broken windows, trauma

ORIGINAL ARTICLE

Using BrokenWindows Theory as the Backdrop for a ProactiveApproach to Threat Identification in Health Care

Albert J. Boquet, PhD,* Tara N. Cohen, MS,* Jennifer S. Cabrera, MS,* Tracy L. Litzinger, MS,*Kevin A. Captain, RN,† Michael A. Fabian, MD,† Steven G. Miles, MD,† and Scott A. Shappell, PhD*

Objectives: Historically, health care has relied on error managementtechniques to measure and reduce the occurrence of adverse events. Thisstudy proposes an alternative approach for identifying and analyzing haz-ardous events. Whereas previous research has concentrated on investigat-ing individual flow disruptions, we maintain the industry should focus onthreat windows, or the accumulation of these disruptions. This methodol-ogy, driven by the broken windows theory, allows us to identify processinefficiencies before they manifest and open the door for the occurrenceof errors and adverse events.Methods: Medical human factors researchers observed disruptions dur-ing 34 trauma cases at a Level II trauma center. Data were collected duringresuscitation and imaging and were classified using a human factors taxon-omy: Realizing Improved Patient Care Through Human-Centered Operat-ing Room Design for Threat Window Analysis (RIPCHORD-TWA).Results: Of the 576 total disruptions observed, communication issues werethe 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 involvingcommunication issues were more prevalent during resuscitation, whereas co-ordination 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 ofthreat windows as a means to analyze potential threats to the integrity ofthe trauma care system. This approach allows for the improved iden-tification of system weaknesses or threats, affording us the ability toaddress these inefficiencies and intervene before errors and adverseevents may occur.

Key Words: human factors, trauma care, traumatic injury,flow disruptions, threats, threat windows, patient safety, adverse events,error, Level II trauma center

(J Patient Saf 2016;00: 00–00)

Traumatic injury is a major public health problem in the UnitedStates. Each year, more than 192,000 people lose their lives to

trauma.1 Trauma is currently the leading cause of death for allAmericans from birth to age 46. From 2000 to 2010, deathsresulting from traumatic injury increased by 22.8%.2 Research in-dicates that there is a 25% reduction in the risk of mortality forthose patients receiving treatment at a trauma center as opposedto a nontrauma center.3 Given these statistics, it is not surprisingthat health care has turned to human factors professionals to im-prove safety and efficiency in the delivery of trauma care.

Over the past few years, human factors researchers haveinvested a great deal of time and effort attempting to identify

AUTHOR

From the *Embry-Riddle Aeronautical University; and †Halifax Health Medi-cal Center, Daytona Beach, Florida.Correspondence: Tracy Lynn Litzinger, MS, 600 South Clyde Morris Blvd,

Daytona Beach, FL (e‐mail: [email protected]).The authors disclose no conflicts of interest.There were no sources of support for this study, and there are no disclosures

of funding.Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.

J Patient Saf • Volume 00, Number 00, Month 2016

Copyright © 2016 Lippincott Williams & Wilkins. Unau

process inefficiencies that threaten the optimal delivery of patientcare. These process inefficiencies are often referred to as flow dis-ruptions (FDs). Wiegmann and colleagues more specifically de-fined FDs as deviations from the natural progression of a taskthat potentially compromises the safety and efficiency of the pro-cess.4 A separate study conducted by Palmer and colleagues de-fined FDs as “any disruption to the normal flow of operations,”including equipment and/or communication failures, physical ob-structions, distractions, and usability issues. These may include avariety of types of disruptions to workflow, including slips, trips,and falls; missing, incorrect, or malfunctioning equipment; andtexting, pages, and phone calls.5 Research indicates that FDsthreaten process efficiency, presenting distractions, impairments,and workarounds that divert attention from the task at hand.4 Re-lated research also suggests that FDs can be considered symptomsof underlying failures somewhere in the system.6

Initial investigations of FDs targeted the cardiovascularoperating room (CVOR), a highly structured, regimented surgi-cal suite that lends itself nicely to the investigation of processinefficiencies and related quality improvements.4–10 AlthoughCVOR surgery often follows a fairly predictable course, thesame cannot be said for trauma care. Trauma resuscitation, bydefinition, is unpredictable in nature and occurs in a fast-paced, dynamic environment where health care professionalsmust quickly and accurately evaluate and diagnose potentiallylife-threatening injuries to the patient. This process is informa-tion laden and team dependent, relying heavily on clinical skilland overall system efficiency.11 Previous studies indicate thatcoordination and communication issues make up approxi-mately half of all FDs in Level I trauma centers.6–10

The study of FDs provides an evidence-based approach toinvestigating process inefficiencies in health care. However, thisapproach experiences a lack of clearly identifiable outcomes andmay be misleading when drawing conclusions. For example, con-sider that there can be multiple miscommunications that occurduring a procedure that do not necessarily impact the patient.Furthermore, although multiple team members may experi-ence numerous disruptive events while providing care to thepatient, these events do not necessarily affect the procedure itself.For instance, an anesthesiologist may receive multiple textmessages over the course of CVOR surgery; yet, these disrup-tions do not necessarily interrupt the surgeon or the progressof the procedure.

Examples such as these reveal the difficulty in isolating thoseFDs that truly pose discernable threats to patient outcomes andthe overall process of patient care. Bearing this in mind, it maybe better to conceptualize the accumulation of these FDs asthreat windows. Threat windows can be operationally definedas aggregates of those disruptions and process inefficienciesthat ultimately open the door for errors and adverse eventsto occur.

The concept of threat windows is inspired by the seminal articleby Wilson and Kelling on broken windows, which theorizes thatthere is an intrinsic connection between disorder and the prolifer-ation of crime in urban neighborhoods.12 The theory suggests that

'S COPY

www.journalpatientsafety.com 1

thorized reproduction of this article is prohibited.

mailto:[email protected]

www.journalpatientsafety.com

Boquet et al J Patient Saf • Volume 00, Number 00, Month 2016

minor crimes and nuisances—such as litter, graffiti, and publicdrinking—set the stage for neighborhood decay. As these seem-ingly minor acts flourish, residents become fearful, leading themto conclude that social order has broken down and that their com-munity has become unsafe. The net effect of this process is thatresidents cease to “invest” in the neighborhood, resulting in re-duced oversight and weakened social control. It is the initial prev-alence of disorder, which undermines the basic integrity of thecommunity and fosters a lengthier chain of events that paves theway for more serious crimes to take place.

Broken windows theory inspired a new policing strategy inNew York City in 1993 and was the blueprint for a concept thatbecame known as “community policing.” Based on the brokenwindows theory of deterrence, New York City began an order-maintenance policing strategy, which targeted minor misde-meanor offenses like turnstile jumping, aggressive panhandling,graffiti, and public drinking to prevent and reduce more seriousoffenses. Former Police Commissioner William Bratton creditsthe broken windows theory with reducing crime rates in NewYork City neighborhoods.13

Research conducted by Hesketh and collegues examinedbroken windows theory concepts in the health care domain,specifically with respect to the management of workplace vio-lence. The authors note that even seemingly innocuous orminor events (eg, verbal abuse, aggression, sexual harassment)should be immediately and firmly engaged to prevent thespread of aggressive behavior and more serious forms of vio-lence from incubating within the system. The authors go onto explain that when workplace violence is tolerated by theorganization, the work environment is negatively impacted ona number of levels, including job satisfaction, absenteeism,job performance, and mental health issues.14

Criminological research on the broken windows theory consis-tently supports the idea that no direct link exists between socialdisorder and crime.15–17 Yet, such a simplistic view of complexsocial systems, whether they be inner city neighborhoods orhealth care settings, fails to do justice to the complexities ofhow individuals operate within those systems.Wilson and Kellingmaintain that social disorder and crime are developmentallylinked rather than causally linked.12 This position fits nicelywith the findings of Wiegmann and colleagues, who identifieda relationship, albeit indirect, between FDs and errors in theCVOR.4 Just as social disorder can reduce the stability of acommunity and lead to an increase in serious crime, so can pro-cess disorder lead to reductions in the collective efficacy ofhealth care teams.

Using broken windows theory as a backdrop, we assert thatFDs introduce a level of disorder in the system. The accumulationof these disruptions is indicative of system-induced entropy thatpaves theway for potential errors and adverse events to occur. Per-haps it is time to stop viewing these events in isolation and startconceptualizing them collectively as threat windows. These threatwindows, in and of themselves, stand alone as a process measurefor the delivery of care.

The use of threat windows rather than errors has multiple poten-tial benefits as a metric for patient safety. First, it removes thestigma associated with the capture of errors, which has the poten-tial to lead to blame and limits the focus of investigation to the“sharp end of the spear.” Second, it makes sense to identify thosesystemic weaknesses that may precede errors and adverse events,rather than waiting for them to occur. Finally, because threat win-dows consider FDs in terms of frequency, this sort of aggregatemay provide a richer source of data for developing targeted inter-ventions that may prevent adverse events from occurring in thefirst place.

AUTHOR

2 www.journalpatientsafety.com


The goals of this research are threefold: (1) introduce threatwindows within the context of broken windows theory, providingan enhanced approach to understanding the impact of FDs inhealth care; (2) identify threat windows in a Level II trauma cen-ter; and (3) assess the viability of using an existing human factorstaxonomy to classify those observed threat windows.

METHODS

PopulationExperienced medical human factors researchers observed a

sample of 34 trauma cases at a Level II trauma center. All obser-vations were collected at an East Central Florida community hos-pital with 678 licensed beds. The study was approved by thehospital’s research oversight committee as a quality improvementproject. It was considered exempt from institutional review boardreview as the focus was on disruptions involved in the trauma careprocess rather than clinical outcomes of the patient. Hospital staffand trauma team members were aware of the presence and re-search goals of the observers.

Procedure

Data CollectionProspective data were collected from the time the patient

arrived in the resuscitation bay and continued through imaging(if needed) until disposition to surgery, the medical floor unit, orthe emergency department for further assessment. Specifically,case observation occurred in 2 phases: (1) observation of the teamin the resuscitation bay (area used to stabilize patients) and (2)observation of the team in imaging (CT scan room where patientsare taken after resuscitation for in-depth diagnostic images to helpproviders gain a better understanding of the patient’s medical sta-tus). Researchers observed and recorded FDs during both phases(if applicable), as well as the amount of time the patient and teamspent in each observation area. Flow disruptions were operation-ally defined as those events that resulted in a delay or disturbanceto a team member’s progress.

'S COPY

Data Coding and ClassificationDisruptions were classified using a human factors taxonomy:

Realizing Improved Patient Care through Human-Centered Oper-ating Room Design for Threat Windows Analysis (RIPCHORD-TWA). The initial RIPCHORD framework was designed foridentifying and classifying work flow disruptions in the CVOR.The original taxonomy classifies FDs by type into 6 distinct clus-ters (communication, usability, physical layout, environmentalhazards, general interruptions, and equipment failures).5 In thecurrent study, the researchers used an expanded version of theframework to accommodate threat window analysis (RIPCHORD-TWA). RIPCHORD-TWA includes 6 major categories for clas-sifying human factors–related disruptions: communication,coordination, equipment issues, interruptions, layout, andusability. These major categories represent the threat windowsassociated with the delivery of trauma care over the course ofthe cases observed. These threat windows can be further bro-ken down into minor categories, providing a fine-grainedanalysis of the types of disruptions occurring in a given set-ting (Table 1). Each observation was classified into theRIPCHORD-TWA taxonomy by at least 2 observers via con-sensus coding. Descriptive statistics were calculated, includingfrequency of FDs, percentage of FDs by category, and time

© 2016 Wolters Kluwer Health, Inc. All rights reserved.



TABLE 1. RIPCHORD-TWA Taxonomy

Communication(verbal and non-verbal)

Interruptions (other)Alerts

Confusion DistractionsEnvironmental noise Equipment/suppliesIneffective communication Interaction with biohazardsLack of response Searching activityLack of sharing Spilling/droppingNonessential communication Task deviationSimultaneous communication Teaching moments

Coordination LayoutCharting/documentation Connector positioningPersonnel not available Equipment positioningPersonnel rotation Furniture positioningPlanning/preparation Inadequate spaceProtocol failure Permanent structure positioningUnknown information Wires/tubing

Equipment issues UsabilityAnesthesia equipment Barrier designGeneral equipment Computer designPerfusion equipment Data entry (non-computer) designSurgeon equipment Equipment design

Packaging designSurface design

J Patient Saf • Volume 00, Number 00, Month 2016 Broken Windows in Health Care

elapsed during the case within each area observed (resuscitationbay or imaging).

RESULTSA total of 576 disruptions were identified during the 34

observed cases (1138 patient contact minutes), which trans-lates to roughly 17 disruptions per case or approximately onedisruption every 2 minutes. Issues relating to communicationrepresented the bulk of the 576 disruptions, making this cate-gory the largest threat window in the delivery of trauma care(28%). Interruptions and coordination issues were each evenlyrepresented at 24% each. Layout, usability, and equipmentissues comprised 16%, 5%, and 2% of disruptions, respec-tively. Further analyses examined differences in the natureand frequency of disruptions between resuscitation and imag-ing. Although the pattern of results was similar to that whichwas seen in the overall analysis, some differences betweenphases were observed. Namely, communication and interrup-tions presented more of an issue during resuscitation, whereascoordination issues were more prevalent in imaging (Fig. 1).

To better understand the composition of the threat windows pop-ulating each major RIPCHORD-TWA category, researchers con-ducted a fine-grained analysis of the data. The threat window forcommunication largely consisted of the following 3 minor catego-ries: ineffective communication (32%), lack of response (24%),and nonessential communication (20%). The remaining disruptionswere divided across confusion (12%), simultaneous communica-tion (7%), lack of sharing (4%), and environmental noise (1%).The 2 most populated minor categories within interruptions werespilling and dropping (34%) and distractions (16%). These werefollowed by searching activity (13%), alerts (12%), equipmentand supplies (11%), teaching moments (10%), task deviation

AUTHOR



(3%), and interaction with biohazards (0%). Minor categories com-prising the bulk of coordination issues consisted of planning andpreparation (35%), personnel not available (21%), charting anddocumentation (21%), unknown information (18%), protocol fail-ure (4%), and personnel rotation (0%). The threat window relatedto layout was further broken down into wires and tubing (45%), in-adequate space (33%), equipment positioning (16%), connector po-sitioning (3%), and furniture positioning (3%). Usability wasfurther broken down into data entry (52%), equipment design(24%), barrier design (17%), computer design (3%), and packagingdesign (3%). The only minor category populated within equipmentissues was general equipment (100%).

Fine-grained analysis revealed some differences between imag-ing and resuscitationworth noting. Themost prevalent type of com-munication disruption observed in resuscitation was ineffectivecommunication (35%). However, both ineffective communication(32%) and nonessential communication (32%) represented the larg-est threats to communication in imaging. Lack of response occurredtwice as often in resuscitation (30%) as compared with imaging(12%) (Fig. 2).

The most frequently occurring interruption in resuscitationwas spilling and dropping (44%), compared with 19% in imag-ing. However, distractions posed the largest threat to the teamwhen they were in imaging (27%), as opposed to 10% in resus-citation (Fig. 3).

In both resuscitation and imaging, breakdowns in coordina-tion most often took the form of issues related to planning andpreparation (48% and 32%, respectively). Following planningand preparation issues, unknown information (23%) was thenext largest coordination issue in resuscitation. In imaging,on the other hand, the next largest coordination issues were per-sonnel not available and charting and documentation (24% and23%, respectively) (Fig. 4).

There were several differences involving the specific typesof layout issues observed in resuscitation and imaging. Inade-quate space contributed most to layout disruptions in resuscita-tion (46%); however, this category played a much smaller rolein imaging (16%). On the other hand, issues surrounding wiresand tubing made up a substantial amount (71%) of the disrup-tions in imaging, despite not being very prevalent in resuscita-tion (19%) (Fig. 5).

The final 2 categories, usability and equipment issues, wereboth lightly populated, representing only 5% and 2% of disrup-tions, respectively. With respect to usability, the most populatedminor category was equipment design (44%) in resuscitation.The most prevalent usability issues in imaging were those relatedto noncomputer data entry (92%). Finally, equipment issues weremade up entirely of general equipment issues for both resuscita-tion and imaging.

'S COPY

DISCUSSIONGruen and colleagues aptly describe trauma as a “perfect

storm” for the incidence of medical error. They discuss how “in-complete patient histories, unstable patients, time-critical deci-sions, concurrent and competing tasks, the involvement of manydisciplines, long hours, and often junior personnel working afterhours in busy emergency departments” create the ideal breedingground for medical errors to occur.18 Despite the obvious impor-tance of error management in health care, preventable deaths arestill occurring at a rate of 2 to 22% in trauma care.19

Rather than continuing to solely conduct studies on error andpreventable death in medicine, we propose an additional means ofconceptualizing potential threats to the integrity of the system. Thisstudy used a human factors framework to classify disruptions




FIGURE 1. Overall breakdown of major categories by location.


PY
occurring during the administration of care in a Level II trauma fa-cility. The accumulation of these disruptions are referred to as threatwindows; the greater the number of disruptions that occur, thelarger the threat window becomes. These disruptions represent sys-temic weaknesses or threats to the quality of patient care and mayset the stage for the occurrence of errors and adverse events downthe line. This approach has a major advantage in that we are nowafforded the ability to address these threats proactively and inter-vene before potential errors occur.
According to our observations and subsequent analysis, thetrauma team experienced an average of 17 disruptions per casewhich translated to one disruption every 2 minutes. Communication-related disruptions represented the largest threat window, followedclosely by interruptions and issues related to coordination. Theseresults are similar to those of others conducting analogous re-search in various health care domains.4–10 Layout, usability, andequipment comprised the remaining threat windows primarilyduring the resuscitation phase.

Although the identification of the major types of threat win-dows is no doubt important, this alone does not provide the levelof resolution necessary to generate targeted, data-driven interven-tions. Thus, the researchers conducted a fine-grained analysis ofthe data by classifying it intominor RIPCHORD-TWA categories.To further move the analyses away from the “one size fits all”

HOR

FIGURE 2. Communication breakdown by resuscitation and imaging.


AUT


paradigm, the data were separated by phase (resuscitation orimaging).

The fine-grained analyses revealed that 3 categories accountedfor the majority of the communication-related disruptions in re-suscitation and imaging: ineffective communication, lack of re-sponse, and nonessential communication. Perhaps the prevalenceof these issues speaks to the inherent differences in training receivedby nurses and physicians. As explained by Thomas and colleagues,nurses are generally taught to be descriptive in their thought andlanguage while physicians are concise in thought and speak usingshorter narratives.20

Lack of response posed a greater issue during resuscitation,whereas nonessential communication occurred more frequentlyduring imaging. This is not surprising given the nature of the treat-ment during those phases. A multidisciplinary team composed ofa minimum of 8 to 10 team members is responsible for stabiliza-tion of a critically injured patient in resuscitation. This process oc-curs at a fast pace and demands that multiple simultaneousinteractions occur effectively, thus increasing the propensity fordropped requests.

On the other hand, once the patient is situated in the CT scan-ner, the down time provides the team with a period of reduced in-tensity, fostering a more relaxed environment within whichnonessential communication unrelated to patient care was more

'S CO




FIGURE 3. Interruptions breakdown by resuscitation and imaging.


PY
likely to occur. Nevertheless, the importance of remaining alert tosudden changes in patient status during this time cannot be em-phasized enough.
The most populated categories of interruptions were distractionsand spilling/dropping. Although these categories accounted for thegreatest number of disruptions, they were also the most disparatebetween resuscitation and imaging. Spilling and dropping items oc-curred much more often during resuscitation. The combination offast-paced trauma care efforts alongside the team’s interaction withvarious equipment and supplies may explain the prevalence of is-sues related to spilling and dropping. On the other hand, distractionsrepresented amore pervasive threat during imaging. This is compat-ible with the high frequency of nonessential communication duringthis time. These findings are similar to those of studies conducted ata Level I trauma center, which found FDs to be most prevalent dur-ing the imaging phase of patient care.9

Coordination is difficult to manage in a hectic trauma environ-ment, where space restrictions, time constraints, and the numberof personnel involved pose unique challenges to providing patientcare. Overall, the fine-grained analysis revealed that planning andpreparation issues occurred most frequently; however, a substan-tial number of these issues were observed during resuscitation.

Although charting/documentation and personnel not availablefollowed planning and preparation in terms of frequency, both

HOR

FIGURE 4. Coordination breakdown by resuscitation and imaging.


AUT


issues were more prevalent during imaging. Disruptions associ-ated with personnel not available speak to one of the more signif-icant differences between a Level I and a Level II trauma center.Unlike a Level I, the trauma team in a Level II facility is madeup of individuals that leave their primary duties to respond tothe unscheduled arrival of a trauma patient. This model may ac-count for the increase in disruptions related to personnel not avail-able in imaging. After the patient is stabilized in resuscitation,personnel conscripted from other departments may migrate backto their primary duty stations, making them unavailable shouldthe need for their services arise again.

Within layout, disruptions related towires and tubing posed thegreatest threat overall. These issues accounted for a much highernumber of disruptions in imaging as compared with resuscitation.This is consistent with the observation that, within the imaging en-vironment, medical personnel have to not only manage physicallymoving the patient into a confined space, but must also arrangeclinical equipment. On the other hand, inadequate space andequipment positioning posed more of an issue during resuscita-tion. This finding is understandable considering the inordinatenumber of personnel operating in a limited space around the pa-tient. This issue is exacerbated by the continuous need to reposi-tion equipment (ie, portable x-ray machines) such that medicalpersonnel can accomplish their individual clinical tasks.

'S CO




FIGURE 5. Layout breakdown by resuscitation and imaging.


OY

Equipment issues and usability issues were far less populatedthan the aforementioned RIPCHORD-TWA categories (12 and29 total disruptions, respectively), making it difficult to draw con-clusions based on the results of the data analysis.

As disruptions accumulate, a system “drifts”more and moretoward disorder. To borrow logic from Dekker, it is the very ab-sence of adverse events associated with disruptions that in-creases the likelihood of drift toward disorder because theseseemingly minor events are perceived as unrelated to safetyand efficiency.21 Dekker further bolsters this argument by sug-gesting that Murphy’s law is wrong in that “what can go wrong,usually does not.”21 Working day-to-day with disruptions thatfail to exhibit their inherent potential for detrimental conse-quence renders them easy to ignore. Conceptualizing these dis-ruptions as threat windows gives weight to these events,making them less likely to “fly under the radar.” By reframingour approach in such a manner, researchers and health care pro-fessionals are better equipped to identify and mitigate threats topatient safety and the system of care.

Traditionally, the health care industry has used error manage-ment systems that are reactive in their approach to treating systembreakdown. However, managing errors and adverse events is nodifferent than treating a patient who is already sick. The industrywould be better served to address errors by proactively identifyingunderlying symptoms before they have the opportunity to mani-fest themselves. Those threats associated with communication, in-terruptions, and coordination issues identified in this investigationpoint specifically to areas of weakness in the system. This empir-ical evidence is instrumental in developing comprehensive inter-ventions that can help to optimize the delivery of care to thetrauma patient.

Future efforts will focus on the implementation of severaltargeted interventions that are geared toward remediating processinefficiencies revealed by the identification of threat windows intrauma care. Rather than using a traditional interventional ap-proach, which tends to fall flat in the real world, we defer to theapplication of an interventional technique driven by empiricalfindings that rely on input from multidisciplinary health care pro-fessionals. This method allows practitioners on the front line toimplement customized interventions to problems they face on aregular basis. The benefit of this method is that it allows the "peo-ple in the trenches," or those individuals who have intimateknowledge of the existing threats to safety, to weigh in with re-spect to improving performance in their workplace.

The next stage of this study will involve the staggered imple-mentation of several specific interventions. Subsequent analysisof these data will provide a new point of comparison for measureagainst the baseline/pre-intervention data to gauge the effectiveness

AUTHOR



of relevant interventions. Outcome measures such as reductions inmedical errors, adverse events, and mortality would also provideevidence of the efficacy of these specific interventions. Ultimately,this approach may prove more successful and lasting in mitigatingthreats to the delivery of traumatic injury care.

P
CONCLUSIONS
Historically, the health care industry has placed emphasis onthe connection between adverse events and negative patient out-comes. Despite enormous efforts to address lapses in patientsafety, the medical community is still struggling to reduce patientmorbidity and mortality directly related to preventable errors andadverse events. It is clear that we should explore additional ap-proaches to solving this problem.

Borrowing models from other disciplines, such as the brokenwindows theory, allows us to shed new light on process inefficien-cies that potentially jeopardize the integrity of patient care. Con-ceptualizing the accumulation of FDs as threat windows allowsresearchers to systematically identify underlying process ineffi-ciencies and present them in terms of tangible threats to safetyand quality of care. The study of threat windows exposes processinefficiencies earlier in the chain of events, thereby affording usthe opportunity to intervene well before a potential error or ad-verse event occurs.

'S C

REFERENCES1. Centers for Disease Control and Prevention (CDC). Guidelines for field

triage of injured patients: recommendations of the national expert panel onfield triage (MMWR, 61(1), 1–20). Atlanta, GA: Division of InjuryResponse, National Center for Injury Prevention and Control, CDC; 2015.

2. Rhee P, Joseph B, Pandit V, et al. Increasing trauma deaths in theUnited States. Ann Surg. 2014;260:13–21.

3. MacKenzie E, Rivara F, Jurkovich G, et al. A national evaluation of theeffect of trauma-center care on mortality. N Engl J Med. 2006;354:366–378.

4. Wiegmann D, ElBardissi A, Dearani J, et al. Disruptions in surgical flowand their relationship to surgical errors: an exploratory investigation.Surgery. 2007;142:658–665.

5. Palmer G 2nd, Abernathy J 3rd, Swinton G, et al. Realizing improvedpatient care through human-centered operating room design: a humanfactors methodology for observing flow disruptions in the cardiothoracicoperating room. Anesthesiology. 2013;119:1066–1077.

6. Blocker R, Duff S, Wiegmann D, et al. Flow disruptions in trauma surgery:type, impact, and affect. Proc Hum Fact Ergon Soc Annu Meet. 2012;56:811–815.





7. Blocker RC, Shouhed D, Gangi A, et al. Barriers to trauma patient careassociated with CT scanning. J Am Coll Surg. 2013;217:135–143.

8. Catchpole K, Gangi A, Blocker R, et al. Flow disruptions in trauma carehandoffs. J Surg Res. 2013;184:586–591.

9. Shouhed D, Blocker R, Gangi A, et al. Flow disruptions during trauma care.World J Surg. 2014;38:314–321.

10. Catchpole K, Ley E, Wiegmann D, et al. Human factors subsystems intrauma care. JAMA Surg. 2014. Available at: http://archsurg.jamanetwork.com/article.aspx?articleid=1891733. Accessed August 28, 2014.

11. Sarcevic A. A study of collaborative information behavior in traumaresuscitation [presentation]. GROUP ’09 Sanibel, FL; 2009.

12. Wilson J, Kelling G. Broken windows: the police and neighborhood safety.Atl Mon. 1982;211:29–38.

13. Harcourt B. Reflecting on the subject: a critique of the social influenceconception of deterrence, the broken windows theory, andorder-maintenance policing New York style.Mich Law Rev. 1998;97:291–389.

14. Hesketh K, Duncan S, Estabrooks C, et al. Workplace violence in Albertaand British Columbia hospitals. Health Policy. 2003;63:311–321.


AUTHOR


15. Sampson R, Raudenbush S. Systematic social observation of public spaces:a new look at disorder in urban neighborhoods. AJS. 1999;105:603–651.

16. Taylor R. Crime, grime, fear, and decline: a longitudinal look [research inbrief]. Washington, DC: National Institute of Justice; 1999.

17. Taylor R. Breaking away from broken windows: Baltimore neighborhoodsand the nationwide fight against crime, grime, fear, and decline. Boulder,CO: Westview Press; 2001.

18. Gruen R, Jurkovich G, McIntyre L, et al. Patterns of errors contributing totrauma mortality: lessons learned from 2,594 deaths. Ann Surg. 2006;244:371–380.

19. Pucher PH, Aggarwal R, Twaij A, et al. Identifying and addressingpreventable process errors in trauma care. World J Surg. 2013;37:752–758.

20. Thomas C, Bertram E, Johnson D. The SBAR communication technique:teaching nursing students professional communication skills. Nurse Educ.2009;34:176–180.

21. Dekker S. The Field Guide to Human Error. Bedford, UK: CranfieldUniversity Press; 2006:163–164.

PY


'S CO


http://archsurg.jamanetwork.com/article.aspx?articleid=1891733

http://archsurg.jamanetwork.com/article.aspx?articleid=1891733


broken windows, trauma

Documents