BackgroundThe feedback of patient-reported outcome measures (PROMs) data is intended to support the care of individual patients and to act as a quality improvement (QI) strategy.ObjectivesTo (1) identify the ideas and assumptions underlying how individual and aggregated PROMs data are intended to improve patient care, and (2) review the evidence to examine the circumstances in which and processes through which PROMs feedback improves patient care.DesignTwo separate but related realist syntheses: (1) feedback of aggregate PROMs and performance data to improve patient care, and (2) feedback of individual PROMs data to improve patient care.InterventionsAggregate – feedback and public reporting of PROMs, patient experience data and performance data to hospital providers and primary care organisations. Individual – feedback of PROMs in oncology, palliative care and the care of people with mental health problems in primary and secondary care settings.Main outcome measuresAggregate – providers’ responses, attitudes and experiences of using PROMs and performance data to improve patient care. Individual – providers’ and patients’ experiences of using PROMs data to raise issues with clinicians, change clinicians’ communication practices, change patient management and improve patient well-being.Data sourcesSearches of electronic databases and forwards and backwards citation tracking.Review methodsRealist synthesis to identify, test and refine programme theories about when, how and why PROMs feedback leads to improvements in patient care.ResultsProviders were more likely to take steps to improve patient care in response to the feedback and public reporting of aggregate PROMs and performance data if they perceived that these data were credible, were aimed at improving patient care, and were timely and provided a clear indication of the source of the problem. However, implementing substantial and sustainable improvement to patient care required system-wide approaches. In the care of individual patients, PROMs function more as a tool to support patients in raising issues with clinicians than they do in substantially changing clinicians’ communication practices with patients. Patients valued both standardised and individualised PROMs as a tool to raise issues, but thought is required as to which patients may benefit and which may not. In settings such as palliative care and psychotherapy, clinicians viewed individualised PROMs as useful to build rapport and support the therapeutic process. PROMs feedback did not substantially shift clinicians’ communication practices or focus discussion on psychosocial issues; this required a shift in clinicians’ perceptions of their remit.Strengths and limitationsThere was a paucity of research examining the feedback of aggregate PROMs data to providers, and we drew on evidence from interventions with similar programme theories (other forms of performance data) to test our theories.ConclusionsPROMs data act as ‘tin openers’ rather than ‘dials’. Providers need more support and guidance on how to collect their own internal data, how to rule out alternative explanations for their outlier status and how to explore the possible causes of their outlier status. There is also tension between PROMs as a QI strategy versus their use in the care of individual patients; PROMs that clinicians find useful in assessing patients, such as individualised measures, are not useful as indicators of service quality.Future workFuture research should (1) explore how differently performing providers have responded to aggregate PROMs feedback, and how organisations have collected PROMs data both for individual patient care and to improve service quality; and (2) explore whether or not and how incorporating PROMs into patients’ electronic records allows multiple different clinicians to receive PROMs feedback, discuss it with patients and act on the data to improve patient care.Study registrationThis study is registered as PROSPERO CRD42013005938.FundingThe National Institute for Health Research Health Services and Delivery Research programme.
How will artificial intelligence and bioinformatics change our understanding of IgA in the next decade?
