Reliability, feasibility, and validity of the quality of interactions schedule (QuIS) in acute hospital care: an observational study

Abstract

Background: Research into relational care in hospitals will be facilitated by a focus on staff-patient interactions. The Quality of Interactions Schedule (QuIS) uses independent observers to measure the number of staff-patient interactions within a healthcare context, and to rate these interactions as ‘positive social’; ‘positive care’; ‘neutral’; ‘negative protective’; or ‘negative restrictive’. QuIS was developed as a research instrument in long term care settings and has since been used for quality improvement in acute care. Prior to this study, its use had not been standardised, and reliability and validity in acute care had not been established. Methods: In 2014 and 2015 a three - phase study was undertaken to develop and test protocols for the use of QuIS across three acute wards within one NHS trust in England. The phases were: (1) A pilot of 16 h observation which developed implementation strategies for QuIS in this context; (2) training two observers and undertaking 16 h of paired observation to inform the development of training protocols; (3) training four nurses and two lay volunteers according to a finalised protocol followed by 36 h of paired observations to test inter-rater agreement. Additionally, patients were asked to rate interactions and to complete a shortened version of the Patient Evaluation of Emotional Care during Hospitalisation (PEECH) questionnaire. Results: Protocols were developed for the use of QuIS in acute care. Patients experienced an average of 6.7 interactions/patient/h (n = 447 interactions). There was close agreement between observers in relation to the number of interactions observed (Intraclass correlation coefficient (ICC) = 0.97) and moderate to substantial agreement on the quality of interactions (absolute agreement 73%, kappa 0.53 to 0.62 depending on weighting scheme). There was 79% agreement (weighted kappa 0.40: P < 0.001; indicating fair agreement) between patients and observers over whether interactions were positive, negative or neutral. Conclusions: Observers using clear QuIS protocols can achieve levels of agreement that are acceptable for the use of QuIS as a research instrument. There is fair agreement between observers and patients’ rating of interactions. Further research is needed to explore the relationship between QuIS measures and reported patient experience.