Effects of team coordination during cardiopulmonary resuscitation: A...

18
Effects of team coordination during cardiopulmonary resuscitation: A systematic review of the literature Ezequiel Fernandez Castelao a, , Sebastian G. Russo b , Martin Riethmüller a , Margarete Boos a a Department of Social and Communication Psychology, Georg-August-University Göttingen, Germany b Department of Anesthesiology, Emergency and Intensive Care Medicine, University Medical Centre Göttingen, Germany Keywords: Cardiopulmonary resuscitation; Advanced life support; Coordination; Communication; Leadership; Planning; Teamwork Abstract Purpose: The purpose of this study is to identify and evaluate to what extent the literature on team coordination during cardiopulmonary resuscitation (CPR) empirically confirms its positive effect on clinically relevant medical outcome. Material and Methods: A systematic literature search in PubMed, MEDLINE, PsycINFO and CENTRAL databases was performed for articles published in the last 30 years. Results: A total of 63 articles were included in the review. Planning, leadership, and communication as the three main interlinked coordination mechanisms were found to have effect on several CPR performance markers. A psychological theorybased integrative model was expanded upon to explain linkages between the three coordination mechanisms. Conclusions: Planning is an essential element of leadership behavior and is primarily accomplished by a designated team leader. Communication affects medical performance, serving as the vehicle for the transmission of information and directions between team members. Our findings also suggest teams providing CPR must continuously verbalize their coordination plan in order to effectively structure allocation of subtasks and optimize success. © 2013 Elsevier Inc. All rights reserved. 1. Introduction Every 5 years, international resuscitation organiza- tions (eg, such as the European Resuscitation Council) publish updated guidelines based on the latest medical research on cardiopulmonary resuscitation (CPR) [1]. The Advanced Life Support (ALS) Guidelines provide an algorithm for how CPR subtasks should be optimally organized and synchronized during the resuscitation process. Task synchronization requires assignment and coordination of responsibilities among rescue team members, which can be an additional challenge yet essential for efficiency of CPR as well as patientsand team memberssafety [2,3]. During CPR, the team of health care providers typically functions in a setting characterized by high levels of stress [4], time pressure, and impending danger to the patient [5]. It is because of these inherent characteristics of CPRhigh- stakes, complex, team-administered, and clarity of shared goalthat we conduct a systematic literature review of team Corresponding author. Tel.: + 49 551 397954; fax: + 49 551 3912496. E-mail address: [email protected] (E. Fernandez Castelao). 0883-9441/$ see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcrc.2013.01.005 Journal of Critical Care (2013) 28, 504521

Transcript of Effects of team coordination during cardiopulmonary resuscitation: A...

Journal of Critical Care (2013) 28, 504–521

Effects of team coordination during cardiopulmonaryresuscitation: A systematic review of the literatureEzequiel Fernandez Castelao a,⁎, Sebastian G. Russo b,Martin Riethmüller a, Margarete Boos a

aDepartment of Social and Communication Psychology, Georg-August-University Göttingen, GermanybDepartment of Anesthesiology, Emergency and Intensive Care Medicine, University Medical Centre Göttingen, Germany

(E

0h

Keywords:Cardiopulmonaryresuscitation;

Advanced life support;Coordination;Communication;Leadership;Planning;Teamwork

AbstractPurpose: The purpose of this study is to identify and evaluate to what extent the literature on teamcoordination during cardiopulmonary resuscitation (CPR) empirically confirms its positive effect onclinically relevant medical outcome.Material and Methods: A systematic literature search in PubMed, MEDLINE, PsycINFO andCENTRAL databases was performed for articles published in the last 30 years.Results: A total of 63 articles were included in the review. Planning, leadership, and communication asthe three main interlinked coordination mechanisms were found to have effect on several CPRperformance markers. A psychological theory–based integrative model was expanded upon to explainlinkages between the three coordination mechanisms.Conclusions: Planning is an essential element of leadership behavior and is primarily accomplished by adesignated team leader. Communication affects medical performance, serving as the vehicle for thetransmission of information and directions between team members. Our findings also suggest teamsproviding CPR must continuously verbalize their coordination plan in order to effectively structureallocation of subtasks and optimize success.© 2013 Elsevier Inc. All rights reserved.

1. Introduction organized and synchronized during the resuscitation

Every 5 years, international resuscitation organiza-tions (eg, such as the European Resuscitation Council)publish updated guidelines based on the latest medicalresearch on cardiopulmonary resuscitation (CPR) [1].The Advanced Life Support (ALS) Guidelines providean algorithm for how CPR subtasks should be optimally

⁎ Corresponding author. Tel.: +49 551 397954; fax: +49 551 3912496E-mail address: [email protected]

. Fernandez Castelao).

883-9441/$ – see front matter © 2013 Elsevier Inc. All rights reserved.ttp://dx.doi.org/10.1016/j.jcrc.2013.01.005

.

process. Task synchronization requires assignment andcoordination of responsibilities among rescue teammembers, which can be an additional challenge yetessential for efficiency of CPR as well as patients’ andteam members’ safety [2,3].

During CPR, the team of health care providers typicallyfunctions in a setting characterized by high levels of stress[4], time pressure, and impending danger to the patient [5]. Itis because of these inherent characteristics of CPR—high-stakes, complex, team-administered, and clarity of sharedgoal—that we conduct a systematic literature review of team

505Effects of team coordination during CPR

coordination in CPR to establish how team coordinationcontributes to the quality of CPR.

Team coordination is defined as the management ofinterdependencies of subtasks by regulated action andinformation flow in order to achieve a common goal, thatis, the performance of high quality CPR [6]. This comprisestask management, leadership, and communication as essen-tial prerequisites of effective teamwork and, thus, theresulting quality of CPR and patient safety [7]. Faulty teamcoordination has been widely recognized as a factorcontributing to medical errors [8], and the positive relation-ship between effective teamwork and patient outcome issupported in other dynamic domains of healthcare [9].

The coordination requirements of CPR are well describedby Tschan et al [10], who provide a hierarchical task analysis(HTA) on the basis of the aforementioned ALS Guidelines.Their HTA identifies the goal dependencies which emergencyteams have to manage during CPR and what types ofcoordination mechanisms are suitable to successfully accom-plish these goal dependencies. The core goals outlined are (1)diagnosing the cardiac arrest, (2) oxygenating the brain, and(3) attempting to re-establish spontaneous blood circulation,which are further broken down into hierarchies of subgoals.

The benefits of effective team coordination in CPR aresubstantive. For example, Fernandez Castelao et al [11]showed a positive relationship between well phrasedleadership utterances, functioning as a coordination mech-anism, and no-flow time. The provision of less or poorlyarticulated leadership utterances was found to be related totreatment delays, which in turn correlate with morbidity [12].

International guidelines [1,13] all focus on describing theoptimal sequence of medical actions during a CPR butwithout specific reference to coordination as the facilitatorfor seamless transition between completions and beginningsof each subgoal to avoid time loss and thereby increasesurvival rate [14]. Based on task analysis [10] and relatedempirical findings [9,15], team coordination and CPRperformance seem to be closely linked. The literature alsocontains a rich body of research regarding team coordina-tion methods in emergency medical teams, how the natureof a task defines whether explicit or implicit methods areoptimal, and how these 2 general methods interact toachieve team goals [16–18].

This review examines whether and to what extent teamcoordination in CPR is empirically confirmed as affectingclinically relevant outcome, with particular focus on findingsthat contribute to improvement of current ALS Guidelines.Furthermore, the quality of the included articles will beassessed by defining level of evidence (LOE)—whichindicates risk of bias—and level of concreteness (LOC)—which indicates lucidity of the reckoning coordinationbehavior. Finally, we aim to organize the findings intomeaningful patterns by deducing a coordination mechanismmodel based on Hacker's (2003) action regulation theory(ART) [19]. As the ART perceives thought and action ofindividuals as an interlinked process it was widely used as a

theoretical approach for structuring and designing workprocesses [20]. Thus, the ART provides the necessarybackground for creating a model to reflect and analyzegoal-oriented interpersonal action processes like CPR inorder to improve effectiveness [19].

2. Methods

2.1. Search strategy for identification of relevantpapers

The procedure to identify publications relevant for thisreview is based on the Cochrane guidelines for designingsystematic reviews [21].

2.1.1. Step 1A computerized literature search of electronic databases

PubMed, MEDLINE, PsycINFO, and CENTRAL was per-formed in September 2011, limited to the last 30 years. Thesearch focus was journal articles on potential impact ofcoordination behavior regarding CPR, using all meaningfulcombinations of two groups of search terms (Fig. 1). Keywordcombinations within each of the 2 groups were not applied.Duplicateswere excluded from the sample resulting fromStep 1.

2.1.2. Step 2All study titles were screened by the lead author (EFC) for

relevance to this review. To assure selection reliability, asecond author (MR) independently reviewed a random 20%subset of the titles (n = 1008). The inter-rater agreement wasalmost perfect (κ = 0.91) [22].

2.1.3. Step 3The abstracts of the remaining publications were further

screened for relevance. Inter-rater agreement between EFC andMR—based on the decision whether to include or discard eachof the remaining articles (n = 462)—at this step was also almostperfect (κ = 0.81) [22]. The references of the end-selectedarticles were perused to identify further appropriate literature,which were then added to the starting pool of this same step.

2.1.4. Step 4The remaining articles were screened according to our

two criteria: (1) the main findings offer empirical evidence ofteam coordination in CPR or related medical emergencies(eg, trauma resuscitation) concerning a link betweencoordination efforts and performance effectiveness; and (2)the article was published in a peer reviewed journal. Onceagain, inter-rater reliability between EFC and MR wascalculated based on the decision whether to include ordiscard the remaining articles (n= 191) (κ = 0.81) [22].

2.1.5. Step 5The selected articles were sorted into topic categories and

assessed for quality, with particular attention to the

Fig. 1 Selection flowchart. *In step 5, several of the 63 articles were categorized into more than one topic.

506 E. Fernandez Castelao et al.

development of coordination behavior classes influencingCPR performance. The categories were inductively generat-ed out of the focal contents of the studies in order to create acomprehensible overview of the findings, simultaneouslyproviding an expandable basis for the allocation of futureempirical results.

2.2. Article quality assessment

The LOE scales we applied are a unified version of theInternational Liaison Committee on Resuscitation (ILCOR) LOEscales: scale for studies of therapeutic interventions, scale forprognostic studies and scale for diagnostic studies [23]. TheILCOR LOE scales are used as part of a standard approach tocreate systematic reviews in order to support the development andupdate of guidelines for the management of several cardiovas-

cular emergencies, including cardiac arrest. ILCOR expertsdeveloped the LOE scales after a detailed review of comparableclassifications in common use aiming to ensure external validity[24]. Eachof the reviewed classifications—for example,GRADE[25]—was designed by expert task forces who in turn consultedat that time existing rating systems used for the generation ofseveral treatment recommendations for different medical fields[26]. Thus, the quality of the studies included in this review can beassessedwith confidence by applying a rating procedure based onthe ILCOR scales. For our purpose we generalized the ILCORLOE scales to their common quality criteria—the potential forbias—into one single scale to make the assessment easier toapply. Based on our LOE scale (lower number, less potential ofbias), studies rated LOE 1 (highest score) were those that (a)provide qualitative, quantitative, and temporal aspects of theimpact of coordination onCPRoutcome, and (b)minimize risk of

507Effects of team coordination during CPR

bias (eg, randomized controlled studies). Studies rated LOE 2were those without true randomization (eg, alphabeticalrandomization). If information included in a study was archival(eg, analysis of medical records), we rated it LOE 3. Case studieswere rated LOE 4. Studies not directly related to specific contextof CPR were rated LOE 5, including reviews in which nocomparable evidence criteria were defined.

The LOC, defined as “clarity of the described/measuredcoordination behavior,” was rated using a 5-point Likert scaledeveloped by the authors. LOC score (1) was awarded if thedescription of the investigated, discussed, and/or recom-mended coordination behavior was so clearly depicted that thereader would be able to exactly reconstruct the behavior,including the appropriate moment of its initialization. Theinitial inter-rater agreement for the LOE and LOC rating wasκ = 0.84 and κ = 0.74, respectively; all disagreement s werediscussed afterwards by the authors to reach full agreement.Table 1 describes the LOE and LOC scales in detail.

3. Results

3.1. LOE and LOC assessments

The flow diagram (Fig. 1) shows the selection procedureof the papers included in our review, which produced 63

Table 1 Definitions, numbers and percentages of LOE and LOC allo

LOELOE 1: Randomized controlled studies, meta-analyzes a

reviews of randomized controlled studies whichto our research question: identifying and quantiteam coordination and clinically relevant medic

LOE 2: Studies using concurrent controls without true rmeta-analyses or systematic reviews of these ty(eg, non-controlled pre-post studies).

LOE 3: Retrospective studies (eg, studies based on pastLOE 4: Studies with no control group (eg, case studies)LOE 5: Studies are not directly related to the research q

(eg, trauma resuscitation) but they provide infoincluded in this review.

LOCLOC 1: Coordination behavior is depicted in detail. (eg

of verbalization is documented word by word).basis of these descriptions an accurate reconstrucoordination is easily possible.

LOC 2: Coordination behavior is depicted in detail butbehaviors remain intangible to some extent.

LOC 3: Coordination behavior is depicted partly in detaextent these behaviors cannot be simply recons

LOC 4: Coordination behavior is partly depicted on a vperfunctory level.

LOC 5: Coordination behavior is mainly depicted on alevel. A concrete reconstruction of these behavidifficult, due to missing details (eg, “good” com

papers for thematic categorization as well as LOE and LOCratings. Of these 63 papers, 4 (6.34%) were rated LOE 1; 5(7.93%), LOE 2; 2 (3.17%), LOE 3; 18 (28.57%), LOE 4; and34 (53.96%), LOE 5. Regarding LOC, 12 papers (19.04%)were rated LOC 1; 10 (15.87%), LOC 2; 11 (17.46%), LOC3; 14 (22.22%), LOC 4; and 16 (25.39%), LOC 5. Themedian LOE and LOC were 5 and 3, respectively. Table 2presents the characteristics of the included studies.

3.2. Categorization by topic

Empirical evidence of the papers included in this reviewrendered three interlinked team coordination mechanismswithin CPR treatment: (1) planning, (2) leadership, and (3)communication. Because these mechanisms are associated witheach other and in some cases were included within the samearticle, some papers were classified into more than one topiccategory. All papers were thematically allocated into at leastone of the above 3 categories.

3.3. Integrative model of team coordinationmechanisms affecting CPR performance

Our integrative coordination mechanism model (Fig. 2)summarizes the literature findings. This model is acombination of a deductive and inductive reasoning process,

cations

No. of allocated studies (%)

nd systematicare directly relatedfying the link betweenal outcomes.

4 (6.34%)

andomization,pe of studies

5 (7.93%)

reports). 2 (3.17%). 18 (28.57%)uestionrmation to be

34 (53.96%)

, contentOn thection of effective

12 (19.04%)

some of these 10 (15.87%)

il but to sometructed.

11 (17.46%)

ery 14 (22.22%)

very perfunctoryors turns out to bemunication).

16 (25.39%)

508 E. Fernandez Castelao et al.

resulting in a dynamic organization and a linkage between thethree main coordination mechanisms revealed in this review:planning, leadership, and communication. The inductive partof our reasoning process consisted in the definition of ourthree topic categories—planning, leadership, and communi-cation—based on the coordination mechanism(s) mainlyfocused by the authors of each study included in this review.The deductive part in the screening consisted of pertinentteamwork frameworks—eg, ‘Big Five’ of Teamwork [27],the Structure of Teamwork Behaviors [28], or the FEMPIPO

model of group coordination [16]—in order to confirm ourdefined topic categories as coordination mechanisms affect-ing team performance. In fact, planning, leadership andcommunication are interrelated coordination mechanismsempirically confirmed as performance influencing mecha-nisms in all types of team tasks. As planning, leadership andcommunication are coordination mechanisms which cannotbe completely regarded as disjunct, we postulated a partialinterrelation between them. The interrelation is based to thefacts that, effective communication is needed for transmittingplans, planning is defined as a function of leadership and inturn communication is needed to be an effective leader. In ourmodel the interrelations of planning, leadership and commu-nication are represented by the overlapping of the innercircles (Fig. 2, inner core). Excerpts from articles included inthis review, which are highlighted in the next paragraphs,clarify these interrelations.

The structure of our integrated model is based on atheoretical reasoning. Adaptive application of the threecoordination mechanisms affects team members’ actionregulation to obtain an ideal result [19]. According to thestructure of the action process [29], our model exposes therequirement for CPR teams to first orientate themselveswithin the clinical situation (eg, by monitoring the patient),conceptualize a treatment procedure (eg, plan whether toattempt defibrillation depending on the assessment of theinitial rhythm), orchestrate and execute the appropriateactions (eg, first defibrillation and then check breathing), andevaluate and modify strategy (eg, heart rate check) (Fig. 2,four elements of outer ring).

The model is a meaningful complement of Tschan's HTAof the CPR task [10] as it merges empirical findings related tocoordination demands with a theoretical background (Hack-er's ART) [19]. Furthermore, on the basis of the ART—theframe of our model—it becomes apparent which steps teamshave to run through to accomplish each subgoal.

In sum, by inter-connecting the coordination mechanismsand each of the ART phases and explicitly stating that thefour phases of the CPR action cycle are iterative, weunderline that, depending on the subtask at hand, teamsproviding CPR repeatedly run through these action phases.In order to accomplish these phases, teams constantly havethe aforementioned 3 interrelated coordination mechanismsat their disposal: planning, leadership and communication.These 3 coordination mechanisms are equally important tosucceed in CPR, but according to the ART, each phase is

characterized by particular coordination demands. Thus, ourmodel depicts the interrelated coordination mechanisms(Fig. 2, inner core) and the iterative CPR coordinationdemands (Fig. 2, outer ring) to accomplish each subtask.

Evidence for each mechanism of the model and theirinterrelations are depicted in the following 3 paragraphs.

3.4. Planning as a CPR coordination mechanisminfluencing medical performance

Our analysis of the empirical evidence identified“planning” as a key performance mechanism affectingcoordination requirements in CPR situations. A total of 26papers were categorized into this topic.

We also detected in ten studies that leadership tasks ofprioritization and distribution of subtasks among the teammembers—before the CPR begins (pre-process) as well asduring the CPR treatment (in-process)—is statisticallyrelated to a variety of outcome: reduced hands-off time[30], faster treatment completion [31–33], higher algorithmadherence [2,34–37], and prevention of interruptions [38].

Pre-process planning in three studies was positivelyassociated with return of spontaneous circulation (ROSC)[39], and initiation of crucial treatments (eg, time to firstdefibrillation) [40,41].

Simulation training as a means to improve planningbehavior of CPR teams was successfully evaluated in sixstudies; improvements ofmedical outcomeswere reported aftermultidisciplinary resuscitation trainings [42,43]. In one study,the impact of the training on treatment times was rescinded bythe integration of new team members into concomitant plans[44]. Two other team trainings evoked augmented planningactivities by team members [45,46]. Both training conceptsaimed to enhance CPR performance by teachingCPR planningskills, although CPR performance quality as a potential resultof improved planning was not quantified. Students reported tobe aware of the importance of planning within CPR aftercompleting a crisis resource management (CRM) trainingincluding planning recommendations [47].

Planning behavior is empirically proven to directly affectthe improvement of chest compressions (CC)—an essentialCPR performance marker—by ensuring the minute-by-minute rotation of CC providers [48,49].

Planning is closely related to leadership behavior.Participants of 2 interview studies underline that an effectivein-process task distribution can be hindered by “hands-on”acting of the team leader (TL), which in turn reduces theleader's capacity to overview the process [50], and that theearly establishment of a plan should be clearly transmitted tothe team by the TL [51]. One literature review focused onteamwork behavior in dynamic domains of healthcare,including CPR teams, and reported that task distribution(mostly performed by the TL) is one important aspectrelevant to the quality of patient care [9].

In summary, pre- and in-process planning, i.e. taskdistribution, role assignment, and timing of tasks, are crucial

Table 2 Main characteristics of the studies included in this review

Authors, year ofpublication [Ref.]

Aim of the paper LOE LOC Topic category Setting Major findings

PL LDS COM

Adams et al(2005) [67]

Determination of which physician-staffing model ― emergency medicineresidents or staff hospitalist physicians― is best for cardiac arrest teams.

4 5 x Code blueactivation

Teams led by emergency medicineresidents reestablished ROSC faster.No differences in survival rates.

Andersen et al(2010) [50]

Identification of non-technical skillssuitable for improving CPR teamperformance anddescription of barriers to theuse and implementation ofsuch skills.

5 1 x x x Undefined List of recommendednon-technical behavior with regard tothe improvementof resuscitation training.

Arshid et al(2009) [56]

Determination of CPR quality duringpediatric resuscitation training sessionsand to assess the awareness of the teamleaders on the CPR quality and theirteam members.

5 5 x Simulation Suboptimal CPR performance. Teamleaders failed to addresspoor CPR performance.

Bergs et al(2005) [84]

Evaluation of communicationduring multidisciplinary traumaresuscitation.

5 3 x Emergencyroom

Knowledge transfer is not optimal.Trend towards bettercommunication during the exposure ofseverely injured patients. Higherdiversity in major teams (12 persons) incomparison to minorteams (b12).

Bernhard et al(2007) [41]

Investigation of whetherthe intervals between admission,diagnosis, and treatment can bereduced and if mortality can beimproved by employing treatmentalgorithms.

4 3 x Emergencyroom

Significant reduction of intervals andmortalityafter algorithm introduction.

Blum et al(2005) [87]

Evaluation of information sharingduring critical event simulationtraining.

5 5 x Simulation Significant difference ininformation sharing from the 1stscenario to the final scenario. Trainingwas reported as useful.

Bogenstätter et al(2009) [85]

Examination of factorsinfluencing the accuracyof information transmitted to nursesand physiciansjoining a medicalemergency situation.

4 4 x Simulation Information transmitted to teammembers who join a CPR process isonly partly reliable.

(continued on next page)

509Effects

ofteam

coordinationduring

CPR

Table 2 (continued)

Authors, year ofpublication [Ref.]

Aim of the paper LOE LOC Topic category Setting Major findings

PL LDS COM

Bradley et al(2009) [43]

Identification of effects ofinterprofessionalresuscitation skillstraining―including attitudes,leadership, teamwork, andBLS skills―on medical andnursing students.

4 4 x Simulation Training improved leadership,coordination behavior and BLS skills.

Capella et al(2010) [60]

Evaluation of formal teamtraining with regard toimprovement of teamworkin the trauma resuscitationbay and its link to clinicaloutcome.

4 5 x x Simulation Improvement of teamworkand medical performance afterintervention.

Cole & Crichton(2006) [51]

Ethnographic exploration ofthe culture of a trauma teamin relation to human factors.

5 4 x x Emergencyroom

Leadership performance andexperience influence teamperformance. Leadership has to betrained specifically. Role awarenessand consideration of the others’ skillshave positive influence on teambehavior. Conflict is detrimental.Communication affects cohesion.

Cooper (2001) [76] Evaluation of theeffectiveness of a leadershipdevelopment seminar.

2 2 x Simulation Training improves leadershipperformance. Training was widelyaccepted.

Cooper &Wakelam(1999) [58]

Determination of thelinkages between leadershipbehavior, teamcommunication, and CPRperformance.

5 1 x Resuscitationt+eam call

Correlations between structuringbehavior by the team leader andadequate CPR performance. Teamleader structuring behavior isnegatively affected by hands-onparticipation.

De Vita et al(2005) [42]

Experience report of a crisis teamsimulation training that is focused onteam organization.

4 3 x Simulation Role allocation seems to lead to betterperformance when combined with teamskills training.

Driscoll & Vincent(1992) [40]

Determination of to what extent timedifferences within trauma resuscitationare affected by the structure of thetrauma team.

4 4 x Emergencyroom

The fastest teams are those where taskshave been allocated before the patientarrives. The risk of personnel listeningor reporting to the team leadermoderated by the number of teammembers.

510E.

FernandezCastelao

etal.

Dwyer & Williams(2002) [88]

Exploration of factors thatmay be associated with nursesparticipating in resuscitation.

5 5 x Undefined Delays caused by nurses occur due toinsecurity during CPR. Their teammembers might influence this level ofinsecurity.

Fernandez Castelaoet al (2011) [11]

Evaluation of the impact of video-basedinteractive CRM training on NFT and onproportions of team memberverbalizations during simulated CPR.Further, to investigate the link betweenteam leader verbalization accuracy andNFT.

1 1 x x Simulation CRM training reduces NFT andimproves team leader verbalizationproportions.

Flanagan et al(2004) [47]

Exploration of the role of simulation inproviding CRM trainings to manageevolving critical situations.

5 3 x x x Simulation Importance of CRM documented bystatements of course participants.Coordination requirements in criticalsituations are listed.

Gaba et al(1998) [37]

Evaluation of ratings of performancethrough the assessment of clinical andteamwork performance duringsimulated crises.

5 4 x x x Simulation Correlations exist between teambehavior and technical score.Behavioral rating system can beimproved.

Gilfoyle et al(2007) [75]

Presentation and evaluation of aneducational intervention with regard tolong term increase of leadership skills.

2 4 x x Simulation Educational intervention does improveleadership skills. Skills were retainedafter 6 months.

Hayes et al(2007) [73]

Determination of internalmedicine residents’perceptions of the adequacy of theirtraining to serve as in-hospital cardiacarrest team leaders.

5 5 x Undefined Perceived deficits in the training ofresidents ― lack of feedback andsupervision ― to adequately functionas CPR team leaders.

Henderson &Ballesteros(2001) [39]

Assessment of the impact of a formal,structured resuscitation team forin-hospital cardiopulmonaryresuscitation over the year following itscreation.

4 5 x Code blue Formation of formalized, well trainedCPR team was associated with betterROSC rates.

Hoff et al(1997) [66]

Evaluation of the effect of an identifiedcommand–physician on resuscitationperformance.

4 4 x Emergencyroom

Adherence to ALS algorithmsignificantly higher in teams in which acommand-physician was identified.

Holcomb et al(2002) [46]

Validation of a simulator as anevaluation tool of trauma teamresuscitation skills.

4 2 x x Simulation Significant improvement in teamperformance after 28-day traumarefresher course.

Høyer et al(2009) [36]

Description of physician behavior asteam leaders in a simulated cardiacarrest during inter-hospital transfer.

5 5 x Simulation Junior physicians performed CPR well.Deficiencies in task delegation werefound in junior physicians.

(continued on next page)

511Effects

ofteam

coordinationduring

CPR

Table 2 (continued)

Authors, year ofpublication [Ref.]

Aim of the paper LOE LOC Topic category Setting Major findings

PL LDS COM

Hunt et al(2008) [62]

Characterization of the quality ofresuscitation efforts and identificationof targets for educational interventions.

5 3 x x Simulation Low adherence to algorithms, lack ofleadership, and communicationmisdiagnoses were identified duringCPR.

Hunziker et al(2011) [52]

Description of the state of the sciencelinking team interactions to theperformance of CPR.

1 2 x Misc. A prolonged process of team buildingand poor leadership are associated withsignificant shortcomings in CPR.

Hunziker et al(2010) [12]

Impact comparison between leadershipinstruction and general technicalinstruction on simulated CPR scenarioperformance.

2 2 x Simulation Both interventions improve CPRperformance. Leadership instructionwas rated superior.

Hunziker et al(2009) [30]

Exploration of effects of ad hoc team-building on the adherence to thealgorithms of CPR among generalpractitioners and hospital physicians.

2 1 x x Simulation Ad hoc formed teams had less hands-on time during early phase of CPR andmade less leadership statements.

Husebø et al(2011) [96]

Exploration and description of verbaland non-verbal coordinationmechanisms employed by studentsduring a simulated CPR.

5 1 x Simulation Coordination implies a combination ofbodily conduct and gestures andverbalizations.

Huseyin et al(2002) [49]

Establishment of the length of timeover which an individual can maintaineffective chest compressions.

4 5 x Simulation Best performance is maintained if CCroles changes every minute.

Hynes et al(2006) [64]

Specification of leadership challengesduring resuscitations.

5 3 x Undefined To manage team member aggressivebehavior, effective task delegation andcommunication were identified as teamleader challenges.

Khetarpal et al(1999) [68]

Investigation of how the presence of anattending trauma surgeon duringtrauma team activation affectstreatment times.

4 3 x Trauma teamactivations

Presence of trauma surgeon (leadingperson) leads to reduce treatment times.

Klein et al(2006) [54]

Gaining new theoretical insights intoteam leadership in trauma resuscitationunits.

5 1 x Emergencyroom

Recommendation of dynamicdelegation (shared leadership).

Kozer et al(2004) [82]

Examination of the incidence andnature of medication errors duringsimulated pediatric resuscitations.

5 2 x Simulation Physicians and nurses made errors,even lethal errors; errors sometimesdue to incomplete orders.

Künzle et al(2010) [53]

Systematic review of effectiveleadership strategies in critical careteams.

5 3 x Misc. Effective leaders play a pivotal role inpromoting team performance.

512E.

FernandezCastelao

etal.

Lo et al(2009) [44]

Description and evaluation ofeffectiveness of weekly pediatricresuscitation training.

4 4 x Simulation Weekly training improves skills, andthe introduction of new leading andnon-leading team members leads toincreased treatment times.

Lubbert et al(2009) [61]

Analysis of trauma team functioningwith respect to adherence torecommended guidelines.

5 5 x Emergencyroom

Inadequate leadership was related to ahigher number of algorithm violations.

Mäkinen et al(2007) [57]

Development of a method to assessCPR skills of nurses and to contributeto the construction of an educationalprogram.

3 5 x Simulation Defining and teaching leadershipseems to improve CPR performance.

Mann & Heyworth(1996) [59]

Impact evaluation of ALS and ACLSteam leader training on resuscitationteam performance.

4 5 x Resuscitationroom

Teams led by ALS or ACLS trainedleaders showed reduced defibrillationand drug administration times.

Manser (2009)[9]

Identification of aspects of teamworkaffecting quality and safety ofhealthcare.

5 2 x x x Misc. Quality of collaboration, shared mentalmodels, coordination, communication,and leadership are relevant to thequality of healthcare.

Marsch et al(2004) [32]

Determination of whether and howhuman factors affect the quality ofCPR.

5 2 x x Simulation Failure was associated with absence ofleadership and explicit taskdistribution.

Marsch et al(2005) [70]

Determination of first responders’adherence to CPR algorithms CPR insimulated cardiac arrests in intensivecare.

5 4 x x Simulation First responders failed to build a teamstructure that ensured effectivemonitored and ongoing team activity.

Marsch et al(2004) [33]

Testing hypothesis that the process ofteambuilding affects the quality of CPRduring its early phase.

3 5 x Simulation Teams that were able to plan before theCPR treatment showed faster treatmenttimes. The teambuilding process (incl.planning) leads to treatment delays.

Meerabeau &Page (1999) [3]

Reflection upon aspects of teamworkwithin CPR treatment.

5 2 x Cardiologyward

CPR teams do not have time for ateambuilding process. Leadership isimportant.

Niemi-Murolaet al (2007) [77]

Examination of medical and nursingstudents’ beliefs and attitudes towardCPR and current guidelines.

4 4 x Undefined Only 42% of medical students and 20%of nursing students felt confident abouttheir ability to work as a CPR teamleader.

Nishiyama et al(2010) [48]

Comparison of time-dependentdeterioration of CC between two typesof CPR (CC-only vs conventional).

2 4 x Simulation CC quality rapidly declined.Recommendation of changing CC rolesafter 1 min to maintain quality of CC.

O´Brien (2001)[74]

Examination of interns' perceptions ofown performance; confidence inmanaging and/or participating insimulated and real CPR.

5 4 x x Simulation/undefined

Importance of leadership, teamcommunication, and role assignmentduring CPR was reported. Simulationtraining increases confidence anddecreases individual stress level.

(continued on next page)

513Effects

ofteam

coordinationduring

CPR

Table 2 (continued)

Authors, year ofpublication [Ref.]

Aim of the paper LOE LOC Topic category Setting Major findings

PL LDS COM

Oakley et al(2006) [97]

Comparison of two methods (videorecording vs. medical reports) toidentify management errors in traumaresuscitations.

5 5 Pediatricemergencyroom

Video recording is more effective fordetecting errors (drug administration,coordination, communication). 40% ofthe recorded teams showed lack inleadership and team functioning.

Østergaard et al(2004) [86]

Description of connections betweenmultiprofessional team training andpatient safety.

4 4 x Simulation Training participants report that poorperformance is based on lack ofcommunication and coordination skills.After the training, self-evaluationshowed NTS improvements.

Pittman et al(2001) [83]

Determination of CPR teamcommunication before and aftertreatment. Examination of teamcomposition and team leader selectionduring the treatment.

5 1 x Undefined Poor communication between teammembers before and after CPR. Teamleader selection is not always based ontheir CPR experience.

Quintana Rieraet al (2007) [98]

Evaluation of health professionalstolerance providing uninterrupted CC.

5 5 x Simulation Two minutes of uninterrupted CC iswell tolerated physically by healthprofessionals.

Ritchie & Cameron(1999) [72]

Analysis of trauma team leaderperformance using video recordings.

5 4 x Emergencyroom

Reticence of team leaders to verbalizetheir thoughts and clear instructionswas observed. Video recording is afeasible method to measure team leaderperformance.

Streiff et al(2011) [69]

Determination of predictors ofleadership behavior in CPR.

4 2 x x Simulation Amount of leadership statements isdetermined by gender and personalityand not by knowledge or experience.

Sugrue et al(1995) [71]

Assessment of the performance oftrauma team leaders and definition ofscope for improvement.

5 5 x Emergencyroom

High scores for medical performance.Deficiencies in communication anddelegation were found.

Thomas et al(2006) [34]

Measurement of teamwork behaviorsduring delivery room care and its linkto the quality of care.

5 2 x x x Emergencyroom

Teamwork behaviors significantlycorrelated with resuscitationcompliance.

Thomas et al(2010) [31]

Evaluation of two team trainings (high-fidelity and low-fidelity) and its impacton neonatal CPR quality.

1 1 x Simulation Teamwork behaviors significantlycorrelated with medical quality.

514E.

FernandezCastelao

etal.

Thomas et al(2007) [45]

Evaluation of the impact of a 1-dayneonatal CPR training.

1 1 x x Resuscitationroom

Training led to increased amount ofteamwork components (eg, informationsharing, Inquiry, assertion). Link tomedical quality was not investigated.

Tschan et al(2011) [99]

Determination of CPR team memberactivities and their attention focusduring unnecessary interruptions.

4 3 x Simulation Team members mainly observed themonitor or dealt with the defibrillatorduring unnecessary interruptions andtheir attention was focused on theseactivities.

Tschan et al(2006) [2]

Evaluation of the impact of leadershipmechanisms on CPR performance.

5 1 x x x Simulation Leadership and structuring partlyexplain differences in CPRperformance.

Weinstock et al(2008) [78]

Specification of the concept ofteamwork during trauma resuscitation.

5 3 Undefined Emphasis of importance of teamworkin emergencies. Appeal for more usefullearning methods, assessment tools,and qualitative scientific approaches.

Wright et al(2009) [35]

Examination of the link betweenobserved teamwork behavior andobjective measures of medicalperformance.

5 1 x x x Simulation Positive relationship betweenteamwork skills and clinical skills in asimulation scenario.

Xiao et al(2004) [100]

Exploratory analysis of team leaderfunctions relative to situationaldemands.

5 1 x Emergencyroom

Identification of six leadershipfunctions depending on the situation.

Yun et al(2005) [63]

Investigation of leadership andeffectiveness of teams performingtrauma resuscitation.

5 3 x Emergencyroom

Linkage of leadership to teameffectiveness, as moderated byrelatively specific situationalcontingencies.

LDS, Leadership; PL, Planning; COM, Communication.

515Effects

ofteam

coordinationduring

CPR

516 E. Fernandez Castelao et al.

to attain adherence to theALSalgorithms.The results fromourreviewreveal that thedesignationofaTLisameans tomaintainthe required in-process planning (i.e. task coordination).

3.5. Leadership behavior as a factor contributing toeffective CPR performance

The literature documents leadership as a substantialbehavior in managing CPR task complexity and positivelyinfluencing CPR performance outcomes. A total of 38 paperswere categorized into this topic.

The pivotal role of leadership within CPR is supported byfive review articles regarding CPR situations [52], as well asrelated medical fields (eg, trauma resuscitation) [9,51,53,54].Hunziker et al [52] deduce five principles for effective CPRleadership: (1) consider situational demands, (2) facilitatecontributions of the non-leading team members, (3) askproblem related questions, (4) keep hands-off, and (5)promote exchange of information. As a meaningful additionto these principles, St Pierre [55] defines further relevantleadership tasks that stipulate (6) to deduce a treatment planbased on medical knowledge and (7) to communicate anddistribute tasks of this plan by assigning tasks according toindividual skills and knowledge.

The TL role has an impact on various CPR performancemarkers as reported in 16 studies: provision of CC (rate,depth, discharge) [30,56–58], reduction of no-flow time(NFT) [2,11,12,30,56], improvement of ventilation quality[30,56–58], medication conformity [58,59], quality ofdefibrillation [58,59], and the adherence to algorithms(overall medical performance) [2,32,34,35,37,57,60–63].

Several core leadership behaviors, partly overlappingwith the above-mentioned principles, were considered to berequired to manage challenging CPR situations effectively.First of all, it is important to be easily identifiable as a TL andto keep hands-off during the treatment in order to maintain aprocedural overview [50]. Secondly, the TL should becapable of building a team structure by distributing tasks andcommunicating using comprehensible commands and re-sponses [3,50,64]. Finally, the leading team member shouldradiate calmness and positively reinforce the other teammembers on desirable actions at every occasion [50,65].

The presence of a designated TL leads to reducedtreatment times and optimized results [66]. Teams led byexperienced TLs re-establish ROSC faster but show nodifferences regarding survival rates [67]. The presence of aboard-certified surgeon who is considered by the team to bean expert for the case at hand reduces treatment decision timesduring straightforward as well as intrusive trauma resuscita-tions [68]. Interestingly, TL experience does not predict theamount of leadership statements (neither increases nordecreases) of students performing simulated CPR [69].

Failures in leadership—structuring failures, poor moni-toring of the ongoing process [70], delegation deficiencies[71], and verbal reticence [72]—are also empiricallydocumented. These deficits are observed in real-life settings

as well as in ALS training scenarios of medical residents[73]. On the other hand, leadership trainings are found to beeffective by increasing participants’ confidence and bydecreasing individual stress level while performing simulat-ed CPR as a TL [74] by elevating the quality of TLstatements [11,12] and by improving overall leadership skills[75,76]. Some of these studies focus on leadership behaviorand do not empirically take into consideration the impact ofleadership behavior on CPR performance [69,73–76].

Residents as well as students report to be unprepared tolead a CPR team [73,77] and underline the importance oftaking part in leadership trainings [47]. The result of twoexploration studies is the appeal to set a wider focus on thetraining of leadership skills after their examinations revealedthat leadership behavior—as an essential element ofteamwork—positively influences treatment quality [51,78].

In summary, the literature indicates trainable key functions(task distribution, monitoring, and comprehensible communi-cation) of leadership behavior. Furthermore, there is empiricalevidence that the designation of a TL and his/her experiencelevel is related to shorter treatment times. Accordingly, theinfluence of communication properties (eg, common speech,verbalization) on the quality of leadership and results deservesto be especially focused upon, as was repeatedly reported inthe literature [11,30,50,53,55,74,79].

3.6. Effective communication as a factorcontributing to effective CPR performance

The results reveal the influence of the process ofcommunication as a medium for information-sharing andleadership statements on the medical success of CPR. A totalof 23 papers were categorized into this topic.

In medical emergencies, communication (defined as thetransmission of information between one person to anotherperson or group) [80] fulfils four functions: (1) building andmaintaining team structure, (2) coordination of teamprocesses, (3) information exchange, and (4) facilitation ofinterpersonal relationships [55,79].

A direct link between different elements of communica-tion and CPR performance markers are reported in severalarticles: the accuracy level of TL verbalizations are found tobe negatively related to NFT [11], failures of informationsharing (eg, confirmation of drug administration) are linkedto treatment errors [81,82], information sharing and inquiryare positively related to overall CPR performance [34],provision of directions for immediate action and inquirycorrelate with cardiovascular support (provision of CC anddefibrillation) [2], and teamwork skills (including communi-cation defined as repetition of messages, usage of accurateterminology, establishment of a conventional speech, andunsolicited provision of information) are found to bepositively linked to CPR performance algorithm adherence[35,37] and clinical care of trauma patients [60].

Communication is the vehicle for information sharing, aspointed out by ALS instructors [50]. We found evidence of

517Effects of team coordination during CPR

inadequate communication regarding different aspects:insufficient communication about individual team memberskills, no or unclear designation of seniority or leadershiproles before CPR treatment, and infrequent post-process teamreflection of CPR performance (eg, no formal debriefing)[83]. At least one third of the trauma team communicationwas neither comprehensible nor audible [84]. Furthermore,the detail rate of verbalizations was found to positivelycorrelate with the accuracy of ongoing tasks: the lack of detailin verbalizations can lead to inaccurate information trans-mission during treatment [85]; sparse communicationbetween interns, nurses, and senior staff members wasperceived by interns after performing simulated CPR, andalso in real-world CPR situations [74]. A lack of clearmessages was reported by nearly two-thirds of ALS courseparticipants after performing a simulated CPR [86].

Recommendations of how to communicate effectively—that is, to coordinate teams—comprise the application ofclosed-loop communication (ie, to express orders explicitlyand to confirm the receipt of orders verbally), avoidance ofinformation overload, and use of terms known and shared byall team members [9,50].

Communication skill improvements were attained afterdifferent types of interventions as reported in several studies:a video-based CRM lecture that leads to self-reportedimprovements of trainees’ communication skills [87]; a 90-minute CRM seminar that increases the quality of TLverbalizations [11]; and a 2-hour didactic session, a half-dayworkshop, and a 28-day trauma course that includessuggestions for effective team communication [46,60,75].Additionally, simulated CPR scenarios followed by debrief-ing sessions are perceived as useful for improving commu-nication by interns and students [47]. A mixed methodsintervention—including role play, video examples illustrat-ing desired behaviors and group discussions—leads tohigher rates of inquiry and information sharing as comparedto a standard training [45].

The literature also documents that communicationbehavior of medical students and nurses during CPRtreatment is predicted by gender: female students makefewer leadership statements and tend to verbalize moreemotional expressions [69]. In addition, personality traitslike extraversion and negative agreeableness [69] andsubjective attitudes towards own CPR skills [88] are reportedto be related to the amount of communication behavior.

4. Discussion

This systematic literature review is focused on articlesreporting the impact of effective team coordination for highquality medical performance during CPR. Aside from differ-ences in the coordination and performance measurementmethods, all selected papers at least support by trend that teamcoordination is essential for managing CPR situations success-fully. Based on our general understanding of the assessed

literature, we deduce the following: (1) Pre- and in-processplanning, i.e. situation assessment and task distribution/roleassignment, can compensate for the lack of team coordinationdepicted in the ALS algorithms. (2) The designation of a TL is apotentialmeans for adequate in-processCPRplanning activities.(3) How far coordination, ideally provided by the TL, leads tothe desired results is affected by the quality of interpersonalcommunication as the vehicle for the transmission of informa-tion between team members.

Our integrative model of team coordination of the CPRaction cycle was created based on a screening and reviewingprocess of pertinent literature. In the model we deduced threeinterrelated team coordination components: planning, lead-ership, and communication. Planning represents a coordina-tion mechanism essential before CPR treatment begins.According to the Coordination Mechanism CircumplexModel [89], the ALS algorithms can be considered asexplicit pre-process planning mechanisms. Such documentedguidelines were developed with the intention to informhealth care team members about the exact sequence of CPRactions to be administered. However, CPR teams mustcompensate for the lack of outlined team coordination advicegiven in these algorithms with situation-specific in-processplanning, that is, team coordination during clinic adminis-tering of CPR treatment, in order for CPR treatment to beeffective. Planning, including pre-process and in-processplanning, can be equated with the assessment orientation andconceptualization phases within the action regulation theory[19] because it implies upcoming action.

The designation of a leading team member is repeatedlyidentified in the literature to be a necessary and efficientmeans to manage the transition from the planning phase intothe execution phase in CPR treatments. The benefits are clear:because inherent time pressure makes a collective decisionabout the course of action ineffective, role allocation and taskassignment are ideally performed by an experienced andsuitably qualified person available—the TL. By the organi-zation of subtasks and the monitoring of the CPR treatmentwithout distraction of constant hands-on administering ofprocess tasks, the TL is able to maintain and convey a sharedplan of procedures. Further responsibilities of the TL are togather relevant information, to provide feedback regardingthe performance of the team members, and to make decisionsabout further treatments. This compilation of responsibilitiesof the TL reveals the complexity of leadership as a necessarycoordination mechanism.

The reported empirical relationship between non-optimalcommunication and treatment failures requires a closerconsideration of linguistic factors and their influence ontask performance. As mentioned before, verbal communi-cation serves as the best medium for coordinationmechanisms. For example, directive leading via assign-ments can be optimized if the receiver verbally acknowl-edges the assignment and a corresponding feedback of theperson that initiated the assignment follows. Such commu-nication that obeys closed loop principles is found to be

Fig. 2 Integrative model of team coordination in CPR action cycle.

518 E. Fernandez Castelao et al.

positively related to performance quality [90]. Each closedloop represents the completion of a subtask, which in turnreduces the level of workload if the corresponding subtaskwas accomplished in a desired manner. Furthermore, thenon–directly involved team members remain updated aboutthe current status of the process and are therefore able toadjust their own behavior to the given circumstances.According to the action regulation theory [19] and ourintegrative model, this function represents the evaluationand adjustment phase.

The recommendation to express short and accurateutterances can be explained on the basis of the results ofexperimental approaches in the investigation of task loadeffects on language processing [91]. It is empiricallyconfirmed that the communication process becomes vulner-able to both time delays and errors when the task loadincreases. In simulated CPR scenarios, the accuracy level ofTL coordination demands is related to better NFT rates [11].Thus, clear and comprehensible communication should befurther focused on as a key coordination mechanism in CPRand emergency room settings.

This review has some limitations. It does not focusprimarily on the quality of methods for measurement ofteamwork. The rules that we apply for the LOE rating arebased on the assessment of the psychometric characteristicsas long as they are included in the article in question. Thus,we purposely dismiss reports of solely psychometric,thematically focused data in order to not lose our currentstudy focus on coordination requirements. Regarding themethodical quality of the modified LOE scale, we suggestconducting a comprehensive validation in order to increasethe validity of this measurement tool. A systematic validitytest would enable a more veridical interpretation of theresults of this review. Nevertheless, our generalized LOEscale can be regarded as having a certain degree of contentvalidity. According to Moosbrugger and Kelava (2012)[101], content validity relies on experts’ logical andfunctional thoughts related to the research question at hand.

Despite the validity limitation, it is eye-catching that onlya small percentage of the articles were comprehensive enoughto be rated LOE 1. This sheds light on the general state-of-the-art of empirical research in the field. Thus, to expand existent

519Effects of team coordination during CPR

knowledge about the impact of coordination behavior onmedical performance, randomized experimental studiesshould be applied more widely, allowing the controlledelimination of confounding variables (eg, medical knowl-edge, status). Beside experimental approaches, it is alsocrucial to obtain more detailed information about theprocesses of actual interaction in the real clinical world inorder to assess ecological validity. We also suggest applyingmethods to study temporal dynamics, for example, effects ofchanges in staff, institutional values and/or functionalbackgrounds on CPR team coordination over time [92].However, it will remain difficult to apply experimentaldesigns and to assess process and outcome of CPR in the fieldbecause of security and other access problems to data. Hence,high fidelity simulators are likely to provide the best availablecontext to organize experimental designs by creating arealistic situation [93–95].

5. Conclusion

In summary, the results of our literature review confirm thatteam coordination—planning, leadership and communication—are well studied and highly relevant factors predicting CPRperformance quality. Our findings also served as the basis forthe development of an integrative model based on the ARTelements of emergency procedure planning—role allocationand task distribution—through effective leadership applied viaclear and comprehensible verbal communication as crucialcoordination requirements in CPR treatments. Furthermore,we list the resources needed to meet those requirements: CPRskills training, communication workshops, simulated CPRscenarios and debriefing sessions. It is now clear that althoughthe ALS Guidelines provide a detailed description of thesequence of medical actions, the accomplishment of situation-specific clinical treatment requires substantial additionalcoordination efforts to optimize the success of applied CPR.With inclusion of the above resources, we expect that the ALSGuidelines and, in turn, success rates of applied, clinical CPRcan be improved in a significant way. The implementation ofcoordination advice into ALS guidelines will help rescuers toround out a more detailed representation of CPR task and teamdemands. This includes well-reputed medical demands—forexample, chest compression, defibrillation, etc—and alsoknowledge about how to coordinate to effectively adhere to theCPR algorithm, which in turn will free more time forcompleting CPR subtasks. The advantage of saving time isespecially beneficial to CPR teamswho, by virtue of tending tomedical emergencies, work under time pressure.

AcknowledgmentsThis research was partly financed by the Courant

Research Centre “Evolution of Social Behavior”, Universityof Göttingen, Germany. Special thanks to Margarita Neff-Heinrich for her English-for-the-sciences proofreading.

References[1] Deakin CD, Morrison LJ, Morley PT, Callaway CW, Kerber RE,

Kronick SL, et al. Part 8: advanced life support. 2010 InternationalConsensus on Cardiopulmonary Resuscitation and EmergencyCardiovascular Care Science with Treatment Recommendations.Resuscitation 2010;81:e93.

[2] Tschan F, Semmer NK, Gautschi D, Hunziker P, Spychiger M,Marsch SU. Leading to recovery: group performance and coordinativeactivities in medical emergency driven groups. Hum Perform2006;19:277-304.

[3] Meerabeau L, Page S. I'm sorry if I panicked you: nurses' accounts ofteamwork in cardiopulmonary resuscitation. J Interprofessional Care1999;13:29-40.

[4] Hunziker S, Laschinger L, Portmann-Schwarz S, Semmer NK,Tschan F, Marsch S. Perceived stress and team performance during asimulated resuscitation. Intens Care Med 2011;37:1473-9.

[5] Rosen MA, Salas E, Wilson KA, King HB, Salisbury M, AugensteinJS, et al. Measuring team performance in simulation-based training:adopting best practices for healthcare. Simul Healthc 2008;3:33-41.

[6] Boos M, Kolbe M, Kappeler P, Ellwart T, editors. Coordination inhuman and primate groups. Berlin: Springer; 2011.

[7] Soar J, Monsieurs KG, Ballance JH, Barelli A, Biarent D, Greif R,et al. European Resuscitation Council Guidelines for Resuscitation2010 Section 9. Principles of education in resuscitation. Resuscitation2010;81:1434-44.

[8] Kohn LT, editor. To err is human: Building a safer health system.7th ed. Washington, DC: National Academy Press; 1999.

[9] Manser T. Teamwork and patient safety in dynamic domains ofhealthcare: a review of the literature. Acta Anaesth Scand 2009;53:143-51.

[10] Tschan F, Semmer NK, Vetterli M, Gurtner A, Hunziker S, MarschSU. Developing observational categories for group process researchbased on task and coordination requirement analysis: examples fromresearch on medical emergency-driven teams. In: Boos M, Kolbe M,Kappeler P, Ellwart T, editors. Coordination in human and primategroups. Berlin: Springer; 2011. p. 93-118.

[11] Fernandez Castelao E, Russo SG, Cremer S, Strack M, Kaminski L,Eich C, et al. Positive impact of crisis resource management training onno-flow time during simulated cardiopulmonary resuscitation: a closerlook at verbal coordination behaviour. Resuscitation 2011;82:1338-43.

[12] Hunziker S, Buhlmann C, Tschan F, Balestra G, Legeret C,Schumacher C, et al. Brief leadership instructions improvecardiopulmonary resuscitation in a high-fidelity simulation: arandomized controlled trial. Crit Care Med 2010;38:1-6.

[13] Neumar RW, Otto CW, Link MS, Kronick SL, Shuster M, CallawayCW, et al. Part 8: adult advanced cardiovascular life support: 2010American Heart Association guidelines for cardiopulmonary resuscita-tion and emergency cardiovascular care. Circulation 2010;122:S729.

[14] Christenson J, Andrusiek D, Everson-Stewart S, Kudenchuk P,Hostler D, Powell J, et al. Chest compression fraction determinessurvival in patients with out-of-hospital ventricular fibrillation.Circulation 2009;120:1241-7.

[15] Semmer NK, Tschan F, Hunziker S, Marsch S. Leadership andminimally invasive training enhance performance in medicalemergency driven teams: simulator studies. In: Duffy VG, editor.Advances in Human Factors and Ergonomics in Healthcare. BocaRaton: Taylor & Francis; 2011. p. 180-90.

[16] Boos M, Kolbe M, Strack M. An inclusive model of group coordination.In: Boos M, Kolbe M, Kappeler P, Ellwart T, editors. Coordination inhuman and primate groups. Berlin: Springer; 2011. p. 11-35.

[17] Riethmüller M, Fernandez Castelao E, Eberhardt I, Timmermann A,Boos M. Adaptive coordination development in anaesthesia teams: alongitudinal study. Ergonomics 2012;55:55-68.

[18] Kolbe M, Burtscher M, Manser T, Künzle B, Grote G. The role ofcoordination in preventing harm in healthcare groups: research examplesfrom anaesthesia and an integrated model of coordination for action

520 E. Fernandez Castelao et al.

teams in health care: coordination in human and primate groups. In:Boos M, Kolbe M, Kappeler P, Ellwart T, editors. Coordination inhuman and primate groups. Berlin: Springer; 2011. p. 75-92.

[19] Action Hacker W, Theory Regulation. A practical tool for the design ofmodern work processes? Eur J Work Organ Psychol 2003;12:105-30.

[20] Oesterreich R, Volpert W. Task analysis for work design on the basisof action regulation theory. Economic and Industrial Democracy1986;7:503-27.

[21] Higgins JPT, Green S, editors. Cochrane handbook for systematicreviews of interventions. Chichester, England. Hoboken, NJ: Wiley-Blackwell; 2008.

[22] Landis JR, Koch GG. The measurement of observer agreement forcategorical data. Biometrics 1977;33:159-74.

[23] Morley PT. Evidence evaluation worksheets: the systematic reviewsfor the evidence evaluation process for the 2010 InternationalConsensus on Resuscitation Science. Resuscitation 2009;80:719-21.

[24] International Liaison Committee On Resuscitation. Defining levels ofevidence. Available from:http://mc.manuscriptcentral.com:80/societyimages/ilcor/Defining%20Levels%20of%20Evidence.doc.[Accessed June 13, 2012].

[25] Working Group GRADE. Grading quality of evidence and strength ofrecommendations. Brit Med J 2004;328:1490.

[26] Atkins D, Eccles M, Flottorp S, Guyatt GH, Henry D, Hill S, et al.Systems for grading the quality of evidence and the strength ofrecommendations I: Critical appraisal of existing approaches. BMCHealth Serv Res 2004;4:38.

[27] Salas E, Sims DE, Burke CS. Is there a "Big Five" in teamwork?Small Group Research 2005;36:555-99.

[28] Rousseau V, Aubé C, Savoie A. Teamwork behaviors. Small GroupResearch 2006;37:540-70.

[29] Frese M, Zapf D. Action as the core of work psychology: a Germanapproach. In: Triandis HC, Dunnette MD, Hough LM, editors.Handbook of Industrial andOrganizational Psychology. Second edition.Palo Alto, California: Consulting Psychologists Press; 1994. p. 271-340.

[30] Hunziker S, Tschan F, Semmer NK, Zobrist R, Spychiger M, BreuerM, et al. Hands-on time during cardiopulmonary resuscitation isaffected by the process of teambuilding: a prospective randomisedsimulator-based trial. BMC Emerg Med 2009;9:3.

[31] Thomas EJ, Williams AL, Reichman EF, Lasky RE, Crandell S,Taggart WR. Team training in the neonatal resuscitationprogram for interns: teamwork and quality of resuscitations.Pediatrics 2010;125:539-46.

[32] Marsch S, Muller C, Marquardt K, Conrad G, Tschan F, HunzikerPR. Human factors affect the quality of cardiopulmonary resuscita-tion in simulated cardiac arrests. Resuscitation 2004;60:51-6.

[33] Marsch S, Hunziker P, Mueller C, Spychiger M, Breuer M, SemmerN, et al. The process of teambuilding delays cardiopulmonaryresuscitation. Crit Care 2004;8:297.

[34] Thomas EJ, Sexton JB, Lasky RE, Helmreich RL, Crandell DS,Tyson J. Teamwork and quality during neonatal care in the deliveryroom. J Perinatol 2006;26:163-9.

[35] Wright MC, Phillips-bute BG, Petrusa ER, Griffin KL, Hobbs GW,Taekman JM. Assessing teamwork in medical education and practice:relating behavioural teamwork ratings and clinical performance. MedTeach 2009;31:30-8.

[36] Høyer CB, Christensen EF, Eika B. Junior physician skill and behav-iour in resuscitation: a simulation study. Resuscitation 2009;80:244-8.

[37] Gaba DM, Howard SK, Flanagan B, Smith BE, Fish KJ, Botney R.Assessment of clinical performance during simulated crises usingboth technical and behavioral ratings. Anesthesiology 1998;89:8-18.

[38] Marsch SCU, Tschan F, Semmer N, Spychiger M, Breuer M,Hunziker PR. Unnecessary interruptions of cardiac massage duringsimulated cardiac arrests. Eur J Anaesthiol 2005;22:831-3.

[39] Henderson MD. Psychological distance and group judgments: theeffect of physical distance on beliefs about common goals. PersonSoc Psychol Bull 2009;35:1330-41.

[40] Driscoll PA, Vincent CA. Organizing an efficient trauma team. Injury1992;23:107-10.

[41] Bernhard M, Becker TK, Nowe T, Mohorovicic M, Sikinger M,Brenner T, et al. Introduction of a treatment algorithm can improvethe early management of emergency patients in the resuscitationroom. Resuscitation 2007;73:362-73.

[42] DeVita MA, Schaefer J, Lutz J, Wang H, Dongilli T. Improvingmedical emergency team (MET) performance using a novelcurriculum and a computerized human patient simulator. Qual SafHealth Care 2005;14:326-31.

[43] Bradley P, Cooper S, Duncan F. A mixed-methods study ofinterprofessional learning of resuscitation skills. Med Educ 2009;43:912-22.

[44] Lo TYM, Morrison R, Atkins K, Reynolds F. Effective performanceof a new post-operative cardiac resuscitation simulation trainingscheme in the paediatric intensive care unit. Intensive Care Med2009;35:725-9.

[45] Thomas EJ, Taggart B, Crandell S, Lasky RE, Williams AL, Love LJ,et al. Teaching teamwork during the neonatal resuscitation program:A randomized trial. J Perinatol 2007;27:409-14.

[46] Holcomb JB, Dumire RD, Crommett JW, Stamateris CE, Fagert MA,Cleveland JA, et al. Evaluation of trauma team performance using anadvanced human patient simulator for resuscitation training. JTrauma 2002;52:1078-86.

[47] Flanagan B, Nestel D, Joseph M. Making patient safety the focus:Crisis resource management in the undergraduate curriculum. MedEduc 2004;38:56-66.

[48] Nishiyama C, Iwami T, Kawamura T, AndoM, Yonemoto N, HiraideA, et al. Quality of chest compressions during continuous CPR;comparison between chest compression-only CPR and conventionalCPR. Resuscitation 2010;81:1152-5.

[49] Huseyin TS, Matthews AJ, Wills P, O'Neill VM. Improving theeffectiveness of continuous closed chest compressions: an explor-atory study. Resuscitation 2002;54:57-62.

[50] Andersen PO, Jensen MK, Lippert A, Ostergaard D. Identifying non-technical skills and barriers for improvement of teamwork in cardiacarrest teams. Resuscitation 2010;81:695-702.

[51] Cole E, Crichton N. The culture of a trauma team in relation to humanfactors. J Clin Nurs 2006;15:1257-66.

[52] Hunziker S, Johansson AC, Tschan F, Semmer NK, Rock L, HowellMD, et al. Teamwork and Leadership in Cardiopulmonary Resusci-tation. J Am Coll Cardiol 2011;57:2381-8.

[53] Künzle B, Kolbe M, Grote G. Ensuring patient safety througheffective leadership behaviour: A literature review. Safety Sci 2010;48:1-17.

[54] Klein KJ, Ziegert JC, Knight AP, Xiao Y. Dynamic delegation:Shared, hierarchical, and deindividualized leadership in extremeaction teams. Admin Sci Quart 2006;51:590-621.

[55] St. Pierre M, Hofinger G, Buerschaper C. Crisis Management inAcute Care Settings: Human Factors and Team Psychology in a HighStakes Enviroment. 1st ed. Berlin, Heidelberg: Springer; 2008.

[56] Arshid M, Lo TM, Reynolds F. Quality of cardio-pulmonaryresuscitation (CPR) during paediatric resuscitation training:Time to stop the blind leading the blind. Resuscitation 2009;80:558-60.

[57] Mäkinen M, Aune S, Niemi-Murola L, Herlitz J, Varpula T, Nurmi J,et al. Assessment of CPR-D skills of nurses in Goteborg, Sweden andEspoo, Finland: Teaching leadership makes a difference. Resuscita-tion 2007;72:264-9.

[58] Cooper S, Wakelam A. Leadership of resuscitation teams: ‘Light-house Leadership’. Resuscitation 1999;42:27-45.

[59] Mann CJ, Heyworth J. Comparison of cardiopulmonary resuscitationtechniques using video camera recordings. J Accid Emerg Med1996;13:198-9.

[60] Capella J, Smith S, Philp A, Putnam T, Gilbert C, Fry W, et al.Teamwork training improves the clinical care of trauma patients. JSurg Educ 2010;67:439-43.

521Effects of team coordination during CPR

[61] Lubbert PHW, Kaasschieter EG, Hoorntje LE, Leenen LPH. Videoregistration of trauma team performance in the emergency depart-ment: the results of a 2-year analysis in a level 1 trauma center. JTrauma 2009;67:1412-20.

[62] Hunt EA, Walker AR, Shaffner DH, Miller MR, Pronovost PJ.Simulation of in-hospital pediatric medical emergencies andcardiopulmonary arrests: highlighting the importance of the first 5minutes. Pediatrics 2008;121:e34-43.

[63] Yun S, Faraj S, Sims HPJR. Contingent leadership and effectivenessof trauma resuscitation teams. J Appl Psychol 2005;90:1288-96.

[64] Hynes P, Kissoon N, Hamielec CM, Greene AM, Simone C. Dealingwith aggressive behavior within the health care team: a leadershipchallenge. J Crit Care 2006;21:224-8.

[65] Xiao Y, group tL. Understanding coordination in a dynamic medicalenvironment: methods and results. In: McNeese M, Salas E, EndsleyM, editors. New trends in cooperative activities: Understandingsystem dynamics in complex environments. Santa Monica, CA:Human Factors and Ergonomics Society; 2001. p. 242-58.

[66] Hoff WS, Reilly PM, Rotondo MF, DiGiacomo JC, Schwab CW. Theimportance of the command-physician in trauma resuscitation. JTrauma 1997;43:772-7.

[67] Adams BD, Zeiler K, Jackson WO, Hughes B. Emergency medicineresidents effectively direct inhospital cardiac arrest teams. Am JEmerg Med 2005;23:304-10.

[68] KhetarpalS,SteinbrunnBS,McGonigalMD,StaffordR,NeyAL,KalbDC,etal.Traumafacultyandtraumateamactivation:Impactontraumasystemfunctionandpatientoutcome.JTrauma1999;47:576-81.

[69] Streiff S, Tschan F, Hunziker S, Buehlmann C, Semmer NK,Hunziker P, et al. Leadership in medical emergencies depends ongender and personality. Simul Healthc 2011;6:78-83.

[70] Marsch SCU, Tschan F, Semmer N, Spychiger M, Breuer M,Hunziker PR. Performance of first responders in simulated cardiacarrests. Crit Care Med 2005;33:963-7.

[71] Sugrue M, Seger M, Kerridge R, Sloane D, Deane S. A prospectivestudy of the performance of the trauma team leader. J Trauma1995;38:79-82.

[72] Ritchie PD, Cameron PA. An evaluation of trauma team leaderperformance by video recording. Aust N Z J Surg 1999;69:183-6.

[73] Hayes CW, Rhee A, Detsky ME, Leblanc VR, Wax RS. Residentsfeel unprepared and unsupervised as leaders of cardiac arrest teams inteaching hospitals: A survey of internal medicine residents. Crit CareMed 2007;35:1668-72.

[74] O'Brien G, Haughton A, Flanagan B. Interns' perceptions ofperformance and confidence in participating in and managingsimulated and real cardiac arrest situations. Med Teach 2001;23:389-95.

[75] Gilfoyle E, Gottesman R, Razack S. Development of a leadershipskills workshop in paediatric advanced resuscitation. Med Teach2007;29:e276-83.

[76] Cooper S. Developing leaders for advanced life support: evaluation ofa training programme. Resuscitation 2001;49:33-8.

[77] Niemi-Murola L, Mäkinen M, Castren M. Medical and nursingstudents' attitudes toward cardiopulmonary resuscitation and currentpractice guidelines. Resuscitation 2007;72:257-63.

[78] Weinstock P, Halamek LP. Teamwork during resuscitation. PediatrClin North Am 2008;55:1011-24.

[79] Boos M, Mey D. Führung, Team und Kommunikation. In: Welk I,Bauer M, editors. OP-Management—Von der Theorie zur Praxis. 1sted. Heidelberg, Berlin: Springer; 2011. p. 141-52.

[80] Riggio RE. Introduction to industrial/organizational psychology. . 5thed.Upper Saddle River, NJ: Prentice Hall; 2008.

[81] Hunt EA, Hohenhaus SM, Luo X, Frush KS. Simulation of pediatrictrauma stabilization in 35 North Carolina emergency departments:Identification of targets for performance improvement. Pediatrics2006;117:641-8.

[82] Kozer E, Seto W, Verjee Z, Parshuram C, Khattak S, Koren G, et al.Prospective observational study on the incidence of medication errorsduring simulated resuscitation in a paediatric emergency department.Brit Med J 2004;329:1321-4.

[83] Pittman J, Turner B, Gabbott DA. Communication betweenmembers of the cardiac arrest team—a postal survey. Resusci-tation 2001;49:175-7.

[84] Bergs EAG, Rutten FLPA, Tadros T, Krijnen P, Schipper IB.Communication during trauma resuscitation: do we know what ishappening? Injury 2005;36:905-11.

[85] Bogenstätter Y, Tschan F, Semmer NK, Spychiger M, Breuer M,Marsch S. How accurate is information transmitted to medicalprofessionals joining a medical emergency? A simulator study. HumFactors 2009;51:115-25.

[86] Østergaard HT, Østergaard D, Lippert A. Implementation of teamtraining in medical education in Denmark. Qual Saf Health Care2004;13:i91-5.

[87] Blum RH, Raemer DB, Carroll JS, Dufresne RL, Cooper JB. A methodfor measuring the effectiveness of simulation-based team training forimproving communication skills. Anesth Analg 2005;100:1375-80.

[88] Dwyer T, Williams LM. Nurses' behaviour regarding CPR and thetheories of reasoned action and planned behaviour. Resuscitation2002;52:85-90.

[89] Strack M, Kolbe M, Boos M. Dimensions of group coordination:applicability test of the coordination mechanism circumplex model.In: Boos M, Kolbe M, Kappeler P, Ellwart T, editors. Coordination inhuman and primate groups. Berlin: Springer; 2011. p. 57-73.

[90] Tschan F. Communication enhances small group performance if itconforms to task requirements: The concept of ideal communicationcycles. Basic Applied Soc Psychol 1995;17:371-93.

[91] Hohlfeld A, Fukuda R, Neuper S, Sangals J, Sommer W, Sträter O.Task load effects on language processing: experimental approaches.In: Dietrich R, Childress TM, editors. Group Interaction in High RiskEnvironments. Hampshire: Ashgate; 2004. p. 207-40.

[92] Roe RA. Time in applied psychology. Eur Psychol 2008;13:37-52.[93] Müller MP, Hänsel M, Fichtner A, Hardt F, Weber S, Kirschbaum C,

et al. Excellence in performance and stress reduction during twodifferent full scale simulator training courses: a pilot study.Resuscitation 2009;80:919-24.

[94] DeVitaM. Organizational factors affect human resuscitation: the role ofsimulation in resuscitation research. Crit Care Med 2005;33:1150-1.

[95] Hunziker S, Tschan F, Semmer NK, Howell MD, Marsch S. Humanfactors in resuscitation: lessons learned from simulator studies. JEmerg Trauma Shock 2010;3:389-94.

[96] Husebø SE, Rystedt H, Friberg F. Educating for teamwork—nursingstudents’ coordination in simulated cardiac arrest situations. J AdvNurs 2011;67:2239-55.

[97] Oakley E, Stocker S, Staubli G, Young S. Using video recording toidentify management errors in pediatric trauma resuscitation.Pediatrics 2006;117:658-64.

[98] Quintana Riera S, Sánchez González B, Trenado Álvarez J. del MarFernández Fernández M, Mestre Saura J. The physiological effect onrescuers of doing 2 min of uninterrupted chest compressions.Resuscitation 2007;74:108-12.

[99] Tschan F, Vetterli M, Semmer NK, Hunziker S, Marsch SCU.Activities during interruptions in cardiopulmonary resuscitation: asimulator study. Resuscitation 2011;82:1419-23.

[100] Xiao Y, Seagull FJ, Mackenzie CF, Klein K. Adaptiveleadership in trauma resuscitation teams: a grounded theoryapproach to video analysis. Cognit Tech Work 2004;6:158-64.

[101] Moosbrugger H, Kelava A. Qualitätsanforderungen an einenpsychologischen Test (Testgütekriterien). In: Moosbrugger H,Kelava A, editors. Testtheorie und Fragebogenkonstruktion.2nd ed. Berlin, Heidelberg, New York: Springer; 2012.p. 7-25.