", with nurses tending to overestimate patients' perceived sleep quality. Relevance to Clinical Practice: If the validity of this questionnaire is supported in other ICU samples, RCSQ‐C could be used as a routine evaluation instrument to distinguish good and poor sleepers and then direct nurses to form corresponding treatment plans to promote sleep. [ABSTRACT FROM AUTHOR]