FEATURE
Escalation of Care in Surgery A Systematic Risk Assessment to Prevent Avoidable Harm in Hospitalized Patients Maximilian Johnston, MB BCh, MRCS,∗ Sonal Arora, PhD, MRCS,∗ Oliver Anderson, MSc, MRCS,∗ Dominic King, MB ChB, MRCS,∗ † Nebil Behar, MSc, FRCS,‡ and Ara Darzi, MD, FRCS, FACS∗ † Objective: To systematically risk assess and analyze the escalation of care process in surgery so as to identify problems and provide recommendations for intervention. Background: The ability to escalate care appropriately when managing deteriorating patients is a hallmark of surgical competence and safe postoperative care. Healthcare-Failure-Mode-Effects-Analysis (HFMEA) is a methodology adapted from safety-critical industries, which allows for hazardous process failures to be prospectively identified and solutions to be recommended. Methods: Forty-two hours of ethnographic observations on surgical wards in 3 London hospitals (phase 1) formed the basis of an escalation process diagram. A risk-assessment survey identified failures associated with process steps and attributed hazard scores (phase 2). Patient safety and clinical risk experts validated hazard scores through a group consensus meeting (phase 3). Hazardous failures were taken forward to multidisciplinary HFMEA where cause analysis was applied and interventions were recommended (phase 4). Results: Observations identified 33 steps in the escalation process. The riskassessment survey (30 surgical staff members, 100% response) and expert consensus group identified 18 hazardous failures associated with these steps. The HFMEA team identified 3 adequately controlled failures; therefore, 15 were subjected to cause analysis. Outdated communication technology, understaffing, and hierarchical barriers were identified as root causes of failure. Participants recommended interventions based on these findings including defined escalation protocols, human factors education, enhanced communication technology, and improved clinical supervision. Conclusions: Failures in the escalation process amenable to intervention were systematically identified. This mapping of the escalation process will allow tailored interventions to enhance surgical training and patient safety. Keywords: escalation of care, health care failure mode effects analysis, patient safety, risk assessment, surgery (Ann Surg 2015;261:831–838)
T
he principles of quality and patient safety should be at the forefront of modern surgical care. Despite this, the rate of adverse events in surgery can be as high as 12.9%.1 In an attempt to address these concerns, there has been an increased drive to understand conFrom the ∗ Centre for Patient Safety and †Centre for Health Policy, Department of Surgery and Cancer, Imperial College London; and ‡Chelsea and Westminster Hospital NHS Foundation Trust, London, UK. Disclosure: This study did not require formal ethical approval. It was registered and approved as a quality improvement project. Informed written consent was gained from all participants. All authors declare no conflicts of interest. Johnston, Arora, King, Anderson, and Darzi are affiliated with the Centre for Patient Safety and Service Quality (www.cpssq.org) at Imperial College, which is funded by the National Institute for Health Research, UK. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.annalsofsurgery.com). Reprints: Max Johnston, MB BCh, MRCS, Department of Surgery and Cancer, St Mary’s Hospital, Imperial College London, W2 1NY, UK. E-mail: [email protected]. C 2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright ISSN: 0003-4932/14/26105-0831 DOI: 10.1097/SLA.0000000000000762
Annals of Surgery r Volume 261, Number 5, May 2015
tributory factors in the operating room (OR)2 and intensive care unit (ICU) so as to develop and implement evidence-based interventions.3 Despite significant improvements, these studies have neglected to explore the postoperative period on the ward, especially in the setting of acute physiological deterioration. Sound postoperative care, to include timely recognition and management of complications, is an important aspect of patient safety4 and can significantly reduce mortality when performed to the highest standard.5 One of the most critical elements of postoperative care is the ability to recognize a deteriorating patient and communicate this to a more senior team member able to provide definitive treatment. This type of communication will typically occur in academic settings where residents are training; communication pathways may be different in other health care settings. The recognition and communication of patient deterioration to a senior colleague can be termed escalation of care (EOC).6,7 In academic health care settings all over the world, it is typically a junior doctor who is the first point of contact for nursing staff when a postoperative patient becomes unstable. This relatively inexperienced clinician must make a rapid assessment of the patient before deciding whether to ask a senior colleague for help. This process is fraught with difficulty; studies suggest that failures in escalation of care are ubiquitous with at least 1 in 5 patients experiencing delays in urgent treatment.8,9 The effect of these deficiencies can lead to the “failure to rescue” of a patient in need, a potentially catastrophic event. In fact, evidence suggests that patients experiencing delayed escalation have an increased risk of morbidity10 and mortality.11,12 Further exploration of the escalation of care process is therefore warranted to extend patient safety gains garnered in the OR and ICU onto postoperative surgical wards. Initial studies evaluating escalation of care have only explored the role of communication failures in this pathway.13 Although important, this approach has neglected to comprehensively consider the human, technical, and patient factors involved in this safety-critical process. More detailed investigation is required—especially in light of calls from the Joint Commission in the United States14 and the Francis Report in the United Kingdom15 mandating the need for more proactive risk assessments of health care processes. A Healthcare Failure Mode and Effects Analysis (HFMEA) is a systematic method of conducting a proactive risk assessment. It allows for the most hazardous failures in a process to be identified and prioritized, so that interventions can be targeted at the most appropriate point. HFMEA incorporates a multistage approach, which utilizes the expertise of an interprofessional team to develop process flow-charts, hazard scores, and decision trees, which define areas of potential failure where the patient is most susceptible to avoidable harm. Originally used by the US Army,16 this research method has also been employed effectively by the aviation and motor industries.17,18 The HFMEA process is a process which enables prospective identification of the severity, detectability, and frequency of potential risks to patient care.19 The standard HFMEA process involves the following steps: define the HFMEA process, assemble the team, graphically describe the process, conduct a risk assessment, discuss areas for www.annalsofsurgery.com | 831
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Johnston et al
intervention, and feedback recommendations to decision-makers. The HFMEA process significantly differs from root cause analysis (RCA) in its timing of data collection. HFMEA aims to prospectively identify failures within a process to allow prevention before their occurrence. However, RCA involves retrospective analysis of events leading to an adverse outcome using case note analysis and interview techniques. The benefit of using HFMEA rather than RCA methodology is similar to a comparison of a prospective versus retrospective clinical study. As such HFMEA is a powerful tool for patient safety due to its proactive nature allowing identification and prevention of failures before they occur. Within surgery, risk assessments using this approach have identified problems in hand hygiene and medication delivery.20 Others have conducted risk assessments of the postoperative handoff process leading to implementation of a protocol, which has significantly improved handoffs from the OR to the post-anesthesia care unit.21 However, no previous studies have conducted a proactive risk assessment of the escalation of care process—despite its importance in preventing potentially avoidable morbidity and mortality. Not surprisingly therefore, interventions have been developed rather haphazardly with little success in improving either the process of escalation or patient outcomes.22,23 The aims of this study were to systematically risk assess and analyze the escalation of care process so as to identify areas of failure and avoidable patient harm, understand their causes, and make recommendations to improve patient safety on the surgical ward.
METHODS This study used the validated HFMEA technique modified by the Veterans Affairs National Center for Patient Safety.19 According to this criterion for HFMEA methodology, this study was conducted in 4 phases—phase 1: ethnographic observation of escalation episodes on surgical wards; phase 2: a risk assessment survey of the EOC process; phase 3: validation of hazard scores by patient safety and clinical risk experts through a consensus group meeting; phase 4: modified HFMEA to identify root causes of potentially hazardous failures and provide recommendations for intervention to prevent avoidable harm (see Fig. 1).
Setting This study was carried out in North-West London and included 6 general surgery wards in 4 hospitals (academic/teaching and community). Formal ethical approval was sought but not deemed necessary by the local ethics committee, so the study was registered as a quality improvement project.
Phase 1: Ethnographic Observations Ethnographic observations were conducted to ensure comprehensive capture of all of the steps in the escalation of care process. Ethnographic observation allows researchers to witness events in realtime and in-situ with minimal interference. In this study, the observers were external to the organization where the study took place and, as such, were not known to staff members. Ward staff were notified that observers would be conducting an evaluation of clinical communication on the wards and had agreed to the presence of observers before commencement of the study. Observational work also gives the researcher insight into how teams and organizations work; this is important for escalation of care as it is a collaborative rather than individual process.24 Two researchers with a background in general surgery and patient safety (M.J. and S.A.) observed and kept detailed ethnographic field notes on episodes of patient deterioration and escalation on general surgery wards and surgical high-dependency units for 42 hours. Fortunately, there were no serious situations that were observed where the patient was critically unwell and care was not 832 | www.annalsofsurgery.com
escalated. Had such a situation arisen, the observers (who are both medically qualified) would have alerted the appropriate medical personnel on duty for that ward. Patient safety is always the priority over the research study as per ethical requirements. During this time, 28 escalation events were observed. Each stakeholder (nursing assistant, nurse, intern, junior resident, senior resident, and attending surgeon) in the process worked on the postoperative surgery ward and was observed for 2 separate 3.5-hour sessions, one by each observer (a total of 42 hours). As per the study protocol, all observations were conducted in 3.5-hour aliquots. If an event of interest was occurring at the 3.5-hour mark, the observer remained until that event had been concluded. Each stakeholder was observed for 1 session during normal working hours (8 AM–5 PM) and 1 session outside of normal working hours (5 PM–8 AM). Although a single stakeholder was the primary focus of each observation session, the interprofessional nature of escalation of care meant that multiple stakeholders were encountered during each session (eg, nurse contacting resident regarding a deteriorating patient). The observers performed the primary 3.5-hour observation session together to ensure consistency of data capture in future sessions. These sessions took place between January and March 2013. No data were collected that could identify staff members, and as such, no evaluation of individual staff members was conducted. All data were anonymized at the point of collection rendering any synthesis of individual performance measures impossible. From these observations, a comprehensive flow diagram of the surgical escalation process was produced with input from all investigators and the literature6,25,26 (see Supplemental Digital Content Appendix A, available at http://links.lww.com/SLA/A561).
Phase 2: Risk Assessment Survey This phase was conducted to provide quantitative data on stakeholders’ views of the level of risk associated with each step of the escalation process. Data were collected with a detailed risk assessment survey based on the observations from phase 1 in 2 London hospitals (hospitals 1 and 3 in Table 1). All stakeholder groups in the escalation process completed the survey (n = 30). From the first hospital, there were 2 attending surgeons, 7 surgical residents, 2 surgical interns, 3 nurses, and 2 nursing assistants from the postoperative surgery ward. From the second hospital, there were 2 attending surgeons, 4 surgical residents, 3 surgical interns, and 5 nurses from the postoperative surgery ward. Participants used a 4-point scale to rate the potential consequences of failure for each step in the escalation process according to its likelihood of occurrence, potential harm associated with failure to escalate, and the detectability of a problem being identified before patient harm. These 3 ratings were multiplied together to calculate hazard scores for each process step (see Table 2).
Phase 3: Hazard Score Validation The hazard scores from the risk assessment were then validated by a group of patient safety and clinical risk experts. There were 2 consensus group sessions involving 5 participants each of that occurred 2 weeks after the initial risk assessment to allow the surveys to be analyzed in the intervening period. Participants were recruited from the Centre for Patient Safety and Service Quality at Imperial College. The participants in each group were a patient safety manager, clinical risk director, surgeon, physician, and critical care nurse. A facilitator (M.J. or O.A.) was also present but did not express any opinions; their role was to structure discussions and act as a timekeeper. Each process step and hazard score was discussed within the group, and consensus was reached before moving on to the next step. To allow prioritization, a hazard score threshold of 50% was chosen on the basis of previous guidelines for HFMEA methodology in surgical research.20,25 Process steps exceeding this threshold were C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Escalation of Care in Surgery
FIGURE 1. A diagram showing the steps in the method for modified HFMEA. TABLE 1. Details of the Hospitals Participating in the HFMEA
Hospital
Type of Hospital
No. Beds
No. Wards Observed
Hospital 1
Academic/Teaching
418
2
Hospital 2
Academic/Teaching
500
1
Hospital 3
Academic/Teaching
430
1
Hospital 4
Community
463
2
Surgical Specialties on Ward(s)
No. Patients on a Typical Ward
General Vascular General Urology Emergency General Vascular Urology General
18 26 28 32
Resident Make-up per Clinical Team
Surgical HDU (No. Beds)
1xPGY1 1xPGY2-4 1xPGY1 1xPGY2-4 2xPGY1 1xPGY2 1xPGY-1 1xPGY2-4 1xSenior resident
General (4) Vascular (6) No Yes (4) Yes (5)
HDU indicates high dependency unit.
TABLE 2. Hazard Scoring System for Escalation of Care Process Failures Score∗ Category Severity† Frequency Detectability
1
2
3
4
No adverse outcome 1 per year High
Increased stay 1 per month Moderate
Disability 1 per week Low
Death 1 per day Remote
∗
The 3 scores were multiplied together to give a hazard score (maximum score = 64). †The severity was interpreted as the worst possible potential outcome from each failure mode.
classified as hazardous (patient at risk of avoidable harm) and carried forward to the modified HFMEA analysis.
Phase 4: Modified HFMEA Formal modified HFMEA was used to confirm the failures and hazard scores associated with each of the hazardous process steps. These failures and hazard scores were generated on the basis of the literature20,21 and observations in phase 1, the risk assessment in phase 2, and expert opinion in phase 3 so as to provide a triangulated approach ensuring detection of all potential failures in the process. An interprofessional HFMEA team was assembled consisting of a patient, a nursing assistant, and 2 nurses from a postoperative C 2014 Wolters Kluwer Health, Inc. All rights reserved.
surgery ward, 2 surgical interns, 4 surgical residents, and 2 attending surgeons. The patient was recruited from the postoperative outpatient clinic having had major abdominal surgery 3 weeks previously. The team was supplemented by a patient safety researcher (M.J.) experienced in the HFMEA process, who acted as a facilitator and ensured that participants were given equal opportunity to air their views. After explanation of the HFMEA procedure, the team checked the process diagram to ensure completeness. After this, the hazard score for each failure mode was reviewed and discussed using decision-tree analysis. Here each failure mode was assessed in terms of its potential detectability and whether there were preexisting control measures in place, were it to occur. Hazardous failures, which lacked control www.annalsofsurgery.com | 833
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Johnston et al
measures and were not easily detectable, were prioritized for further action. To eliminate the potential for participants with leadership qualities to dominate, each participant individually, and privately, scored each failure before discussion. Final ratings were reached through additional consensus agreement; if consensus was not possible, the failure was automatically included in further analysis. This ensured all potentially hazardous failures were reviewed to minimize selection bias. The HFMEA team explored each potential cause of process failure using the framework for incident investigation from the London protocol RCA method.27 Potential causes were categorized into organizational, environmental, patient, staff (individual), and team- and task-related factors. Serial causes for each failure were suggested until a definitive or root cause within each relevant category of the London protocol could be identified. This meant that each failure could have more than 1 cause if the participants assigned causes to multiple categories within the protocol. Identification of the underlying causes of a failure allowed synthesis of recommendations aimed at preventing the failure in the future. Suggested interventions were then fed back to managers and clinical directors through a clinical governance meeting and report, potentially facilitating implementation and quality improvement. Finally, each participant was asked to fill out a questionnaire describing their perceptions of the success of the research method and their knowledge of patient safety.
RESULTS Phase 1: Ethnographic Observations Ethnographic observations of surgical patients and staff led to the identification of 33 core steps involved in a successful escalation of care process. Additional steps could have been added to this depending on whether a junior doctor escalated directly to an attending surgeon or via a senior resident (see Supplemental Digital Content Appendix A, available at http://links.lww.com/SLA/A561). As this would only involve a repeat cycle of assessments (resident review followed by attending review), these steps were combined and considered together. The observation phase of this study revealed escalation to be a stepwise process prone to failure until a patient received definitive management to enable recovery (see Figure 2).
protocols, so a total of 15 (45%) steps were carried forward to cause analysis and solution synthesis (see Table 3). The 3 controlled failures were the nurse failing to assess the patient correctly, the nurse failing to adhere to the escalation protocol correctly, and the doctor failing to attend the patient. The control measures identified by participants were the use of vital signs measurement (which participants identified as objective physiological evidence of deterioration), the requirement for nurse assistants and junior nurses to check all abnormal vital signs with the senior nurse for the ward, and the directive that empowered nurses to contact senior residents or attendings if more junior doctors were not receptive to their request for assistance. The causes and recommended solutions to hazardous yet uncontrolled failures are described later. These are described according to the sequence of health care professional involvement when a patient becomes unwell and typically include nursing, junior doctor, and senior doctor stages.
Process Steps Involving Nursing Staff Causes There were 7 steps with hazardous, uncontrolled failures to consider in the nursing stage of the escalation process (see Table 4). These were the patient failing to inform the nursing assistant they feel unwell, failure of the nurse to attend the patient, failure to notice the patient is deteriorating, failure to measure the vital signs correctly, failure to document the vital signs correctly, failure to inform the senior nurse that the patient is deteriorating, and failure to inform the junior doctor about the deterioration. Participants felt that clinical understaffing was the principal cause of all these failures. They explained that a greater (and therefore more visible) staff presence would allow patients to raise concerns about their well-being more easily. Interim causes of these failures included patient alarm bells not working, inexperienced/agency staff, patients with dementia or communication difficulties, and transcription errors when recording vital signs. An interim cause means a cause that contributed to a failure but was not felt to be the root cause of that failure by participants. Regarding communication between the nurse and junior doctor; the participants felt that both human and technological factors were contributory. Human factors included the nurses fearing criticism by junior doctors, technological factors principally revolved around the limitations of the pager system.
Recommendations Phases 2 and 3: Risk Assessment Survey and Validation of Hazard Scores Of the 33 process steps involved in escalation of care, the risk assessment participants identified a total of 17 hazardous failures that exceeded the hazard score threshold. The patient safety and clinical risk experts identified 1 additional hazardous failure via consensus agreement, this was the patient informing the nursing assistant they feel unwell, rather than going straight to the nurse. These 18 steps were taken forward to discussion in the modified HFMEA.
Phase 4: Modified HFMEA The steps with associated hazardous failures within the process were presented to the modified HFMEA team. The HFMEA team then repeated the hazard scoring, the lowest possible score was 1 and the highest possible score was 64. Ratings applied by the HFMEA team ranged from 27 to 48. The hazard score threshold was higher than 26; therefore, all of the 18 process steps identified as hazardous in the risk assessment were confirmed by the HFMEA team. During the discussion, participants decided that 3 of these 18 hazardous steps had failures that were already adequately controlled with preexisting 834 | www.annalsofsurgery.com
The recommendations made by participants can be seen in full in Table 4. The main recommendations to improve the nursing steps of the process were as follows:
r Increased nurse: patient ratios r Investment and training in the use of electronic vital sign recording and documentation systems
r Removal of hierarchical barriers through development of an improved escalation protocol, and;
r Increased use of smartphone technology. Process Steps Involving Junior Doctors Causes There were 7 steps with hazardous, uncontrolled failures in the stage involving junior doctors assessing/managing a deteriorating patient (see Table 4). These were failure to take an adequate history, failure to examine the patient thoroughly, failure to check the drug/fluid charts or case notes, failure to commence the correct initial treatment, and failure to inform a senior doctor of patient deterioration. The causes of these failures included hierarchical and language barriers, senior staff being busy in the OR and unable to answer a C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Escalation of Care in Surgery
FIGURE 2. A diagram showing the pathway to successful escalation of care. TABLE 3. Details of Steps Taken Forward to HFMEA Process Phase
Steps With Hazardous Failures
Mean Hazard Score∗
Steps Included in HFMEA
Controlled Steps
Steps Analyzed in HFMEA
Nursing Junior doctor Senior doctor Total (mean)
16 14 3 33
26.8 25.1 28 (26.6)
9 8 1 18
2 1 0 3
7 7 1 15
∗
Threshold for analysis within HFMEA > 26.
pager, understaffing, charts being poorly designed or unavailable on the ward, inexperience of junior doctors, illegible clinical notes, and poor availability of equipment.
Recommendations The recommendations made by participants can be seen in full in Table 4. The main recommendations to improve the junior doctor steps of the process were as follows:
Recommendations The recommendations made by participants can be seen in full in Table 4. The main recommendations to improve the senior doctor steps of the process were as follows:
r Development of a clear escalation protocol r Production of a clinical guideline defining appropriate levels of
crease reliance on agency staff
care according to patient diagnosis, physiological parameters, and predictive scoring systems, and; r Improvement of resource and staffing management to ensure sufficient access to the OR and surgical ICU beds for unstable patients.
system
Evaluation of the Chosen Research Method
r Recruitment of more permanent nursing and medical staff to der Development and implementation of an electronic medical records r Encourage senior surgeons to be more proactive (meaning that the responsibility for initiating contact would not be the sole responsibility of juniors and nurses) r Educate junior doctors about the importance of prompt escalation of care, and; r Develop a software platform with integrated demographics, pathology, radiology, and vital signs to allow prompt escalation in the presence of patient deterioration.
Process Steps Involving Senior Doctors Causes There was 1 step with hazardous, uncontrolled failures in the senior doctor phase of the process (see Table 4), which was the senior doctor failing to arrange definitive management. Participants felt this could be due to the senior doctor failing to contact an appropriate colleague or failing to arrange appropriate patient transfer to the OR or ICU. The root causes of these failures were felt to be fear when faced with contacting the attending surgeon or intensivist on duty and the presence of logistical barriers. These barriers included a lack of intensive care beds and uncertainty regarding the appropriate level of care for each patient. The participants also felt that only having a single OR available for emergent surgeries out of normal working hours was a major safety concern. C 2014 Wolters Kluwer Health, Inc. All rights reserved.
The participants felt that the modified HFMEA process was easy to interpret (75%) and had increased their awareness of patient safety (83%). They expressed enthusiasm about participating in further sessions (75%) and would also recommend participation to a colleague (83%).
DISCUSSION This study represents the first systematic risk assessment of the entire process of recognizing and responding to an unstable patient on a surgical ward. The steps of the escalation of care process were carefully mapped and analyzed to identify the most hazardous failure points where the patient was most susceptible to harm. These hazardous failures were then subjected to detailed cause analysis, which enabled technological innovation, clear escalation protocols, and staff recruitment to be identified as key recommendations for improving patient safety in surgical inpatients. This study exposes the significant vulnerabilities faced by a patient on a surgical ward—at a point when high-quality care is most required. It demonstrates that failures can occur at any step of the escalation of care pathway, highlighting poor resilience and reliability in the overall system. This is in contrast to other high-risk, safety critical industries such as aviation and the military where redundancy mechanisms are incorporated into the system to compensate for a www.annalsofsurgery.com | 835
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Johnston et al
TABLE 4. Causes of Failures in Escalation of Care and Recommendations for Improvement Process Phase and Step With Hazardous Failures Nursing Steps Patient fails to inform nursing assistant they feel unwell Nurse fails to notice that patient is unwell Nurse fails to attend to patient promptly Nurse fails to measure vital signs correctly Nurse fails to document vital signs correctly Nurse fails to inform senior nurse that patient is unwell Nurse fails to contact junior doctor successfully
Junior Doctor Steps Junior doctor fails to take accurate history
Junior doctor fails to complete thorough examination
Causes of Failure Patient decides not to inform nursing assistant Patient cannot inform nursing assistant Decides patient is not unwell Fails to attend to patient promptly
Root Cause Poor knowledge of patient personality Bell not supplied or out of reach Lack of specialty training Use of agency staff High workload
Fails to measure vital signs correctly Documents vital signs incorrectly on chart Fails to inform junior doctor
Lack of appropriate technology
Decides not to contact junior doctor Pages doctor but no response Contacts but junior doctor will not attend
Hierarchical barriers Lack of confidence Previous trouble with pager system Extension engaged when returning page Unable to locate extension Busy with a patient Doctor receives multiple pages at once High workload Poor judgment
Fails to complete history Unable to communicate with patient
Lack of specialty training High workload Agency staff unfamiliar with hospital Language barriers
Fails to complete examination
Patient positioning
Transcribing error Unclear escalation protocol
Equipment not available Completes incorrect examination
High workload Lack of specialty training
Fails to review chart
Chart not on ward
Reviews chart incorrectly Reviews incorrect chart
Illegible chart Human error
Junior doctor fails to review fluid chart
Fails to review chart
Chart not on ward
Junior doctor fails to review patient case notes
Unable to fully review case notes
Not completed accurately Large files for long-term patients
Junior doctor fails to initiate correct treatment
Selects incorrect treatment
Illegible Lack of specialty training
Junior doctor fails to review drug chart
Unable to commence correct treatment
836 | www.annalsofsurgery.com
Equipment not available
Recommendations More regular patient rounds Nominate ward “champions” Investment and training in detection of deterioration for nurses Recruit more permanent staff Recruit more permanent staff Invest in bedside electronic vital signs charts Invest in bedside electronic vital signs charts Develop and publicize a clear escalation protocol Clear escalation protocol Teach handover skills (eg, SBAR∗ ) to nursing staff Use of mobile telephones Develop electronic patient platforms allowing transfer of patient information Use of mobile telephones Use messaging services to send information Use messaging services to send information Recruit more permanent staff Teach prioritization skills Investment and training in detection of deterioration for junior doctors Recruit more permanent staff Recruit more permanent staff Mobile phones able to call interpreter from bedside Increase ward staff numbers and manual handling Nominate ward equipment “champions” Recruit more permanent staff Encourage cross-discipline co-operation Electronic chart available across site Computerized chart Bedside software with electronic chart Electronic chart available across site Education and training for staff Electronic notes system with weekly summaries Computerized notes Support courses/teaching for junior medical staff Nominate ward equipment “champions” (continues)
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Escalation of Care in Surgery
TABLE 4. (Continued) Process Phase and Step With Hazardous Failures Junior doctor fails to contact senior doctor successfully
Senior Doctor Steps Senior doctor fails to arrange definitive management
Causes of Failure
Root Cause
Decides not to contact senior doctor Attempts contact but no response Contacts but senior doctor will not attend Contacts senior but delayed response
Hierarchical barriers Busy with a patient Unable to locate extension Extension engaged when returning contact High workload Senior off-site
Clear escalation protocols Use messaging services to send information Use of mobile telephones Develop electronic patient platforms allowing transfer of patient information Recruit more permanent staff Seniors from busy specialties to remain on-site when on-call
Unable to transfer patient
No ICU beds
Set aside surgical ICU beds or a separate unit Increase OR availability for emergency surgery Physiological parameter thresholds to decide level of care Clear escalation protocol
No OR available
Fails to contact appropriate colleague
Uncertainty regarding appropriate level of care for patient Fear of Attending Surgeon or Intensivist
Recommendations
∗
SBAR is a method of structuring handover. It stands for Situation, Background, Assessment, Recommendation. ICU indicates intensive care unit; OR, operating room.
potential failure at any one point—for example, through the use of back-up behaviors, dual-tasking, and debriefing.28 The concerns and recommendations described in this study are consistent with previous studies evaluating the safety of patient care. In particular, clinical understaffing was highlighted as a significant contributor to delays in recognizing and responding to a deteriorating patient. This finding is supported by research from the United States29 and worldwide30 reporting better patient outcomes in centers with higher nurse staffing levels. Staffing levels are also a focus for the Agency for Healthcare Research and Quality and the Joint Commission.31 They highlight that recruitment and retention of staff requires determined action from health care providers. The identification of communication failures leading to failed escalation of care is consistent with other work highlighting the ubiquitous nature of problems in information transfer across the surgical pathway.32 Human factors and technological failure were identified as the root causes of these communication failures. Regarding the former, this study described the importance of overcoming hierarchical barriers between junior and senior surgeons as crucial for ensuring successful escalation of care. A flattened hierarchy is a key property of a safe system7 ; recommendations were made for a clear escalation protocol to facilitate this. Others have reported that such escalation protocols can reduce ICU admission33 and mortality34 making them a cost-effective proposition, which could be readily implemented. However, if measurable improvements in patient safety are to be attained, the safety culture of the health care institution must also be improved.35,36 Recommendations on staff education and training were described as integral to this campaign so as to build capacity and capability in promoting patient safety. Evidencebased train-the-trainers courses and faculty development programs could be used to provide effective training at a large-scale, institutional level.37,38 The technology available to staff was also felt to be suboptimal. In particular, the view that the hospital pager is an outdated technology with significant flaws is also supported by the literature.39,40 The strengths of this study include the systematic manner of risk assessment that captured all the vulnerabilities in the EOC pro C 2014 Wolters Kluwer Health, Inc. All rights reserved.
cess. Specifically, this method enabled the prospective capture of all potential failures as opposed to a retrospective analysis following an adverse event, which could be limited to the factors surrounding that event. The length and depth of the observation period also allowed researchers to form a comprehensive map of EOC in surgery. The validation of process steps and associated failure modes by clinical risk and safety specialists strengthened the analysis. The inclusion of core stakeholders in the surgical care pathway meant that multiple perspectives allowed robust recommendations to be made appropriate and acceptable to end-users. The limitations to this study are present in all HFMEA processes. The subjective nature of participant’s views inevitably introduces bias; attempts were made to reduce this by ensuring group consensus at appropriate points throughout the process. In addition, the HFMEA process is built to consider process failures in isolation. If failures in the escalation process were to occur simultaneously, the recommendations made in this study may fail to control them. This study was conducted in hospitals run by the National Health Service in the United Kingdom, as such, the findings may not be generalizable to other countries or private health systems. However, the findings regarding a fear of criticism might be universally applied. This study has wide-reaching implications if adverse events in postsurgical patients are to be prevented. Breakdowns in the escalation of care process can ultimately lead to a “failure to rescue” or death after a complication. With failure to rescue rates ranging from 29% to 41%, early escalation of care is critical to prevent avoidable harm and improve outcomes.41 This study highlights the complexity of this process and, for the first time, describes the antecedents of the escalation of care process, which, if suboptimal, can lead to a failure to rescue event. In doing so, it draws attention to the numerous areas where patients on surgical wards are susceptible to avoidable harm. The design and implementation of tailored interventions, based on the recommendations of this study, should serve as a basis for future research. This can ensure that the quality of care received by a postoperative patient on a ward is commensurate to the same exacting standards seen in the OR, thereby promoting patient safety across the entire surgical pathway. www.annalsofsurgery.com | 837
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Johnston et al
CONCLUSIONS This study is the first to systematically explore potential failures in escalation of care on surgical wards using HFMEA. The identification of failures in the process and recommendation of solutions to resolve these failures will allow targeted interventions to be trialled. Future work should concentrate on development of welldesigned interventions and analysis of their effect on surgical patient outcome.
ACKNOWLEDGMENTS The authors thank all of the study participants. Their contributions were as follows: Study concept and design: All authors; Drafting/revising study protocol: M.J., S.A., O.A., D.K., A.D.; Data collection/analysis: M.J., S.A., N.B.; Writing of the first draft: M.J., S.A.; Revising draft and approval of final version: All authors.
REFERENCES 1. Sukumar S, Roghmann F, Trinh VQ, et al. National trends in hospital-acquired preventable adverse events after major cancer surgery in the USA. BMJ Open. 2013;3. 2. Haynes AB, Weiser TG, Berry WR, et al. A surgical safety checklist to reduce morbidity and mortality in a global population. N Engl J Med. 2009;360:491– 499. 3. Petrovic MA, Martinez EA, Aboumatar H. Implementing a perioperative handoff tool to improve postprocedural patient transfers. Jt Comm J Qual Patient Saf. 2012;38:135–142. 4. Ghaferi AA, Birkmeyer JD, Dimick JB. Hospital volume and failure to rescue with high-risk surgery. Med Care. 2011;49:1076–1081. 5. Glance LG, Dick AW, Meredith JW, et al. Variation in hospital complication rates and failure-to-rescue for trauma patients. Ann Surg. 2011;253:811–816. 6. Johnston M, Arora S, King D, et al. Escalation of care and failure to rescue: a multicenter, multiprofessional qualitative study. Surgery. 2014;155:989–994. 7. Mackintosh N, Sandall J. Overcoming gendered and professional hierarchies in order to facilitate escalation of care in emergency situations: the role of standardised communication protocols. Soc Sci Med. 2010;71:1683–1686. 8. Quach JL, Downey AW, Haase M, et al. Characteristics and outcomes of patients receiving a medical emergency team review for respiratory distress or hypotension. J Crit Care. 2008;23:325–331. 9. Boniatti MM, Azzolini N, Viana MV, et al. Delayed medical emergency team calls and associated outcomes. Crit Care Med. 2013;24:26–30. 10. Calzavacca P, Licari E, Tee A, et al. A prospective study of factors influencing the outcome of patients after a Medical Emergency Team review. Intens Care Med. 2008;34:2112–2116. 11. Pattison N, Eastham E. Critical care outreach referrals: a mixed-method investigative study of outcomes and experiences. Nurs Crit Care. 2012;17: 71–82. 12. Downey AW, Quach JL, Haase M, et al. Characteristics and outcomes of patients receiving a medical emergency team review for acute change in conscious state or arrhythmias. Crit Care Med. 2008;36:477–481. 13. Greenberg CC, Regenbogen SE, Studdert DM, et al. Patterns of communication breakdowns resulting in injury to surgical patients. J Am Coll Surg. 2007;204:533–540. 14. The Joint Commission announces 2014 National Patient Safety Goal. Jt Comm Perspective. 2013;33:3–4. 15. Francis R. Lessons from Stafford. Med Leg J. 2013;81:153–165. 16. Stalhandske E, DeRosier J, Wilson R, et al. Healthcare FMEA in the Veterans Health Administration. Patient Saf Qual Healthc. 2009;6:30–33. 17. Dale BG, Shaw P. Failure mode and effects analysis in the UK motor industry: a state-of-the-art study. Qual Reliab Eng Int. 1990;6:179–188.
838 | www.annalsofsurgery.com
18. Yang J, Hong-Zhong H, Li-Ping H, et al. Risk evaluation in failure mode and effects analysis of aircraft turbine rotor blades using Dempster–Shafer evidence theory under uncertainty. Eng Fail Anal. 2011;18:2084–2092. 19. DeRosier J, Stalhandske E, Bagian JP, et al. Using health care Failure Mode and Effect Analysis: the VA National Center for Patient Safety’s prospective risk analysis system. Jt Comm J Qual Improv. 2002;28:209, 248–267. 20. Anderson O, Brodie A, Vincent CA, et al. A systematic proactive risk assessment of hazards in surgical wards: a quantitative study. Ann Surg. 2012;255:1086–1092. 21. Nagpal K, Vats A, Ahmed K, et al. A systematic quantitative assessment of risks associated with poor communication in surgical care. Arch Surg. 2010;145:582–588. 22. Robb G, Seddon M. A multi-faceted approach to the physiologically unstable patient. Qual Saf Health Care. 2010;19:e47. 23. Kansal A, Havill K. The effects of introduction of new observation charts and calling criteria on call characteristics and outcome of hospitalised patients. Crit Care Resusc. 2012;14:38–43. 24. Dimick JB, Greenberg CC. Understanding gaps in surgical quality: learning to count what cannot be counted. Ann Surg. 2013;257:6–7. 25. Nagpal K, Abboudi M, Fischler L, et al. Evaluation of postoperative handover using a tool to assess information transfer and teamwork. Ann Surg. 2011;253:831–837. 26. Catchpole KR, de Leval MR, McEwan A, et al. Patient handover from surgery to intensive care: using Formula 1 pit-stop and aviation models to improve safety and quality. Paediatr Anaesth. 2007;17:470–488. 27. Vincent C, Taylor-Adams S, Stanhope N. Framework for analysing risk and safety in clinical medicine. BMJ 1998;316:1154–1157. 28. Catchpole KR, Dale TJ, Hirst DG, et al. A multicenter trial of aviation-style training for surgical teams. Journal of Patient Safety. 2010;6:180–186. 29. Aiken LH, Clarke SP, Sloane DM, et al. Hospital nurse staffing and patient mortality, nurse burnout, and job dissatisfaction. JAMA. 2002;288:1987–1993. 30. Zhu XW, You LM, Zheng J, et al. Nurse staffing levels make a difference on patient outcomes: a multisite study in Chinese hospitals. J Nurs Scholarsh. 2012;44:266–273. 31. Kane RL, Shamliyan T, Mueller C, et al. Nurse staffing and quality of patient care. Evid Rep Technol Assess (Full Rep). 2007:1–115. 32. Solvoll T, Scholl J, Hartvigsen G. Physicians interrupted by mobile devices in hospitals: understanding the interaction between devices, roles, and duties. J Med Internet Res. 2013;15:e56. 33. Adelstein BA, Piza MA, Nayyar V, et al. Rapid response systems: a prospective study of response times. J Crit Care. 2011;26:635.e11–635.e18. 34. Mitchell IA, McKay H, Van Leuvan C, et al. A prospective controlled trial of the effect of a multi-faceted intervention on early recognition and intervention in deteriorating hospital patients. Resuscitation. 2010;81:658–666. 35. Paine LA, Rosenstein BJ, Sexton JB, et al. Republished paper: assessing and improving safety culture throughout an academic medical centre: a prospective cohort study. Postgrad Med J. 2011;87:428–435. 36. Pronovost PJ, Weast B, Holzmueller CG, et al. Evaluation of the culture of safety: survey of clinicians and managers in an academic medical center. Qual Saf Health Care. 2003;12:405–410. 37. Ahmed M, Arora S, Tiew S, et al. Building a safer foundation: the Lessons Learnt patient safety training programme. BMJ Qual Saf. 2014;23:78–86. 38. Hull L, Arora S, Symons NR, et al. Training faculty in nontechnical skill assessment: national guidelines on program requirements. Ann Surg. 2013;258:370– 375. 39. Coiera E, Tombs V. Communication behaviours in a hospital setting: an observational study. BMJ. 1998;316:673–676. 40. Perna G. Study: pagers, outdated communication tech costing hospitals. Available at: http://www.healthcare-informatics.com/news-item/study-pagersoutdated-communication-tech-costing-hospitals. Accessed May 10, 2014. 41. Sheetz KH, Waits SA, Krell RW, et al. Improving mortality following emergent surgery in older patients requires focus on complication rescue. Ann Surg. 2013;258:614–617.
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Escalation of Care in Surgery A Systematic Risk Assessment to Prevent Avoidable Harm in Hospitalized Patients Maximilian Johnston, MB BCh, MRCS,∗ Sonal Arora, PhD, MRCS,∗ Oliver Anderson, MSc, MRCS,∗ Dominic King, MB ChB, MRCS,∗ † Nebil Behar, MSc, FRCS,‡ and Ara Darzi, MD, FRCS, FACS∗ † Objective: To systematically risk assess and analyze the escalation of care process in surgery so as to identify problems and provide recommendations for intervention. Background: The ability to escalate care appropriately when managing deteriorating patients is a hallmark of surgical competence and safe postoperative care. Healthcare-Failure-Mode-Effects-Analysis (HFMEA) is a methodology adapted from safety-critical industries, which allows for hazardous process failures to be prospectively identified and solutions to be recommended. Methods: Forty-two hours of ethnographic observations on surgical wards in 3 London hospitals (phase 1) formed the basis of an escalation process diagram. A risk-assessment survey identified failures associated with process steps and attributed hazard scores (phase 2). Patient safety and clinical risk experts validated hazard scores through a group consensus meeting (phase 3). Hazardous failures were taken forward to multidisciplinary HFMEA where cause analysis was applied and interventions were recommended (phase 4). Results: Observations identified 33 steps in the escalation process. The riskassessment survey (30 surgical staff members, 100% response) and expert consensus group identified 18 hazardous failures associated with these steps. The HFMEA team identified 3 adequately controlled failures; therefore, 15 were subjected to cause analysis. Outdated communication technology, understaffing, and hierarchical barriers were identified as root causes of failure. Participants recommended interventions based on these findings including defined escalation protocols, human factors education, enhanced communication technology, and improved clinical supervision. Conclusions: Failures in the escalation process amenable to intervention were systematically identified. This mapping of the escalation process will allow tailored interventions to enhance surgical training and patient safety. Keywords: escalation of care, health care failure mode effects analysis, patient safety, risk assessment, surgery (Ann Surg 2015;261:831–838)
T
he principles of quality and patient safety should be at the forefront of modern surgical care. Despite this, the rate of adverse events in surgery can be as high as 12.9%.1 In an attempt to address these concerns, there has been an increased drive to understand conFrom the ∗ Centre for Patient Safety and †Centre for Health Policy, Department of Surgery and Cancer, Imperial College London; and ‡Chelsea and Westminster Hospital NHS Foundation Trust, London, UK. Disclosure: This study did not require formal ethical approval. It was registered and approved as a quality improvement project. Informed written consent was gained from all participants. All authors declare no conflicts of interest. Johnston, Arora, King, Anderson, and Darzi are affiliated with the Centre for Patient Safety and Service Quality (www.cpssq.org) at Imperial College, which is funded by the National Institute for Health Research, UK. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.annalsofsurgery.com). Reprints: Max Johnston, MB BCh, MRCS, Department of Surgery and Cancer, St Mary’s Hospital, Imperial College London, W2 1NY, UK. E-mail: [email protected]. C 2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright ISSN: 0003-4932/14/26105-0831 DOI: 10.1097/SLA.0000000000000762
Annals of Surgery r Volume 261, Number 5, May 2015
tributory factors in the operating room (OR)2 and intensive care unit (ICU) so as to develop and implement evidence-based interventions.3 Despite significant improvements, these studies have neglected to explore the postoperative period on the ward, especially in the setting of acute physiological deterioration. Sound postoperative care, to include timely recognition and management of complications, is an important aspect of patient safety4 and can significantly reduce mortality when performed to the highest standard.5 One of the most critical elements of postoperative care is the ability to recognize a deteriorating patient and communicate this to a more senior team member able to provide definitive treatment. This type of communication will typically occur in academic settings where residents are training; communication pathways may be different in other health care settings. The recognition and communication of patient deterioration to a senior colleague can be termed escalation of care (EOC).6,7 In academic health care settings all over the world, it is typically a junior doctor who is the first point of contact for nursing staff when a postoperative patient becomes unstable. This relatively inexperienced clinician must make a rapid assessment of the patient before deciding whether to ask a senior colleague for help. This process is fraught with difficulty; studies suggest that failures in escalation of care are ubiquitous with at least 1 in 5 patients experiencing delays in urgent treatment.8,9 The effect of these deficiencies can lead to the “failure to rescue” of a patient in need, a potentially catastrophic event. In fact, evidence suggests that patients experiencing delayed escalation have an increased risk of morbidity10 and mortality.11,12 Further exploration of the escalation of care process is therefore warranted to extend patient safety gains garnered in the OR and ICU onto postoperative surgical wards. Initial studies evaluating escalation of care have only explored the role of communication failures in this pathway.13 Although important, this approach has neglected to comprehensively consider the human, technical, and patient factors involved in this safety-critical process. More detailed investigation is required—especially in light of calls from the Joint Commission in the United States14 and the Francis Report in the United Kingdom15 mandating the need for more proactive risk assessments of health care processes. A Healthcare Failure Mode and Effects Analysis (HFMEA) is a systematic method of conducting a proactive risk assessment. It allows for the most hazardous failures in a process to be identified and prioritized, so that interventions can be targeted at the most appropriate point. HFMEA incorporates a multistage approach, which utilizes the expertise of an interprofessional team to develop process flow-charts, hazard scores, and decision trees, which define areas of potential failure where the patient is most susceptible to avoidable harm. Originally used by the US Army,16 this research method has also been employed effectively by the aviation and motor industries.17,18 The HFMEA process is a process which enables prospective identification of the severity, detectability, and frequency of potential risks to patient care.19 The standard HFMEA process involves the following steps: define the HFMEA process, assemble the team, graphically describe the process, conduct a risk assessment, discuss areas for www.annalsofsurgery.com | 831
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Johnston et al
intervention, and feedback recommendations to decision-makers. The HFMEA process significantly differs from root cause analysis (RCA) in its timing of data collection. HFMEA aims to prospectively identify failures within a process to allow prevention before their occurrence. However, RCA involves retrospective analysis of events leading to an adverse outcome using case note analysis and interview techniques. The benefit of using HFMEA rather than RCA methodology is similar to a comparison of a prospective versus retrospective clinical study. As such HFMEA is a powerful tool for patient safety due to its proactive nature allowing identification and prevention of failures before they occur. Within surgery, risk assessments using this approach have identified problems in hand hygiene and medication delivery.20 Others have conducted risk assessments of the postoperative handoff process leading to implementation of a protocol, which has significantly improved handoffs from the OR to the post-anesthesia care unit.21 However, no previous studies have conducted a proactive risk assessment of the escalation of care process—despite its importance in preventing potentially avoidable morbidity and mortality. Not surprisingly therefore, interventions have been developed rather haphazardly with little success in improving either the process of escalation or patient outcomes.22,23 The aims of this study were to systematically risk assess and analyze the escalation of care process so as to identify areas of failure and avoidable patient harm, understand their causes, and make recommendations to improve patient safety on the surgical ward.
METHODS This study used the validated HFMEA technique modified by the Veterans Affairs National Center for Patient Safety.19 According to this criterion for HFMEA methodology, this study was conducted in 4 phases—phase 1: ethnographic observation of escalation episodes on surgical wards; phase 2: a risk assessment survey of the EOC process; phase 3: validation of hazard scores by patient safety and clinical risk experts through a consensus group meeting; phase 4: modified HFMEA to identify root causes of potentially hazardous failures and provide recommendations for intervention to prevent avoidable harm (see Fig. 1).
Setting This study was carried out in North-West London and included 6 general surgery wards in 4 hospitals (academic/teaching and community). Formal ethical approval was sought but not deemed necessary by the local ethics committee, so the study was registered as a quality improvement project.
Phase 1: Ethnographic Observations Ethnographic observations were conducted to ensure comprehensive capture of all of the steps in the escalation of care process. Ethnographic observation allows researchers to witness events in realtime and in-situ with minimal interference. In this study, the observers were external to the organization where the study took place and, as such, were not known to staff members. Ward staff were notified that observers would be conducting an evaluation of clinical communication on the wards and had agreed to the presence of observers before commencement of the study. Observational work also gives the researcher insight into how teams and organizations work; this is important for escalation of care as it is a collaborative rather than individual process.24 Two researchers with a background in general surgery and patient safety (M.J. and S.A.) observed and kept detailed ethnographic field notes on episodes of patient deterioration and escalation on general surgery wards and surgical high-dependency units for 42 hours. Fortunately, there were no serious situations that were observed where the patient was critically unwell and care was not 832 | www.annalsofsurgery.com
escalated. Had such a situation arisen, the observers (who are both medically qualified) would have alerted the appropriate medical personnel on duty for that ward. Patient safety is always the priority over the research study as per ethical requirements. During this time, 28 escalation events were observed. Each stakeholder (nursing assistant, nurse, intern, junior resident, senior resident, and attending surgeon) in the process worked on the postoperative surgery ward and was observed for 2 separate 3.5-hour sessions, one by each observer (a total of 42 hours). As per the study protocol, all observations were conducted in 3.5-hour aliquots. If an event of interest was occurring at the 3.5-hour mark, the observer remained until that event had been concluded. Each stakeholder was observed for 1 session during normal working hours (8 AM–5 PM) and 1 session outside of normal working hours (5 PM–8 AM). Although a single stakeholder was the primary focus of each observation session, the interprofessional nature of escalation of care meant that multiple stakeholders were encountered during each session (eg, nurse contacting resident regarding a deteriorating patient). The observers performed the primary 3.5-hour observation session together to ensure consistency of data capture in future sessions. These sessions took place between January and March 2013. No data were collected that could identify staff members, and as such, no evaluation of individual staff members was conducted. All data were anonymized at the point of collection rendering any synthesis of individual performance measures impossible. From these observations, a comprehensive flow diagram of the surgical escalation process was produced with input from all investigators and the literature6,25,26 (see Supplemental Digital Content Appendix A, available at http://links.lww.com/SLA/A561).
Phase 2: Risk Assessment Survey This phase was conducted to provide quantitative data on stakeholders’ views of the level of risk associated with each step of the escalation process. Data were collected with a detailed risk assessment survey based on the observations from phase 1 in 2 London hospitals (hospitals 1 and 3 in Table 1). All stakeholder groups in the escalation process completed the survey (n = 30). From the first hospital, there were 2 attending surgeons, 7 surgical residents, 2 surgical interns, 3 nurses, and 2 nursing assistants from the postoperative surgery ward. From the second hospital, there were 2 attending surgeons, 4 surgical residents, 3 surgical interns, and 5 nurses from the postoperative surgery ward. Participants used a 4-point scale to rate the potential consequences of failure for each step in the escalation process according to its likelihood of occurrence, potential harm associated with failure to escalate, and the detectability of a problem being identified before patient harm. These 3 ratings were multiplied together to calculate hazard scores for each process step (see Table 2).
Phase 3: Hazard Score Validation The hazard scores from the risk assessment were then validated by a group of patient safety and clinical risk experts. There were 2 consensus group sessions involving 5 participants each of that occurred 2 weeks after the initial risk assessment to allow the surveys to be analyzed in the intervening period. Participants were recruited from the Centre for Patient Safety and Service Quality at Imperial College. The participants in each group were a patient safety manager, clinical risk director, surgeon, physician, and critical care nurse. A facilitator (M.J. or O.A.) was also present but did not express any opinions; their role was to structure discussions and act as a timekeeper. Each process step and hazard score was discussed within the group, and consensus was reached before moving on to the next step. To allow prioritization, a hazard score threshold of 50% was chosen on the basis of previous guidelines for HFMEA methodology in surgical research.20,25 Process steps exceeding this threshold were C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Escalation of Care in Surgery
FIGURE 1. A diagram showing the steps in the method for modified HFMEA. TABLE 1. Details of the Hospitals Participating in the HFMEA
Hospital
Type of Hospital
No. Beds
No. Wards Observed
Hospital 1
Academic/Teaching
418
2
Hospital 2
Academic/Teaching
500
1
Hospital 3
Academic/Teaching
430
1
Hospital 4
Community
463
2
Surgical Specialties on Ward(s)
No. Patients on a Typical Ward
General Vascular General Urology Emergency General Vascular Urology General
18 26 28 32
Resident Make-up per Clinical Team
Surgical HDU (No. Beds)
1xPGY1 1xPGY2-4 1xPGY1 1xPGY2-4 2xPGY1 1xPGY2 1xPGY-1 1xPGY2-4 1xSenior resident
General (4) Vascular (6) No Yes (4) Yes (5)
HDU indicates high dependency unit.
TABLE 2. Hazard Scoring System for Escalation of Care Process Failures Score∗ Category Severity† Frequency Detectability
1
2
3
4
No adverse outcome 1 per year High
Increased stay 1 per month Moderate
Disability 1 per week Low
Death 1 per day Remote
∗
The 3 scores were multiplied together to give a hazard score (maximum score = 64). †The severity was interpreted as the worst possible potential outcome from each failure mode.
classified as hazardous (patient at risk of avoidable harm) and carried forward to the modified HFMEA analysis.
Phase 4: Modified HFMEA Formal modified HFMEA was used to confirm the failures and hazard scores associated with each of the hazardous process steps. These failures and hazard scores were generated on the basis of the literature20,21 and observations in phase 1, the risk assessment in phase 2, and expert opinion in phase 3 so as to provide a triangulated approach ensuring detection of all potential failures in the process. An interprofessional HFMEA team was assembled consisting of a patient, a nursing assistant, and 2 nurses from a postoperative C 2014 Wolters Kluwer Health, Inc. All rights reserved.
surgery ward, 2 surgical interns, 4 surgical residents, and 2 attending surgeons. The patient was recruited from the postoperative outpatient clinic having had major abdominal surgery 3 weeks previously. The team was supplemented by a patient safety researcher (M.J.) experienced in the HFMEA process, who acted as a facilitator and ensured that participants were given equal opportunity to air their views. After explanation of the HFMEA procedure, the team checked the process diagram to ensure completeness. After this, the hazard score for each failure mode was reviewed and discussed using decision-tree analysis. Here each failure mode was assessed in terms of its potential detectability and whether there were preexisting control measures in place, were it to occur. Hazardous failures, which lacked control www.annalsofsurgery.com | 833
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Johnston et al
measures and were not easily detectable, were prioritized for further action. To eliminate the potential for participants with leadership qualities to dominate, each participant individually, and privately, scored each failure before discussion. Final ratings were reached through additional consensus agreement; if consensus was not possible, the failure was automatically included in further analysis. This ensured all potentially hazardous failures were reviewed to minimize selection bias. The HFMEA team explored each potential cause of process failure using the framework for incident investigation from the London protocol RCA method.27 Potential causes were categorized into organizational, environmental, patient, staff (individual), and team- and task-related factors. Serial causes for each failure were suggested until a definitive or root cause within each relevant category of the London protocol could be identified. This meant that each failure could have more than 1 cause if the participants assigned causes to multiple categories within the protocol. Identification of the underlying causes of a failure allowed synthesis of recommendations aimed at preventing the failure in the future. Suggested interventions were then fed back to managers and clinical directors through a clinical governance meeting and report, potentially facilitating implementation and quality improvement. Finally, each participant was asked to fill out a questionnaire describing their perceptions of the success of the research method and their knowledge of patient safety.
RESULTS Phase 1: Ethnographic Observations Ethnographic observations of surgical patients and staff led to the identification of 33 core steps involved in a successful escalation of care process. Additional steps could have been added to this depending on whether a junior doctor escalated directly to an attending surgeon or via a senior resident (see Supplemental Digital Content Appendix A, available at http://links.lww.com/SLA/A561). As this would only involve a repeat cycle of assessments (resident review followed by attending review), these steps were combined and considered together. The observation phase of this study revealed escalation to be a stepwise process prone to failure until a patient received definitive management to enable recovery (see Figure 2).
protocols, so a total of 15 (45%) steps were carried forward to cause analysis and solution synthesis (see Table 3). The 3 controlled failures were the nurse failing to assess the patient correctly, the nurse failing to adhere to the escalation protocol correctly, and the doctor failing to attend the patient. The control measures identified by participants were the use of vital signs measurement (which participants identified as objective physiological evidence of deterioration), the requirement for nurse assistants and junior nurses to check all abnormal vital signs with the senior nurse for the ward, and the directive that empowered nurses to contact senior residents or attendings if more junior doctors were not receptive to their request for assistance. The causes and recommended solutions to hazardous yet uncontrolled failures are described later. These are described according to the sequence of health care professional involvement when a patient becomes unwell and typically include nursing, junior doctor, and senior doctor stages.
Process Steps Involving Nursing Staff Causes There were 7 steps with hazardous, uncontrolled failures to consider in the nursing stage of the escalation process (see Table 4). These were the patient failing to inform the nursing assistant they feel unwell, failure of the nurse to attend the patient, failure to notice the patient is deteriorating, failure to measure the vital signs correctly, failure to document the vital signs correctly, failure to inform the senior nurse that the patient is deteriorating, and failure to inform the junior doctor about the deterioration. Participants felt that clinical understaffing was the principal cause of all these failures. They explained that a greater (and therefore more visible) staff presence would allow patients to raise concerns about their well-being more easily. Interim causes of these failures included patient alarm bells not working, inexperienced/agency staff, patients with dementia or communication difficulties, and transcription errors when recording vital signs. An interim cause means a cause that contributed to a failure but was not felt to be the root cause of that failure by participants. Regarding communication between the nurse and junior doctor; the participants felt that both human and technological factors were contributory. Human factors included the nurses fearing criticism by junior doctors, technological factors principally revolved around the limitations of the pager system.
Recommendations Phases 2 and 3: Risk Assessment Survey and Validation of Hazard Scores Of the 33 process steps involved in escalation of care, the risk assessment participants identified a total of 17 hazardous failures that exceeded the hazard score threshold. The patient safety and clinical risk experts identified 1 additional hazardous failure via consensus agreement, this was the patient informing the nursing assistant they feel unwell, rather than going straight to the nurse. These 18 steps were taken forward to discussion in the modified HFMEA.
Phase 4: Modified HFMEA The steps with associated hazardous failures within the process were presented to the modified HFMEA team. The HFMEA team then repeated the hazard scoring, the lowest possible score was 1 and the highest possible score was 64. Ratings applied by the HFMEA team ranged from 27 to 48. The hazard score threshold was higher than 26; therefore, all of the 18 process steps identified as hazardous in the risk assessment were confirmed by the HFMEA team. During the discussion, participants decided that 3 of these 18 hazardous steps had failures that were already adequately controlled with preexisting 834 | www.annalsofsurgery.com
The recommendations made by participants can be seen in full in Table 4. The main recommendations to improve the nursing steps of the process were as follows:
r Increased nurse: patient ratios r Investment and training in the use of electronic vital sign recording and documentation systems
r Removal of hierarchical barriers through development of an improved escalation protocol, and;
r Increased use of smartphone technology. Process Steps Involving Junior Doctors Causes There were 7 steps with hazardous, uncontrolled failures in the stage involving junior doctors assessing/managing a deteriorating patient (see Table 4). These were failure to take an adequate history, failure to examine the patient thoroughly, failure to check the drug/fluid charts or case notes, failure to commence the correct initial treatment, and failure to inform a senior doctor of patient deterioration. The causes of these failures included hierarchical and language barriers, senior staff being busy in the OR and unable to answer a C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Escalation of Care in Surgery
FIGURE 2. A diagram showing the pathway to successful escalation of care. TABLE 3. Details of Steps Taken Forward to HFMEA Process Phase
Steps With Hazardous Failures
Mean Hazard Score∗
Steps Included in HFMEA
Controlled Steps
Steps Analyzed in HFMEA
Nursing Junior doctor Senior doctor Total (mean)
16 14 3 33
26.8 25.1 28 (26.6)
9 8 1 18
2 1 0 3
7 7 1 15
∗
Threshold for analysis within HFMEA > 26.
pager, understaffing, charts being poorly designed or unavailable on the ward, inexperience of junior doctors, illegible clinical notes, and poor availability of equipment.
Recommendations The recommendations made by participants can be seen in full in Table 4. The main recommendations to improve the junior doctor steps of the process were as follows:
Recommendations The recommendations made by participants can be seen in full in Table 4. The main recommendations to improve the senior doctor steps of the process were as follows:
r Development of a clear escalation protocol r Production of a clinical guideline defining appropriate levels of
crease reliance on agency staff
care according to patient diagnosis, physiological parameters, and predictive scoring systems, and; r Improvement of resource and staffing management to ensure sufficient access to the OR and surgical ICU beds for unstable patients.
system
Evaluation of the Chosen Research Method
r Recruitment of more permanent nursing and medical staff to der Development and implementation of an electronic medical records r Encourage senior surgeons to be more proactive (meaning that the responsibility for initiating contact would not be the sole responsibility of juniors and nurses) r Educate junior doctors about the importance of prompt escalation of care, and; r Develop a software platform with integrated demographics, pathology, radiology, and vital signs to allow prompt escalation in the presence of patient deterioration.
Process Steps Involving Senior Doctors Causes There was 1 step with hazardous, uncontrolled failures in the senior doctor phase of the process (see Table 4), which was the senior doctor failing to arrange definitive management. Participants felt this could be due to the senior doctor failing to contact an appropriate colleague or failing to arrange appropriate patient transfer to the OR or ICU. The root causes of these failures were felt to be fear when faced with contacting the attending surgeon or intensivist on duty and the presence of logistical barriers. These barriers included a lack of intensive care beds and uncertainty regarding the appropriate level of care for each patient. The participants also felt that only having a single OR available for emergent surgeries out of normal working hours was a major safety concern. C 2014 Wolters Kluwer Health, Inc. All rights reserved.
The participants felt that the modified HFMEA process was easy to interpret (75%) and had increased their awareness of patient safety (83%). They expressed enthusiasm about participating in further sessions (75%) and would also recommend participation to a colleague (83%).
DISCUSSION This study represents the first systematic risk assessment of the entire process of recognizing and responding to an unstable patient on a surgical ward. The steps of the escalation of care process were carefully mapped and analyzed to identify the most hazardous failure points where the patient was most susceptible to harm. These hazardous failures were then subjected to detailed cause analysis, which enabled technological innovation, clear escalation protocols, and staff recruitment to be identified as key recommendations for improving patient safety in surgical inpatients. This study exposes the significant vulnerabilities faced by a patient on a surgical ward—at a point when high-quality care is most required. It demonstrates that failures can occur at any step of the escalation of care pathway, highlighting poor resilience and reliability in the overall system. This is in contrast to other high-risk, safety critical industries such as aviation and the military where redundancy mechanisms are incorporated into the system to compensate for a www.annalsofsurgery.com | 835
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Johnston et al
TABLE 4. Causes of Failures in Escalation of Care and Recommendations for Improvement Process Phase and Step With Hazardous Failures Nursing Steps Patient fails to inform nursing assistant they feel unwell Nurse fails to notice that patient is unwell Nurse fails to attend to patient promptly Nurse fails to measure vital signs correctly Nurse fails to document vital signs correctly Nurse fails to inform senior nurse that patient is unwell Nurse fails to contact junior doctor successfully
Junior Doctor Steps Junior doctor fails to take accurate history
Junior doctor fails to complete thorough examination
Causes of Failure Patient decides not to inform nursing assistant Patient cannot inform nursing assistant Decides patient is not unwell Fails to attend to patient promptly
Root Cause Poor knowledge of patient personality Bell not supplied or out of reach Lack of specialty training Use of agency staff High workload
Fails to measure vital signs correctly Documents vital signs incorrectly on chart Fails to inform junior doctor
Lack of appropriate technology
Decides not to contact junior doctor Pages doctor but no response Contacts but junior doctor will not attend
Hierarchical barriers Lack of confidence Previous trouble with pager system Extension engaged when returning page Unable to locate extension Busy with a patient Doctor receives multiple pages at once High workload Poor judgment
Fails to complete history Unable to communicate with patient
Lack of specialty training High workload Agency staff unfamiliar with hospital Language barriers
Fails to complete examination
Patient positioning
Transcribing error Unclear escalation protocol
Equipment not available Completes incorrect examination
High workload Lack of specialty training
Fails to review chart
Chart not on ward
Reviews chart incorrectly Reviews incorrect chart
Illegible chart Human error
Junior doctor fails to review fluid chart
Fails to review chart
Chart not on ward
Junior doctor fails to review patient case notes
Unable to fully review case notes
Not completed accurately Large files for long-term patients
Junior doctor fails to initiate correct treatment
Selects incorrect treatment
Illegible Lack of specialty training
Junior doctor fails to review drug chart
Unable to commence correct treatment
836 | www.annalsofsurgery.com
Equipment not available
Recommendations More regular patient rounds Nominate ward “champions” Investment and training in detection of deterioration for nurses Recruit more permanent staff Recruit more permanent staff Invest in bedside electronic vital signs charts Invest in bedside electronic vital signs charts Develop and publicize a clear escalation protocol Clear escalation protocol Teach handover skills (eg, SBAR∗ ) to nursing staff Use of mobile telephones Develop electronic patient platforms allowing transfer of patient information Use of mobile telephones Use messaging services to send information Use messaging services to send information Recruit more permanent staff Teach prioritization skills Investment and training in detection of deterioration for junior doctors Recruit more permanent staff Recruit more permanent staff Mobile phones able to call interpreter from bedside Increase ward staff numbers and manual handling Nominate ward equipment “champions” Recruit more permanent staff Encourage cross-discipline co-operation Electronic chart available across site Computerized chart Bedside software with electronic chart Electronic chart available across site Education and training for staff Electronic notes system with weekly summaries Computerized notes Support courses/teaching for junior medical staff Nominate ward equipment “champions” (continues)
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Escalation of Care in Surgery
TABLE 4. (Continued) Process Phase and Step With Hazardous Failures Junior doctor fails to contact senior doctor successfully
Senior Doctor Steps Senior doctor fails to arrange definitive management
Causes of Failure
Root Cause
Decides not to contact senior doctor Attempts contact but no response Contacts but senior doctor will not attend Contacts senior but delayed response
Hierarchical barriers Busy with a patient Unable to locate extension Extension engaged when returning contact High workload Senior off-site
Clear escalation protocols Use messaging services to send information Use of mobile telephones Develop electronic patient platforms allowing transfer of patient information Recruit more permanent staff Seniors from busy specialties to remain on-site when on-call
Unable to transfer patient
No ICU beds
Set aside surgical ICU beds or a separate unit Increase OR availability for emergency surgery Physiological parameter thresholds to decide level of care Clear escalation protocol
No OR available
Fails to contact appropriate colleague
Uncertainty regarding appropriate level of care for patient Fear of Attending Surgeon or Intensivist
Recommendations
∗
SBAR is a method of structuring handover. It stands for Situation, Background, Assessment, Recommendation. ICU indicates intensive care unit; OR, operating room.
potential failure at any one point—for example, through the use of back-up behaviors, dual-tasking, and debriefing.28 The concerns and recommendations described in this study are consistent with previous studies evaluating the safety of patient care. In particular, clinical understaffing was highlighted as a significant contributor to delays in recognizing and responding to a deteriorating patient. This finding is supported by research from the United States29 and worldwide30 reporting better patient outcomes in centers with higher nurse staffing levels. Staffing levels are also a focus for the Agency for Healthcare Research and Quality and the Joint Commission.31 They highlight that recruitment and retention of staff requires determined action from health care providers. The identification of communication failures leading to failed escalation of care is consistent with other work highlighting the ubiquitous nature of problems in information transfer across the surgical pathway.32 Human factors and technological failure were identified as the root causes of these communication failures. Regarding the former, this study described the importance of overcoming hierarchical barriers between junior and senior surgeons as crucial for ensuring successful escalation of care. A flattened hierarchy is a key property of a safe system7 ; recommendations were made for a clear escalation protocol to facilitate this. Others have reported that such escalation protocols can reduce ICU admission33 and mortality34 making them a cost-effective proposition, which could be readily implemented. However, if measurable improvements in patient safety are to be attained, the safety culture of the health care institution must also be improved.35,36 Recommendations on staff education and training were described as integral to this campaign so as to build capacity and capability in promoting patient safety. Evidencebased train-the-trainers courses and faculty development programs could be used to provide effective training at a large-scale, institutional level.37,38 The technology available to staff was also felt to be suboptimal. In particular, the view that the hospital pager is an outdated technology with significant flaws is also supported by the literature.39,40 The strengths of this study include the systematic manner of risk assessment that captured all the vulnerabilities in the EOC pro C 2014 Wolters Kluwer Health, Inc. All rights reserved.
cess. Specifically, this method enabled the prospective capture of all potential failures as opposed to a retrospective analysis following an adverse event, which could be limited to the factors surrounding that event. The length and depth of the observation period also allowed researchers to form a comprehensive map of EOC in surgery. The validation of process steps and associated failure modes by clinical risk and safety specialists strengthened the analysis. The inclusion of core stakeholders in the surgical care pathway meant that multiple perspectives allowed robust recommendations to be made appropriate and acceptable to end-users. The limitations to this study are present in all HFMEA processes. The subjective nature of participant’s views inevitably introduces bias; attempts were made to reduce this by ensuring group consensus at appropriate points throughout the process. In addition, the HFMEA process is built to consider process failures in isolation. If failures in the escalation process were to occur simultaneously, the recommendations made in this study may fail to control them. This study was conducted in hospitals run by the National Health Service in the United Kingdom, as such, the findings may not be generalizable to other countries or private health systems. However, the findings regarding a fear of criticism might be universally applied. This study has wide-reaching implications if adverse events in postsurgical patients are to be prevented. Breakdowns in the escalation of care process can ultimately lead to a “failure to rescue” or death after a complication. With failure to rescue rates ranging from 29% to 41%, early escalation of care is critical to prevent avoidable harm and improve outcomes.41 This study highlights the complexity of this process and, for the first time, describes the antecedents of the escalation of care process, which, if suboptimal, can lead to a failure to rescue event. In doing so, it draws attention to the numerous areas where patients on surgical wards are susceptible to avoidable harm. The design and implementation of tailored interventions, based on the recommendations of this study, should serve as a basis for future research. This can ensure that the quality of care received by a postoperative patient on a ward is commensurate to the same exacting standards seen in the OR, thereby promoting patient safety across the entire surgical pathway. www.annalsofsurgery.com | 837
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Johnston et al
CONCLUSIONS This study is the first to systematically explore potential failures in escalation of care on surgical wards using HFMEA. The identification of failures in the process and recommendation of solutions to resolve these failures will allow targeted interventions to be trialled. Future work should concentrate on development of welldesigned interventions and analysis of their effect on surgical patient outcome.
ACKNOWLEDGMENTS The authors thank all of the study participants. Their contributions were as follows: Study concept and design: All authors; Drafting/revising study protocol: M.J., S.A., O.A., D.K., A.D.; Data collection/analysis: M.J., S.A., N.B.; Writing of the first draft: M.J., S.A.; Revising draft and approval of final version: All authors.
REFERENCES 1. Sukumar S, Roghmann F, Trinh VQ, et al. National trends in hospital-acquired preventable adverse events after major cancer surgery in the USA. BMJ Open. 2013;3. 2. Haynes AB, Weiser TG, Berry WR, et al. A surgical safety checklist to reduce morbidity and mortality in a global population. N Engl J Med. 2009;360:491– 499. 3. Petrovic MA, Martinez EA, Aboumatar H. Implementing a perioperative handoff tool to improve postprocedural patient transfers. Jt Comm J Qual Patient Saf. 2012;38:135–142. 4. Ghaferi AA, Birkmeyer JD, Dimick JB. Hospital volume and failure to rescue with high-risk surgery. Med Care. 2011;49:1076–1081. 5. Glance LG, Dick AW, Meredith JW, et al. Variation in hospital complication rates and failure-to-rescue for trauma patients. Ann Surg. 2011;253:811–816. 6. Johnston M, Arora S, King D, et al. Escalation of care and failure to rescue: a multicenter, multiprofessional qualitative study. Surgery. 2014;155:989–994. 7. Mackintosh N, Sandall J. Overcoming gendered and professional hierarchies in order to facilitate escalation of care in emergency situations: the role of standardised communication protocols. Soc Sci Med. 2010;71:1683–1686. 8. Quach JL, Downey AW, Haase M, et al. Characteristics and outcomes of patients receiving a medical emergency team review for respiratory distress or hypotension. J Crit Care. 2008;23:325–331. 9. Boniatti MM, Azzolini N, Viana MV, et al. Delayed medical emergency team calls and associated outcomes. Crit Care Med. 2013;24:26–30. 10. Calzavacca P, Licari E, Tee A, et al. A prospective study of factors influencing the outcome of patients after a Medical Emergency Team review. Intens Care Med. 2008;34:2112–2116. 11. Pattison N, Eastham E. Critical care outreach referrals: a mixed-method investigative study of outcomes and experiences. Nurs Crit Care. 2012;17: 71–82. 12. Downey AW, Quach JL, Haase M, et al. Characteristics and outcomes of patients receiving a medical emergency team review for acute change in conscious state or arrhythmias. Crit Care Med. 2008;36:477–481. 13. Greenberg CC, Regenbogen SE, Studdert DM, et al. Patterns of communication breakdowns resulting in injury to surgical patients. J Am Coll Surg. 2007;204:533–540. 14. The Joint Commission announces 2014 National Patient Safety Goal. Jt Comm Perspective. 2013;33:3–4. 15. Francis R. Lessons from Stafford. Med Leg J. 2013;81:153–165. 16. Stalhandske E, DeRosier J, Wilson R, et al. Healthcare FMEA in the Veterans Health Administration. Patient Saf Qual Healthc. 2009;6:30–33. 17. Dale BG, Shaw P. Failure mode and effects analysis in the UK motor industry: a state-of-the-art study. Qual Reliab Eng Int. 1990;6:179–188.
838 | www.annalsofsurgery.com
18. Yang J, Hong-Zhong H, Li-Ping H, et al. Risk evaluation in failure mode and effects analysis of aircraft turbine rotor blades using Dempster–Shafer evidence theory under uncertainty. Eng Fail Anal. 2011;18:2084–2092. 19. DeRosier J, Stalhandske E, Bagian JP, et al. Using health care Failure Mode and Effect Analysis: the VA National Center for Patient Safety’s prospective risk analysis system. Jt Comm J Qual Improv. 2002;28:209, 248–267. 20. Anderson O, Brodie A, Vincent CA, et al. A systematic proactive risk assessment of hazards in surgical wards: a quantitative study. Ann Surg. 2012;255:1086–1092. 21. Nagpal K, Vats A, Ahmed K, et al. A systematic quantitative assessment of risks associated with poor communication in surgical care. Arch Surg. 2010;145:582–588. 22. Robb G, Seddon M. A multi-faceted approach to the physiologically unstable patient. Qual Saf Health Care. 2010;19:e47. 23. Kansal A, Havill K. The effects of introduction of new observation charts and calling criteria on call characteristics and outcome of hospitalised patients. Crit Care Resusc. 2012;14:38–43. 24. Dimick JB, Greenberg CC. Understanding gaps in surgical quality: learning to count what cannot be counted. Ann Surg. 2013;257:6–7. 25. Nagpal K, Abboudi M, Fischler L, et al. Evaluation of postoperative handover using a tool to assess information transfer and teamwork. Ann Surg. 2011;253:831–837. 26. Catchpole KR, de Leval MR, McEwan A, et al. Patient handover from surgery to intensive care: using Formula 1 pit-stop and aviation models to improve safety and quality. Paediatr Anaesth. 2007;17:470–488. 27. Vincent C, Taylor-Adams S, Stanhope N. Framework for analysing risk and safety in clinical medicine. BMJ 1998;316:1154–1157. 28. Catchpole KR, Dale TJ, Hirst DG, et al. A multicenter trial of aviation-style training for surgical teams. Journal of Patient Safety. 2010;6:180–186. 29. Aiken LH, Clarke SP, Sloane DM, et al. Hospital nurse staffing and patient mortality, nurse burnout, and job dissatisfaction. JAMA. 2002;288:1987–1993. 30. Zhu XW, You LM, Zheng J, et al. Nurse staffing levels make a difference on patient outcomes: a multisite study in Chinese hospitals. J Nurs Scholarsh. 2012;44:266–273. 31. Kane RL, Shamliyan T, Mueller C, et al. Nurse staffing and quality of patient care. Evid Rep Technol Assess (Full Rep). 2007:1–115. 32. Solvoll T, Scholl J, Hartvigsen G. Physicians interrupted by mobile devices in hospitals: understanding the interaction between devices, roles, and duties. J Med Internet Res. 2013;15:e56. 33. Adelstein BA, Piza MA, Nayyar V, et al. Rapid response systems: a prospective study of response times. J Crit Care. 2011;26:635.e11–635.e18. 34. Mitchell IA, McKay H, Van Leuvan C, et al. A prospective controlled trial of the effect of a multi-faceted intervention on early recognition and intervention in deteriorating hospital patients. Resuscitation. 2010;81:658–666. 35. Paine LA, Rosenstein BJ, Sexton JB, et al. Republished paper: assessing and improving safety culture throughout an academic medical centre: a prospective cohort study. Postgrad Med J. 2011;87:428–435. 36. Pronovost PJ, Weast B, Holzmueller CG, et al. Evaluation of the culture of safety: survey of clinicians and managers in an academic medical center. Qual Saf Health Care. 2003;12:405–410. 37. Ahmed M, Arora S, Tiew S, et al. Building a safer foundation: the Lessons Learnt patient safety training programme. BMJ Qual Saf. 2014;23:78–86. 38. Hull L, Arora S, Symons NR, et al. Training faculty in nontechnical skill assessment: national guidelines on program requirements. Ann Surg. 2013;258:370– 375. 39. Coiera E, Tombs V. Communication behaviours in a hospital setting: an observational study. BMJ. 1998;316:673–676. 40. Perna G. Study: pagers, outdated communication tech costing hospitals. Available at: http://www.healthcare-informatics.com/news-item/study-pagersoutdated-communication-tech-costing-hospitals. Accessed May 10, 2014. 41. Sheetz KH, Waits SA, Krell RW, et al. Improving mortality following emergent surgery in older patients requires focus on complication rescue. Ann Surg. 2013;258:614–617.
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
