A systems analysis of the COVID-19 pandemic response in the United Kingdom - Part 1 - The overall context.

The most common reaction to suggesting that we could learn valuable lessons from the way the current pandemic has been/ is being handled, is to discourage the attempt; as it is suggested that it can all be done more accurately and authoritatively after the inevitable Public Inquiry (Slater, 2019). On the other hand, a more constructive approach, is to capture and understand the work that was actually done.This would include normal activities, as well as positive adaptations to challenges and failures that may have occurred. Such an approach aimed at improving what worked, rather than blaming people for what went wrong, has the potential to contribute more successfully to controlling the consequences of the current crisis. Such an approach should thus be aimed at detecting and feeding back lessons from emerging and probably unexpected behaviours and helping to design the system to adapt better to counter the effects. The science and discipline of Human Factors (HF) promotes system resilience. This can be defined as an organisation's ability to adjust its functioning before, during or after significant disturbances (such as a pandemic), enabling adaptation and operation under both anticipated and unanticipated circumstances. A "functional" approach methodology enables the identification of where the system and its various interdependent functions (an activity or set of activities that are required to give a certain output), could be improved and strengthened; if not immediately, at least for the future. Along these lines, suggestions for adding key resilience functions are additionally identified and outlined. The application and insights gained from this functional approach to the 2015 MERS-Cov pandemic in South Korea has been seen as contributing substantially to the effective response to the current crisis in that country (Min, submitted for publication). In this paper, we present an overarching framework for a series of projects that are planned to carry out focussed systems-based analysis to generate learning from key aspects of the COVID-19 pandemic response in the United Kingdom.

On the brink of disruption: Applying Resilience Engineering to anticipate system performance under crisis.

As COVID-19 spread across Brazil, it quickly reached remote regions including Amazon's ultra-peripheral locations where patient transportation through rivers is added to the list of obstacles to overcome. This article analyses the pandemic's effects in the access of riverine communities to the prehospital emergency healthcare system in the Brazilian Upper Amazon River region. To do so, we present two studies that by using a Resilience Engineering approach aimed to predict the functioning of the Brazilian Mobile Emergency Medical Service (SAMU) for riverside and coastal areas during the COVID-19 pandemic, based on the normal system functioning. Study I, carried out before the pandemic, applied ethnographic methods for data collection and the Functional Resonance Analysis Method - FRAM for data analysis in order to develop a model of the mobile emergency care in the region during typical conditions of operation. Study II then estimated how changes in variability dynamics would alter system functioning during the pandemic, arriving at three trends that could lead the service to collapse. Finally, the accuracy of predictions is discussed after the pandemic first peaked in the region. Findings reveal that relatively small changes in variability dynamics can deliver strong implications to operating care and safety of expeditions aboard water ambulances. Also, important elements that add to the resilient capabilities of the system are extra-organizational, and thus during the pandemic safety became jeopardized as informal support networks grew fragile. Using FRAM for modelling regular operation enabled prospective scenario analysis that accurately predicted disruptions in providing emergency care to riverine population.

Safer Systems: People Training or System Tuning?

Safety is usually seen as a problem when it is absent rather than when it is present, where accidents, incidents, and the like represent a lack of safety rather than the presence of safety. To explain this lack of safety, one or more causes must be found. In the management of industrial safety, the human factor has traditionally been seen as a weak element; human error is often offered as the first, and sometimes the only cause of lack of safety and human factors have since the early days offered three principal solutions, namely training, design, and automation. Of these, training has considerable face value as an effective means to improve human performance. The drawback of safety training, however, is that it focuses on a single system component, the human, instead of on the system as a whole. Safety training further takes for granted that humans are a liability and focuses on overcoming the weakness of this specific component through simplistic models of what determines human performance. But humans may also be seen as an asset which changes the focus to strengthening how a complex socio-technical system functions. A socio-technical system comprises multiple functions that must be finely tuned in order to ensure expected and acceptable performance. Since systems cannot be made safer without developing effective ways of managing the conditions in which people work, system tuning offers an alternative solution to an old problem.

A novel approach to explore Safety-I and Safety-II perspectives in in situ simulations-the structured what if functional resonance analysis methodology.

OBJECTIVES: With ever increasingly complex healthcare settings, technology enhanced simulation (TES) is well positioned to explore all perspectives to enhance patient safety and patient outcomes. Analysis from a Safety-II stance requires identification of human adjustments in daily work that are key to maintaining safety. The aim of this paper is to describe an approach to explore the consequences of human variability from a Safety-II perspective and describe the added value of this to TES. METHODS: The reader is guided through a novel application of functional resonance analysis methodology (FRAM), a method to analyse how a system or activity is affected by human variability, to explore human adaptations observed in in situ simulations (ISS). The structured applicability of this novel approach to TES is described by application to empirical data from the standardised ISS management of paediatric time critical head injuries (TCHI). RESULTS: A case series is presented to illustrate the step-wise observation of key timings during ISSs, the construction of FRAM models and the visualisation of the propagation of human adaptations through the FRAM models. The key functions/actions that ensure the propagation are visible, as are the sequelae of the adaptations. CONCLUSIONS: The approach as described in this paper is a first step to illuminating how to explore, analyse and observe the consequences of positive and negative human adaptations within simulated complex systems. This provides TES with a structured methodology to visualise and reflect upon both Safety-I and Safety-II perspectives to enhance patient safety and patient outcomes.

Measurement of resilience potential - development of a resilience assessment grid for emergency departments.

BACKGROUND: Resilience engineering has been advocated as an alternative to the management of safety over the last decade in many domains. However, to facilitate metrics for measuring and helping analyze the resilience potential for emergency departments (EDs) remains a significant challenge. The study aims to redesign the Hollnagel's resilience assessment grid (RAG) into a custom-made RAG (ED-RAG) to support resilience management in EDs. METHODS: The study approach had three parts: 1) translation of Hollnagel's RAG into Chinese version, followed by generation of a tailored set of ED-RAG questions adapted to EDs; 2) testing and revising the tailored sets until to achieve satisfactory validity for application; 3) design of a new rating scale and scoring method. The test criteria of the ED-RAG questionnaire adopted the modified three-level scoring criteria proposed by Bloom and Fischer. The study setting of the field test is a private regional hospital. RESULTS: The fifth version of ED-RAG was acceptable after a field test. It has three sets of open structured questions for the potentials to respond, monitor, and anticipate, and a set of structured questions for the potential to learn. It contained 38 questions corresponding to 32 foci. A new 4-level rating scale along with a novel scaling method can improve the scores conversion validity and communication between team members and across investigations. This final version is set to complete an interview for around 2 hours. CONCLUSIONS: The ED-RAG represents a snapshot of EDs'resilience under specific conditions. It might be performed multiple times by a single hospital to monitor the directions and contents of improvement that can supplement conventional safety management toward resilience. Some considerations are required to be successful when hospitals use it. Future studies to overcome the potential methodological weaknesses of the ED-RAG are needed.

Staff acting resiliently at two hospital wards.

PURPOSE: The purpose of this paper is to understand how the hospital staff (nurses and physicians) at two hospital wards have coped with everyday work having leaders in conflict or longer periods without one or the other leader and whether the way the staff handled the challenges was resilient. DESIGN/METHODOLOGY/APPROACH: Through semi-structured interviews with the staff at the two wards, the authors analysed how the staff were working, if they had cooperation and interdisciplinary cooperation, how they would handle uncertainties and how they coped with the absence of their leaders. FINDINGS: The staff at both wards were handling the everyday work in a resilient way. The authors argue that to increase the resilience in an organisation, leaders should acknowledge the need to establish strong emotional ties among staff and at the same time ensure role structures that make sense in the everyday work. ORIGINALITY/VALUE: This study reports on original work and shows what decision makers could do to increase resilience in an organisation. This paper shows that the organisational context is important for the staff to act resiliently. As leaders come and go, it can be important for the stability of the organisation to promote the staff in acting resiliently independent of the leader situation.

Leadership set-up: wishful thinking or reality?

PURPOSE: The purpose of the study is to determine whether one leader set-up is better than the others according to interdisciplinary cooperation and leader legitimacy. DESIGN/METHODOLOGY/APPROACH: The study is a qualitative study based on semi-structured interviews at three Danish hospitals. FINDINGS: The study found that the leadership set-up did not have any clear influence on interdisciplinary cooperation, as all wards had a high degree of interdisciplinary cooperation independent of which leadership set-up they had. Instead, the authors found a relation between leadership set-up and leader legitimacy. In cases where staff only referred to a leader from their own profession, that leader had legitimacy within the staff group. When there were two leaders from different professions, they only had legitimacy within the staff group from their own profession. Furthermore, clinical specialty also could influence legitimacy. ORIGINALITY/VALUE: The study shows that leadership set-up is not the predominant factor that creates interdisciplinary cooperation; but rather, leader legitimacy also should be considered. Additionally, the study shows that leader legitimacy can be difficult to establish and that it cannot be taken for granted. This is something chief executive officers should bear in mind when they plan and implement new leadership structures. Therefore, it would also be useful to look more closely at how to achieve legitimacy in cases where the leader is from a different profession to the staff.

Hospital accreditation: staff experiences and perceptions.

Purpose The purpose of this paper is to understand how staff at various levels perceive and understand hospital accreditation generally and in relation to quality improvement (QI) specifically. Design/methodology/approach In a newly accredited Danish hospital, the authors conducted semi-structured interviews to capture broad ranging experiences. Medical doctors, nurses, a quality coordinator and a quality department employee participated. Interviews were audio recorded and subjected to framework analysis. Findings Staff reported that The Danish Healthcare Quality Programme affected management priorities: office time and working on documentation, which reduced time with patients and on improvement activities. Organisational structures were improved during preparation for accreditation. Staff perceived that the hospital was better prepared for new QI initiatives after accreditation; staff found disease specific requirements unnecessary. Other areas benefited from accreditation. Interviewees expected that organisational changes, owing to accreditation, would be sustained and that the QI focus would continue. Practical implications Accreditation is a critical and complete hospital review, including areas that often are neglected. Accreditation dominates hospital agendas during preparation and surveyor visits, potentially reducing patient care and other QI initiatives. Improvements are less likely to occur in areas that other QI initiatives addressed. Yet, accreditation creates organisational foundations for future QI initiatives. Originality/value The authors study contributes new insights into how hospital staff at different organisational levels perceive and understand accreditation.

"Failure-to-Identify" Hunting Incidents: A Resilience Engineering Approach.

Objective The objective was to develop an understanding, using the Functional Resonance Analysis Method (FRAM), of the factors that could cause a deer hunter to misidentify their intended target. Background Hunting is a popular activity in many communities. However, hunters vary considerably based on training, experience, and expertise. Surprisingly, safety in hunting has not received much attention, especially failure-to-identify hunting incidents. These are incidents in which the hunter, after spotting and targeting their quarry, discharge their firearm only to discover they have been spotting and targeting another human, an inanimate object, or flora by mistake. The hunter must consider environment, target, time of day, weather, and many other factors-continuously evaluating whether the hunt should continue. To understand how these factors can relate to one another is fundamental to begin to understand how incidents happen. Method Workshops with highly experienced and active hunters led to the development of a FRAM model detailing the functions of a "Hunting FRAM." The model was evaluated for correctness based on confidential and anonymous near-miss event submissions by hunters. Results A FRAM model presenting the functions of a hunt was produced, evaluated, and accepted. Using the model, potential sources of incidents or other unintended outcomes were identified, which in turn helped to improve the model. Conclusion Utilizing principles of understanding and visualization tools of the FRAM, the findings create a foundation for safety improvements potentially through training or safety messages based on an increased understanding of the complexity of hunting.

Compliance with accreditation and recommended hospital care-a Danish nationwide population-based study.

OBJECTIVE: To examine the association between compliance with accreditation and recommended hospital care. DESIGN: A Danish nationwide population-based follow-up study based on data from six national, clinical quality registries between November 2009 and December 2012. SETTING: Public, non-psychiatric Danish hospitals. PARTICIPANTS: Patients with acute stroke, chronic obstructive pulmonary disease, diabetes, heart failure, hip fracture and bleeding/perforated ulcers. INTERVENTIONS: All hospitals were accredited by the first version of The Danish Healthcare Quality Programme. Compliance with accreditation was defined by level of accreditation awarded the hospital after an announced onsite survey; hence, hospitals were either fully (n = 11) or partially accredited (n = 20). MAIN OUTCOME MEASURES: Recommended hospital care included 48 process performance measures reflecting recommendations from clinical guidelines. We assessed recommended hospital care as fulfilment of the measures individually and as an all-or-none composite score. RESULTS: In total 449 248 processes of care were included corresponding to 68 780 patient pathways. Patients at fully accredited hospitals had a significantly higher probability of receiving care according to clinical guideline recommendations than patients at partially accredited hospitals across conditions (individual measure: adjusted odds ratio (OR) = 1.20, 95% CI: 1.01-1.43, all-or-none: adjusted OR = 1.27, 95% CI: 1.02-1.58). For five of the six included conditions there were an association; the pattern appeared particular strong among patients with acute stroke and hip fracture (all-or-none; acute stroke: adjusted OR = 1.39, 95% CI: 1.05-1.83, hip fracture: adjusted OR = 1.57, 95% CI: 1.00-2.49). CONCLUSION: High compliance with accreditation standards was associated with a higher level of evidence-based hospital care in Danish hospitals.

Predictors of the effectiveness of accreditation on hospital performance: A nationwide stepped-wedge study.

OBJECTIVE: To identify predictors of the effectiveness of hospital accreditation on process performance measures. DESIGN: A multi-level, longitudinal, stepped-wedge, nationwide study. PARTICIPANTS: All patients admitted for acute stroke, heart failure, ulcers, diabetes, breast cancer and lung cancer at Danish hospitals. INTERVENTION: The Danish Healthcare Quality Programme that was designed to create a framework for continuous quality improvement. MAIN OUTCOME MEASURE(S): Changes in week-by-week trends of hospitals' process performance measures during the study period of 269 weeks prior to, during and post-accreditations. Process performance measures were based on 43 different processes of care obtained from national clinical quality registries. Analyses were stratified according to condition, type of care (i.e. treatment, diagnostics, secondary prevention and patient monitoring) and hospital characteristics (i.e. university affiliation, location, size, experience with accreditation and accreditation compliance). RESULTS: A total of 1 624 518 processes of care were included. The impact of accreditation differed across the conditions. During accreditation, heart failure and breast cancer showed less improvement than other disease areas. Across all conditions, diagnostic processes improved less rapidly than other types of processes. However, after stratifying the data by hospital characteristics, process performance measures improved more uniformly. In respect of the measures that had an unsatisfactory level of quality, the processes related to diabetes, diagnostics and patient monitoring all responded to accreditation and showed an increased improvement during the preparatory work. CONCLUSION: Hospital characteristics were not found to be predictors for the effects of accreditation, whereas conditions and types of care to some extent predicted the effectiveness.

Dual leadership in a hospital practice.

Purpose Despite the practice of dual leadership in many organizations, there is relatively little research on the topic. Dual leadership means two leaders share the leadership task and are held jointly accountable for the results of the unit. To better understand how dual leadership works, this study aims to analyse three different dual leadership pairs at a Danish hospital. Furthermore, this study develops a tool to characterize dual leadership teams from each other. Design/methodology/approach This is a qualitative study using semi-structured interviews. Six leaders were interviewed to clarify how dual leadership works in a hospital context. All interviews were transcribed and coded. During coding, focus was on the nine principles found in the literature and another principle was found by looking at the themes that were generic for all six interviews. Findings Results indicate that power balance, personal relations and decision processes are important factors for creating efficient dual leaderships. The study develops a categorizing tool to use for further research or for organizations, to describe and analyse dual leaderships. Originality/value The study describes dual leadership in the hospital context and develops a categorizing tool for being able to distinguish dual leadership teams from each other. It is important to reveal if there are any indicators that can be used for optimising dual leadership teams in the health-care sector and in other organisations.

Blood sampling - Two sides to the story.

This study aimed to investigate why there is variability in taking blood. A multi method Pilot study was completed in four National Health Service Scotland hospitals. Human Factors/Ergonomics principles were applied to analyse data from 50 observations, 15 interviews and 12-months of incident data from all Scottish hospitals. The Functional Resonance Analysis Method (FRAM) was used to understand why variability may influence blood sampling functions. The analysis of the 61 pre blood transfusion sampling incidents highlighted limitations in the data collected to understand factors influencing performance. FRAM highlighted how variability in the sequence of blood sampling functions and the number of practitioners involved in a single blood sampling activity was influenced by the working environment, equipment, clinical context, work demands and staff resources. This pilot study proposes a realistic view of why blood sampling activities vary and proposes the need to consider the system's resilience in future safety management strategies.

Improvement in quality of hospital care during accreditation: A nationwide stepped-wedge study.

OBJECTIVE: To assess changes over time in quality of hospital care in relation to the first accreditation cycle in Denmark. DESIGN, SETTING AND PARTICIPANTS: We performed a multi-level, longitudinal, stepped-wedge, nationwide study of process performance measures to evaluate the impact of a mandatory accreditation programme in all Danish public hospitals. Patient-level data (n = 1 624 518 processes of care) on stroke, heart failure, ulcer, diabetes, breast cancer and lung cancer care were obtained from national clinical quality registries. INTERVENTION: The Danish Healthcare Quality Programme was introduced in 2009, aiming to create a framework for continuous quality improvement. MAIN OUTCOME: Changes in week-by-week trends of hospital care during the study period of 269 weeks prior to, during and post-accreditation. RESULTS: The quality of hospital care improved over time throughout the study period. The overall positive change in trend odds ratio (OR) = 1.002 per week; 95% confidence interval (CI: 0.997-1.006) observed when comparing the period during accreditation with the period prior to accreditation was not significant. However, when restricting the analyses to processes of care where the performance did not meet target values for satisfactory quality prior to accreditation, we found a significant positive change in trend (OR = 1.006 per week; 95% CI: 1.001-1.011). When comparing the post-accreditation period with the period during accreditation, we found a significantly reduced trend (OR = 0.994 per week; 95% CI: 0.988-0.999), indicating the improvement in quality of care continued but at a lower rate than during accreditation. CONCLUSION: These findings support the hypothesis that hospital accreditation leads to improvements in patient care.

Is compliance with hospital accreditation associated with length of stay and acute readmission? A Danish nationwide population-based study.

OBJECTIVE: To examine the association between compliance with hospital accreditation and length of stay (LOS) and acute readmission (AR). DESIGN: A nationwide population-based follow-up study from November 2009 to December 2012. SETTING: Public, non-psychiatric Danish hospitals. PARTICIPANTS: In-patients admitted with one of 80 diagnoses. INTERVENTION: Accreditation by the first version of The Danish Healthcare Quality Programme. Using an on-site survey, surveyors assessed the level of compliance with the standards. The hospital was awarded either fully (n = 11) or partially accredited (n = 20). MAIN OUTCOME MEASURES: LOS including transfers between hospitals and all-cause AR within 30 days after discharge. The Cox Proportional Hazard regression was used to compute hazard ratios (HRs) adjusted for potential confounding factors and cluster effect at hospital level. RESULTS: For analyses of LOS, 275 589 in-patients were included of whom 266 532 were discharged alive and included in the AR analyses. The mean LOS was 4.51 days (95% confidence interval (CI): 4.46-4.57) at fully and 4.54 days (95% CI: 4.50-4.57) at partially accredited hospitals, respectively. After adjusting for confounding factors, the adjusted HR for time to discharge was 1.07 (95% CI: 1.01-1.14). AR within 30 days after discharge was 13.70% (95% CI: 13.45-13.95) at fully and 12.72% (95% CI: 12.57-12.86) at partially accredited hospitals, respectively, corresponding to an adjusted HR of 1.01 (95% CI: 0.92-1.10). CONCLUSION: Admissions at fully accredited hospitals were associated with a shorter LOS compared with admissions at partially accredited hospitals, although the difference was modest. No difference was observed in AR within 30 days after discharge.

Where the rubber meets the road: using FRAM to align work-as-imagined with work-as-done when implementing clinical guidelines.

BACKGROUND: Uptake of guidelines in healthcare can be variable. A focus on behaviour change and other strategies to improve compliance, however, has not increased implementation success. The contribution of other factors such as clinical setting and practitioner workflow to guideline utilisation has recently been recognised. In particular, differences between work-as-imagined by those who write procedures, and work-as-done-or actually enacted-in the clinical environment, can render a guideline difficult or impossible for clinicians to follow. The Functional Resonance Analysis Method (FRAM) can be used to model workflow in the clinical setting. The aim of this study was to investigate whether FRAM can be used to identify process elements in a draft guideline that are likely to impede implementation by conflicting with current ways of working. METHODS: Draft guidelines in two intensive care units (ICU), one in Australia and one in Denmark, were modelled and analysed using FRAM. The FRAM was used to guide collaborative discussion with healthcare professionals involved in writing and implementing the guidelines and to ensure that the final instructions were compatible with other processes used in the workplace. RESULTS: Processes that would have impeded implementation were discovered early, and the guidelines were modified to maintain compatibility with current work processes. Missing process elements were also identified, thereby, avoiding the confusion that would have ensued had the guideline been introduced as originally written. CONCLUSIONS: Using FRAM to reconcile differences between work-as-imagined and work-as-done when implementing a guideline can reduce the need for clinicians to adjust performance and create workarounds, which may be detrimental to both safety and quality, once the guideline is introduced.

Accreditation and improvement in process quality of care: a nationwide study.

OBJECTIVE: To examine whether performance measures improve more in accredited hospitals than in non-accredited hospital. DESIGN AND SETTING: A historical follow-up study was performed using process of care data from all public Danish hospitals in order to examine the development over time in performance measures according to participation in accreditation programs. PARTICIPANTS: All patients admitted for acute stroke, heart failure or ulcer at Danish hospitals. INTERVENTION: Hospital accreditation by either The Joint Commission International or The Health Quality Service. MEASUREMENTS: The primary outcome was a change in opportunity-based composite score and the secondary outcome was a change in all-or-none scores, both measures were based on the individual processes of care. These processes included seven processes related to stroke, six processes to heart failure, four to bleeding ulcer and four to perforated ulcer. RESULTS: A total of 27 273 patients were included. The overall opportunity-based composite score improved for both non-accredited and accredited hospitals (13.7% [95% CI 10.6; 16.8] and 9.9% [95% 5.4; 14.4], respectively), but the improvements were significantly higher for non-accredited hospitals (absolute difference: 3.8% [95% 0.8; 8.3]). No significant differences were found at disease level. The overall all-or-none score increased significantly for non-accredited hospitals, but not for accredited hospitals. The absolute difference between improvements in the all-or-none score at non-accredited and accredited hospitals was not significant (3.2% [95% -3.6:9.9]). CONCLUSIONS: Participating in accreditation was not associated with larger improvement in performance measures for acute stroke, heart failure or ulcer.

Resilient health care: turning patient safety on its head.

The current approach to patient safety, labelled Safety I, is predicated on a 'find and fix' model. It identifies things going wrong, after the event, and aims to stamp them out, in order to ensure that the number of errors is as low as possible. Healthcare is much more complex than such a linear model suggests. We need to switch the focus to what we have come to call Safety II: a concerted effort to enable things to go right more often. The key is to appreciate that healthcare is resilient to a large extent, and everyday performance succeeds much more often than it fails. Clinicians constantly adjust what they do to match the conditions. Facilitating work flexibility, and actively trying to increase the capacity of clinicians to deliver more care more effectively, is key to this new paradigm. At its heart, proactive safety management focuses on how everyday performance usually succeeds rather than on why it occasionally fails, and actively strives to improve the former rather than simply preventing the latter.

Compliance with hospital accreditation and patient mortality: a Danish nationwide population-based study.

OBJECTIVE: To examine the association between compliance with hospital accreditation and 30-day mortality. DESIGN: A nationwide population-based, follow-up study with data from national, public registries. SETTING: Public, non-psychiatric Danish hospitals. PARTICIPANTS: In-patients diagnosed with one of the 80 primary diagnoses. INTERVENTION: Accreditation by the first version of The Danish Healthcare Quality Programme for hospitals from 2010 to 2012. Compliance were assessed by surveyors on an on-site survey and awarded the hospital as a whole; fully (n = 11) or partially accredited (n = 20). A follow-up activity was requested for partially accredited hospitals; submitting additional documentation (n = 11) or by having a return-visit (n = 9). MAIN OUTCOME MEASURES: All-cause mortality within 30-days after admission. Multivariable logistic regression was used to compute odds ratios (ORs) for 30-day mortality adjusted for six confounding factors and for cluster effect at hospital level. RESULTS: A total of 276 980 in-patients were identified. Thirty-day mortality risk for in-patients at fully (n = 76 518) and partially accredited hospitals (n = 200 462) was 4.14% (95% confidence interval (CI):4.00-4.28) and 4.28% (95% CI: 4.20-4.37), respectively. In-patients at fully accredited hospitals had a lower risk of dying within 30-days after admission than in-patients at partially accredited hospitals (adjusted OR of 0.83; 95% CI: 0.72-0.96). A lower risk of 30-day mortality was observed among in-patients at partially accredited hospitals required to submit additional documentation compared with in-patients at partially accredited hospitals requiring a return-visit (adjusted OR 0.83; 95% CI: 0.67-1.02). CONCLUSION: Admissions at fully accredited hospitals were associated with a lower 30-day mortality risk than admissions at partially accredited hospitals.