Search

Introduction: Many patients diverted to hyper acute stroke units (HASUs) with suspected stroke have other diagnoses. This may delay treatment for non-stroke patients and overburden stroke teams. We sought to describe real world pre-hospital stroke recognition in a UK centralised stroke system.Methods: We extracted all patients presenting to Salford HASU on the stroke pathway from 01/08/2015 to 28/02/2017. Case notes were manually reviewed by a Consultant Stroke Physician where discharge diagnosis was not coded. To identify false negative cases, all patients with a discharge diagnosis of stroke were extracted. Linked ambulance data were manually extracted from scanned records. Additional qualitative work was undertaken.Results: 5,809 patients were conveyed via ambulance, to the HASU on the stroke pathway during the study period; currently, 5,125 cases have been analysed. 2003 (39%) were false-positives. Of these, the 5 most common diagnoses were sepsis (n = 360 [7%]), migraine (n = 282 [6%]), epilepsy (n = 277 [5%]), limb/face pathology (n = 124 [2%]), syncope (n = 110 [2%]). To date, 84 stroke patients were identified as false-negatives, but further analysis of ambulance records is required. Analysis of focus group and interview data with 16 pre-hospital clinicians identifies respondents receive limited feedback from jobs which impedes their ability to learn from experiences. Respondents report difficulty in recognising differential diagnoses, expressing lack of confidence to rule out stroke and greater concern for ‘missed strokes’. An enhanced FAST tool, better relations with hospital clinicians, and education on differential diagnoses were all considered necessary to improve accurate stroke detection.Conclusion: Findings support the development of pre-hospital interventions to improve stroke recognition.

Professionals with managerial responsibilities are often seen as natural boundary spanners, brokering knowledge between the professional and managerial domains. However, the tensions implicit in their knowledge brokering activities, especially when the latter are formalized, remain under-researched. Using interviews and observational data, we examine the case of an inter-organizational collaborative research partnership that deployed clinical professionals as designated knowledge brokers across multiple sites. We identify three strategies that such hybrid professionals utilized to surmount challenges associated with knowledge brokering in the fragmented healthcare context: (1) relying on additional boundary 'bridges'; (2) conforming to existing ways of doing things; and (3) shifting from 'facilitating' to 'doing'. An analysis of these strategies highlights the tensions between different dimensions of brokering reinforced by macro-level institutional arrangements, the intertwining of formalized and emergent elements of brokering as a collectively enacted phenomenon, and the limitations of knowledge brokering professionals arising from their professional expertise and authority.

Policies and interventions in the health-care system may have a wide range of effects on multiple patient outcomes and operate through many clinical processes. This presents a challenge for their evaluation, especially when the effect on any one patient is small. In this essay, we explore the nature of the health-care system and discuss how the empirical evidence produced within it relates to the underlying processes governing patient outcomes. We argue for an evidence synthesis framework that first models the underlying phenomena common across different health-care settings and then makes inferences regarding these phenomena from data. Bayesian methods are recommended. We provide the examples of electronic prescribing and increased consultant provision at the weekend.