B Lloyd-Evans1, M Slade, D P Osborn, R Skinner, S Johnson. 1. Department of Mental Health Sciences, University College London (Bloomsbury Campus), Charles Bell House, 67-73 Riding House Street, London W1W 7EJ, UK. b.lloyd-evans@ucl.ac.uk
Abstract
BACKGROUND: Measurement of what happens in mental health services is needed to describe services, identify variation in care provision and understand service outcomes. However, there is no consensus about appropriate methods or measures for content of care. Previous research has primarily used a single information source and prioritised staff over patient perspectives on content of care. This study aims to enhance understanding of how to measure content of care by developing and evaluating four instruments, each using a different measurement method. METHODS: Development is described by four instruments-CaSPAR, CaRICE, CCCQ-S and CCCQ-P-which use researcher observation, staff-report and patient-report to measure the intensity and nature of care at services. Inter-rater reliability of CaRICE and CCCQ-S was investigated. Concordance between staff and patient perspectives was explored through assessing inter-rater agreement of CCCQ-S and CCCQ-P questionnaires. The convergence of data from the measures collected in an inpatient multi-site study was investigated. RESULTS: CaRICE demonstrated good inter-rater reliability (κ = 0.71). CCCQ-S inter-rater reliability was poor. Concordance between staff and patient reports was low: there was a trend for patients to report less care received than staff reported had been provided. Results from CaSPAR, CaRICE and CCCQ-P exhibited divergence, indicating possible differences in patient, staff and researcher perspectives. DISCUSSION: Information about content of care should be sought as close as possible to the point of delivery. There may be differing, valid perspectives about care provided by services. Further development of methods to measure content of care is required. Meanwhile, a multi-methods approach should be adopted, which allows inclusion of different perspectives (specifically including the patient's perspective) and triangulation of results from different measurement methods. CaSPAR, CaRICE and CCCQ-P can provide multi-perspective content of care measurement in inpatient services.
BACKGROUND: Measurement of what happens in mental health services is needed to describe services, identify variation in care provision and understand service outcomes. However, there is no consensus about appropriate methods or measures for content of care. Previous research has primarily used a single information source and prioritised staff over patient perspectives on content of care. This study aims to enhance understanding of how to measure content of care by developing and evaluating four instruments, each using a different measurement method. METHODS: Development is described by four instruments-CaSPAR, CaRICE, CCCQ-S and CCCQ-P-which use researcher observation, staff-report and patient-report to measure the intensity and nature of care at services. Inter-rater reliability of CaRICE and CCCQ-S was investigated. Concordance between staff and patient perspectives was explored through assessing inter-rater agreement of CCCQ-S and CCCQ-P questionnaires. The convergence of data from the measures collected in an inpatient multi-site study was investigated. RESULTS:CaRICE demonstrated good inter-rater reliability (κ = 0.71). CCCQ-S inter-rater reliability was poor. Concordance between staff and patient reports was low: there was a trend for patients to report less care received than staff reported had been provided. Results from CaSPAR, CaRICE and CCCQ-P exhibited divergence, indicating possible differences in patient, staff and researcher perspectives. DISCUSSION: Information about content of care should be sought as close as possible to the point of delivery. There may be differing, valid perspectives about care provided by services. Further development of methods to measure content of care is required. Meanwhile, a multi-methods approach should be adopted, which allows inclusion of different perspectives (specifically including the patient's perspective) and triangulation of results from different measurement methods. CaSPAR, CaRICE and CCCQ-P can provide multi-perspective content of care measurement in inpatient services.