Understanding the "Swiss Cheese Model" and Its Application to Patient Safety.

ABSTRACT: This article reviews several key aspects of the Theory of Active and Latent Failures, typically referred to as the Swiss cheese model of human error and accident causation. Although the Swiss cheese model has become well known in most safety circles, there are several aspects of its underlying theory that are often misunderstood. Some authors have dismissed the Swiss cheese model as an oversimplification of how accidents occur, whereas others have attempted to modify the model to make it better equipped to deal with the complexity of human error in health care. This narrative review aims to provide readers with a better understanding and greater appreciation of the Theory of Active and Latent Failures upon which the Swiss cheese model is based. The goal is to help patient safety professionals fully leverage the model and its associated tools when performing a root cause analysis as well as other patient safety activities.

Implementing a human factors approach to RCA2 : Tools, processes and strategies.

Root Cause Analysis and Action (RCA2 ) guidelines offer fundamental improvements to traditional RCA. Yet, these guidelines lack robust methods to support a human factors analysis of patient harm events and development of systems-level interventions. We recently integrated a complement of human factors tools into the RCA2 process to address this gap. These tools include the Human Factors Analysis and Classification System (HFACS), the Human Factors Intervention Matrix (HFIX), and a multiple-criterion decision tool called FACES, for selecting effective HFIX solutions. We describe each of these tools and illustrate how they can be integrated into RCA2 to create a robust human factors RCA process called HFACS-RCA2 . We also present qualitative results from an 18-month implementation study within a large academic health center. Results demonstrate how HFACS-RCA2 can foster a more comprehensive, human factors analysis of serious patient harm events and the identification of broader system interventions. Following HFACS-RCA2 implementation, RCA team members (risk managers and quality improvement advisors) also experienced greater satisfaction in their work, leadership gained more trust in RCA findings and recommendations, and the transparency of the RCA process increased. Effective strategies for overcoming implementation barriers, including changes in roles, responsibilities and workload will also be presented.

Using Broken Windows Theory as the Backdrop for a Proactive Approach to Threat Identification in Health Care.

OBJECTIVES: Historically, health care has relied on error management techniques to measure and reduce the occurrence of adverse events. This study proposes an alternative approach for identifying and analyzing hazardous events. Whereas previous research has concentrated on investigating individual flow disruptions, we maintain the industry should focus on threat windows, or the accumulation of these disruptions. This methodology, driven by the broken windows theory, allows us to identify process inefficiencies before they manifest and open the door for the occurrence of errors and adverse events. METHODS: Medical human factors researchers observed disruptions during 34 trauma cases at a Level II trauma center. Data were collected during resuscitation and imaging and were classified using a human factors taxonomy: Realizing Improved Patient Care Through Human-Centered Operating Room Design for Threat Window Analysis (RIPCHORD-TWA). RESULTS: Of the 576 total disruptions observed, communication issues were the most prevalent (28%), followed by interruptions and coordination issues (24% each). Issues related to layout (16%), usability (5%), and equipment (2%) comprised the remainder of the observations. Disruptions involving communication issues were more prevalent during resuscitation, whereas coordination problems were observed more frequently during imaging. CONCLUSIONS: Rather than solely investigating errors and adverse events, we propose conceptualizing the accumulation of disruptions in terms of threat windows as a means to analyze potential threats to the integrity of the trauma care system. This approach allows for the improved identification of system weaknesses or threats, affording us the ability to address these inefficiencies and intervene before errors and adverse events may occur.

A Theoretical Model of Flow Disruptions for the Anesthesia Team During Cardiovascular Surgery.

OBJECTIVES: This investigation explores flow disruptions observed during cardiothoracic surgery and how they serve to disconnect anesthesia providers from their primary task. We can improve our understanding of this disengagement by exploring what we call the error space or the accumulated time required to resolve disruptions. METHODS: Trained human factors students observed 10 cardiac procedures for disruptions impacting the anesthesia team and recorded the time required to resolve these events. Observations were classified using a human factors taxonomy. RESULTS: Of 301 disruptions observed, interruptions (e.g., those events related to alerts, distractions, searching activity, spilling/dropping, teaching moment, and task deviations) accounted for the greatest frequency of events (39.20%). The average amount of time needed for each disruption to be resolved was 48 seconds. Across 49.87 hours of observation, more than 4 hours were spent resolving disruptions to the anesthesia team's work flow. CONCLUSIONS: By defining a calculable error space associated with these disruptions, this research provides a conceptual metric that can serve in the identification and design of targeted interventions. This method serves as a proactive approach for recognizing systemic threats, affording healthcare workers the opportunity to mitigate the development and incidence of preventable errors precedently.

A Data-Driven Approach to Team Training for Nurses in a Level II Trauma Center.

This prospective investigation describes the process of designing a targeted, data-driven team training aimed at reducing identified process inefficiencies or flow disruptions (FDs) that threaten the optimal delivery of trauma care. Trained researchers observed and classified FDs during 34 trauma cases in a Level II trauma center. Multidisciplinary trauma personnel generated interventions to identified issues using the human factors intervention matrix (HFIX). This article focuses on one intervention: a formal trauma nurse training program centered around leadership, teamwork, and communication. The training was well perceived and was found to have a significant impact on participant knowledge of course content; t (65) = -13.92, p ≤ .01. By using hospital-specific data to drive intervention development from multidisciplinary team members, it is possible to develop effective solutions aimed at addressing individual threats.

Using HFACS-Healthcare to Identify Systemic Vulnerabilities During Surgery.

The Human Factors Analysis and Classification System for Healthcare (HFACS-Healthcare) was used to classify surgical near miss events reported via a hospital's event reporting system over the course of 1 year. Two trained analysts identified causal factors within each event narrative and subsequently categorized the events using HFACS-Healthcare. Of 910 original events, 592 could be analyzed further using HFACS-Healthcare, resulting in the identification of 726 causal factors. Most issues (n = 436, 60.00%) involved preconditions for unsafe acts, followed by unsafe acts (n = 257, 35.39%), organizational influences (n = 27, 3.72%), and supervisory factors (n = 6, 0.82%). These findings go beyond the traditional methods of trending incident data that typically focus on documenting the frequency of their occurrence. Analyzing near misses based on their underlying contributing human factors affords a greater opportunity to develop process improvements to reduce reoccurrence and better provide patient safety approaches.

Proactive Safety Management in Trauma Care: Applying the Human Factors Analysis and Classification System.

INTRODUCTION: This article examines the reliability of the Human Factors Analysis and Classification System (HFACS) for classifying observational human factors data collected prospectively in a trauma resuscitation center. METHODS: Three trained human factors analysts individually categorized 1,137 workflow disruptions identified in a previously collected data set involving 65 observed trauma care cases using the HFACS framework. RESULTS: Results revealed that the framework was substantially reliable overall (κ = 0.680); agreement increased when only the preconditions for unsafe acts were investigated (κ = 0.757). Findings of the analysis also revealed that the preconditions for unsafe acts category was most highly populated (91.95%), consisting mainly of failures involving communication, coordination, and planning. CONCLUSION: This study helps validate the use of HFACS as a tool for classifying observational data in a variety of medical domains. By identifying preconditions for unsafe acts, health care professionals may be able to construct a more robust safety management system that may provide a better understanding of the types of threats that can impact patient safety.

Coding Human Factors Observations in Surgery.

The reliability of the Human Factors Analysis and Classification System (HFACS) for classifying retrospective observational human factors data in the cardiovascular operating room is examined. Three trained analysts independently used HFACS to categorize observational human factors data collected at a teaching and nonteaching hospital system. Results revealed that the framework was substantially reliable overall (Study I: k = 0.635; Study II: k = 0.642). Reliability increased when only preconditions for unsafe acts were investigated (Study I: k =0.660; Study II: k = 0.726). Preconditions for unsafe acts were the most commonly identified issues, with HFACS categories being similarly populated across both hospitals. HFACS is a reliable tool for systematically categorizing observational data of human factors issues in the operating room. Findings have implications for the development of a HFACS tool for proactively collecting observational human factors data, eliminating the necessity for classification post hoc.

Evaluating the Reliability of the Human Factors Analysis and Classification System.

INTRODUCTION: This paper examines the reliability of the Human Factors Analysis and Classification System (HFACS) as tool for coding human error and contributing factors associated with accidents and incidents. METHODS: A systematic review of articles published across a 13-yr period between 2001 and 2014 revealed a total of 14 peer-reviewed manuscripts that reported data concerning the reliability of HFACS. RESULTS: Results revealed that the majority of these papers reported acceptable levels of interrater and intrarater reliability. CONCLUSION: Reliability levels were higher with increased training and sample sizes. Likewise, when deviations from the original framework were minimized, reliability levels increased. Future applications of the framework should consider these factors to ensure the reliability and utility of HFACS as an accident analysis and classification tool.

Reflex PCA3 messenger ribonucleic acid testing: validation of postbiopsy urine samples and correlation with prostate biopsy findings in ∼2000 patients.

OBJECTIVE: To validate post-transrectal ultrasonography (TRUS) prostate biopsy (bx) urine samples for PCA3 messenger ribonucleic acid testing, including correlation of PCA3 score with concurrent bx findings. METHODS: From July 2008 to July 2010, 2015 patients had urine collected immediately after a TRUS-guided prostate bx. Excluded were men with history of prostate carcinoma (CaP), <6 or ≥24 bx cores, and/or prostate-specific antigen (PSA) level ≥50 ng/mL, resulting in 1909 included men. PCA3 and PSA messenger ribonucleic acid were quantitated using transcription-mediated amplification. A PCA3 score of ≥35 was considered positive. RESULTS: Mean and median ages were 66 years. Mean and median PSA levels were 6.7 and 5.1 ng/mL, respectively. Bxs were benign in 970 (50.8%), CaP in 726 (38%), high-grade prostatic intraepithelial neoplasia (HGPIN) in 124 (6.5%), and atypical in 89 (4.7%). PCA3 test was informative in 1887 (98.8%) patients. Means ± standard deviations (median) of PCA3 scores for benign, HGPIN, atypical, and CaP were 22.3 ± 27.9 (12.8), 37.6 ± 43.2 (24.1), 35.7 ± 36.2 (25.7), and 46.9 ± 48.1 (31.6; P <.05 benign vs CaP, benign vs HGPIN and atypical, HGPIN and atypical vs CaP). Sensitivity and specificity of PCA3 for CaP were 46.3% and 78.7%, respectively. CaP risk increased with progressively higher PCA3 score ranges from 14.8% for PCA3 <5 to 66.7% for PCA3 >100. Area under the curve (AUC) for the PCA3 receiver operating characteristics was not significantly different in men without prior bx (AUC = 0.716) compared with men with at least 1 prior nonpositive bx (AUC = 0.702). CONCLUSION: Post-TRUS bx urine is a valid sample for PCA3 testing. Patients with a negative bx and a positive PCA3 test may have a higher likelihood of unsampled CaP.

The Human Factors Analysis Classification System (HFACS) applied to health care.

In spite of efforts to improve patient safety since the 1999 report, To Error Is Human, recent studies have shown limited progress toward preventing serious error. Most hospitals use root cause analysis as a method of serious event investigation. The authors postulate that this method suffers from 4 problems: (a) the use of root cause analysis is neither standardized nor reliable between organizations, (b) hospitals focus on "who" did "what" rather than on "why" the error occurred, (c) the identified causes are often too nonspecific to develop actionable correction plans, and (d) a standardized nomenclature does not exist to allow analysis of recurring errors across the organization. This article describes the modification of the Human Factors Analysis Classification System based on James Reason's theory of error causation for use in health care. This method resolves the 4 deficiencies noted above. The authors' experience investigating 105 serious events over 2 years is described.

Realizing improved patient care through human-centered operating room design: a human factors methodology for observing flow disruptions in the cardiothoracic operating room.

BACKGROUND: Human factors engineering has allowed a systematic approach to the evaluation of adverse events in a multitude of high-stake industries. This study sought to develop an initial methodology for identifying and classifying flow disruptions in the cardiac operating room (OR). METHODS: Two industrial engineers with expertise in human factors workflow disruptions observed 10 cardiac operations from the moment the patient entered the OR to the time they left for the intensive care unit. Each disruption was fully documented on an architectural layout of the OR suite and time-stamped during each phase of surgery (preoperative [before incision], operative [incision to skin closure], and postoperative [skin closure until the patient leaves the OR]) to synchronize flow disruptions between the two observers. These disruptions were then categorized. RESULTS: The two observers made a total of 1,158 observations. After the elimination of duplicate observations, a total of 1,080 observations remained to be analyzed. These disruptions were distributed into six categories such as communication, usability, physical layout, environmental hazards, general interruptions, and equipment failures. They were further organized into 33 subcategories. The most common disruptions were related to OR layout and design (33%). CONCLUSIONS: By using the detailed architectural diagrams, the authors were able to clearly demonstrate for the first time the unique role that OR design and equipment layout has on the generation of physical layout flow disruptions. Most importantly, the authors have developed a robust taxonomy to describe the flow disruptions encountered in a cardiac OR, which can be used for future research and patient safety improvements.

Comparative analysis of whole mount processing and systematic sampling of radical prostatectomy specimens: pathological outcomes and risk of biochemical recurrence.

PURPOSE: Whole mount processing is more resource intensive than routine systematic sampling of radical retropubic prostatectomy specimens. We compared whole mount and systematic sampling for detecting pathological outcomes, and compared the prognostic value of pathological findings across pathological methods. MATERIALS AND METHODS: We included men (608 whole mount and 525 systematic sampling samples) with no prior treatment who underwent radical retropubic prostatectomy at Vanderbilt University Medical Center between January 2000 and June 2008. We used univariate and multivariate analysis to compare the pathological outcome detection rate between pathological methods. Kaplan-Meier curves and the log rank test were used to compare the prognostic value of pathological findings across pathological methods. RESULTS: There were no significant differences between the whole mount and the systematic sampling groups in detecting extraprostatic extension (25% vs 30%), positive surgical margins (31% vs 31%), pathological Gleason score less than 7 (49% vs 43%), 7 (39% vs 43%) or greater than 7 (12% vs 13%), seminal vesicle invasion (8% vs 10%) or lymph node involvement (3% vs 5%). Tumor volume was higher in the systematic sampling group and whole mount detected more multiple surgical margins (each p <0.01). There were no significant differences in the likelihood of biochemical recurrence between the pathological methods when patients were stratified by pathological outcome. CONCLUSIONS: Except for estimated tumor volume and multiple margins whole mount and systematic sampling yield similar pathological information. Each method stratifies patients into comparable risk groups for biochemical recurrence. Thus, while whole mount is more resource intensive, it does not appear to result in improved detection of clinically important pathological outcomes or prognostication.

Operator error and system deficiencies: analysis of 508 mining incidents and accidents from Queensland, Australia using HFACS.

Historically, mining has been viewed as an inherently high-risk industry. Nevertheless, the introduction of new technology and a heightened concern for safety has yielded marked reductions in accident and injury rates over the last several decades. In an effort to further reduce these rates, the human factors associated with incidents/accidents needs to be addressed. A modified version of the Human Factors Analysis and Classification System was used to analyze incident and accident cases from across the state of Queensland to identify human factor trends and system deficiencies within mining. An analysis of the data revealed that skill-based errors were the most common unsafe act and showed no significant differences across mine types. However, decision errors did vary across mine types. Findings for unsafe acts were consistent across the time period examined. By illuminating human causal factors in a systematic fashion, this study has provided mine safety professionals the information necessary to reduce mine incidents/accidents further.