BackgroundHigh performing hospitals attain excellence across multiple measures of performance and multiple departments. Studying high performing hospitals can be valuable if factors associated with high performance can be identified and applied. Factors leading to high performance are complex and an exclusive quantitative approach may fail to identify richly descriptive or relevant contextual factors. The objective of this study was to undertake a systematic review of qualitative literature to identify methods used to identify high performing hospitals, the factors associated with high performers, and practical strategies for improvement.MethodsMethods used to collect and summarise the evidence contributing to this review followed the ‘enhancing transparency in reporting the synthesis of qualitative research’ protocol. Peer reviewed studies were identified through Medline, Embase and Cinahl (Jan 2000-Feb 2014) using specified key words, subject terms, and medical subject headings. Eligible studies required the use of a quantitative method to identify high performing hospitals, and qualitative methods or tools to identify factors associated with high performing hospitals or hospital departments. Title, abstract, and full text screening was undertaken by four reviewers, and inter-rater reliability statistics were calculated for each review phase. Risk of bias was assessed. Following data extraction, thematic syntheses identified contextual factors important for explaining success. Practical strategies for achieving high performance were then mapped against the identified themes.ResultsA total of 19 studies from a possible 11,428 were included in the review. A range of process, output, outcome and other indicators were used to identify high performing hospitals. Seven themes representing factors associated with high performance (and 25 sub-themes) emerged from the thematic syntheses: positive organisational culture, senior management support, effective performance monitoring, building and maintaining a proficient workforce, effective leaders across the organisation, expertise-driven practice, and interdisciplinary teamwork. Fifty six practical strategies for achieving high performance were catalogued.ConclusionsThis review provides insights into methods used to identify high performing hospitals, and yields ideas about the factors important for success. It highlights the need to advance approaches for understanding what constitutes high performance and how to harness factors associated with high performance.Electronic supplementary materialThe online version of this article (doi:10.1186/s12913-015-0879-z) contains supplementary material, which is available to authorized users.

Objective Patients can provide a unique perspective on the safety of care in hospitals. Understanding that the extent to which the way hospitals are organized for quality and safety is associated with patient perceptions of care is becoming increasingly valued and necessary for the direction of targeted interventions across healthcare systems. The UK-developed patient measure of safety (PMOS) assesses eight domains of ward safety from the patient point of view and has recently been adapted and piloted in Australia. The aim of this study is to test the psychometric properties of PMOS-Australia (PMOS-A) amongst a large cohort of hospitalized patients. Design Cross-sectional questionnaire validation assessment. Setting and participants As part of the DUQuA project, the PMOS-A survey was distributed within acute myocardial infarction, hip fracture and stroke departments across 32 large public hospitals in Australia. Patients could complete the PMOS-A independently, or request the assistance of a family member/guardian, or staff on the wards—space was included to record mode of completion. Main outcome measures Confirmatory factor analysis (CFA) was undertaken on a calibration sample to generate the model, and a validation sample was used to cross-validate the model. A subset of only those participants who received assistance for PMOS-A completion was also tested using CFA on a calibration and validation sample. Model fit indices (chi-square to degrees of freedom ratio [Chi-square:DF], root mean square error of approximation [RMSEA], comparative fit indices [CFI], standardized root mean squared residual [SRMR]), Cronbach’s α, average inter-item correlations, construct reliability and cross-loadings were examined with reference to recommended thresholds to establish the extent of convergent validity and discriminant validity. A marker of criterion validity was assessed through testing associations between the PMOS-A and adherence to clinical guidelines. Results Across the calibration and validation samples of the full (N = 911) and assisted completers only subset (N = 490), three (Chi-square:DF, SRMR, RMSEA) of the four indices consistently or almost always met thresholds for acceptable model fit. CFI indices did not meet the recommended limits (0.72–0.78, against a target > 0.9). Positive relationships were found for all tests between PMOS-A and adherence to clinical guidelines, and these were significant when assessed in the calibration datasets for the full and assisted completion samples. Conclusion A sufficiently reliable and valid measure of patient perceptions of safety has been developed. These findings should provide adequate support to justify the use of this measure to assess patient perceptions of safety in Australian hospitals and can be modified for use elsewhere.

Translating evidence into complex health systems is an ongoing challenge. Building the capacity of healthcare workers in behavioral and implementation science methods may facilitate the use of evidence-based implementation approaches, leading to sustainable and effective translation. The aim was to describe the development, contents and evaluation of a training workshop aimed at upskilling hospital-embedded staff to deliver an evidence-based implementation approach. The Hide and Seek Project (HaSP) is a cluster randomized controlled trial testing two implementation approaches for improving hereditary cancer referral at eight Australian hospitals. Healthcare workers were recruited as “Implementation Leads” and trained via a one-day workshop—TRAining in evideNce-baSed ImpLementATion for hEalth (TRANSLATE). The purpose of TRANSLATE was to upskill Implementation Leads in the delivery of HaSP, as well as implementation science methods more broadly. Implementation Leads participated in semi-structured evaluation interviews, which were analyzed using inductive thematic analysis. Nine Implementation Leads from various professional backgrounds completed the training. Four key themes were identified: (i) training day reactions, (ii) learning, (iii) implementation barriers and facilitators, and (iv) building health system capacity for implementation. Participants reported high levels of satisfaction, and anticipated that the knowledge and skills may be useful in the future. We describe a novel training program focused on the delivery of evidence-based implementation within health systems. Guided by insights from this study, methods to deliver the training on a larger scale and across different contexts are being explored. The prolonged impact of TRANSLATE will be further evaluated at trial completion. Trial registration: ANZCTR, ACTRN12618001072202. Registered on June 27, 2018

Background Lynch syndrome (LS) is an inherited, cancer predisposition syndrome associated with an increased risk of colorectal, endometrial and other cancer types. Identifying individuals with LS allows access to cancer risk management strategies proven to reduce cancer incidence and improve survival. However, LS is underdiagnosed and genetic referral rates are poor. Improving LS referral is complex, and requires multisystem behaviour change. Although barriers have been identified, evidence-based strategies to facilitate behaviour change are lacking. The aim of this study is to compare the effectiveness of a theory-based implementation approach against a non-theory based approach for improving detection of LS amongst Australian patients with colorectal cancer (CRC). Methods A two-arm parallel cluster randomised trial design will be used to compare two identical, structured implementation approaches, distinguished only by the use of theory to identify barriers and design targeted intervention strategies, to improve LS referral practices in eight large Australian hospital networks. Each hospital network will be randomly allocated to a trial arm, with stratification by state. A trained healthcare professional will lead the following phases at each site: (1) undertake baseline clinical practice audits, (2) form multidisciplinary Implementation Teams, (3) identify target behaviours for practice change, (4) identify barriers to change, (5) generate intervention strategies, (6) support staff to implement interventions and (7) evaluate the effectiveness of the intervention using post-implementation clinical data. The theoretical and non-theoretical components of each trial arm will be distinguished in phases 4–5. Study outcomes include a LS referral process map for each hospital network, with evaluation of the proportion of patients with risk-appropriate completion of the LS referral pathway within 2 months of CRC resection pre and post implementation. Discussion This trial will determine the more effective approach for improving the detection of LS amongst patients with CRC, whilst also advancing understanding of the impact of theory-based implementation approaches in complex health systems and the feasibility of training healthcare professionals to use them. Insights gained will guide the development of future interventions to improve LS identification on a larger scale and across different contexts, as well as efforts to address the gap between evidence and practice in the rapidly evolving field of genomic research. Trial registration ANZCTR, ACTRN12618001072202 . Registered on 27 June 2018. Electronic supplementary material The online version of this article (10.1186/s13063-019-3457-6) contains supplementary material, which is available to authorized users.