Person-centredness in the care of older adults: a systematic review of questionnaire-based scales and their measurement properties

Abstract Background Person-centredness is promoted as a central feature of the long-term care of older adults. Measures are needed to assist researchers, service planners and regulators in assessing this feature of quality. However, no systematic review exists to identify potential instruments and to provide a critical appraisal of their measurement properties. Method A systematic review of measures of person-centredness was undertaken. Inclusion criteria restricted references to multi-item instruments designed for older adult services, or otherwise with measurement properties tested in an older adult population. A two-stage critical appraisal was conducted. First, the methodological quality of included references was assessed using the COSMIN toolkit. Second, seven measurement properties were rated using widely-recognised thresholds of acceptability. These results were then synthesised to provide an overall appraisal of the strength of evidence for each measurement property for each instrument. Results Eleven measures tested in 22 references were included. Six instruments were designed principally for use in long-stay residential facilities, and four were for ambulatory hospital or clinic-based services. Only one measure was designed mainly for completion by users of home care services. No measure could be assessed across all seven measurement properties. Despite some instruments having promising measurement properties, this was consistently undermined by the poor methodological quality underpinning them. Testing of hypotheses to support construct validity was of particularly low quality, whilst measurement error was rarely assessed. Two measures were identified as having been the subject of the most rigorous testing. Conclusion The review is unable to unequivocally recommend any measures of person-centredness for use in older adult care. Researchers are advised to improve methodological rigour when testing instruments. Efforts may be best focused on testing a narrower range of measurement properties but to a higher standard, and ensuring that translations to new languages are resisted until strong measurement properties are demonstrated in the original tongue. Limitations of the review include inevitable semantic and conceptual challenges involved in defining ‘person-centredness’. The review protocol was registered with PROSPERO (ref: CRD42014005935).