Article Text

Download PDFPDF

Anticipation, teamwork and cognitive load: chasing efficiency during robot-assisted surgery
  1. Kevin Sexton1,
  2. Amanda Johnson1,
  3. Amanda Gotsch1,
  4. Ahmed A Hussein1,2,
  5. Lora Cavuoto3,
  6. Khurshid A Guru1
  1. 1 Department of Urology, ATLAS Program, Roswell Park Cancer Institute, Buffalo, New York, USA
  2. 2 Urology, Cairo University, Cairo, Egypt
  3. 3 Industrial and Systems Engineering, University at Buffalo - The State University of New York, Buffalo, New York, USA
  1. Correspondence to Dr Khurshid A Guru, Department of Urology, ATLAS Program, Roswell Park Cancer Institute, Buffalo, New York, USA; Khurshid.Guru{at}


Introduction Robot-assisted surgery (RAS) has changed the traditional operating room (OR), occupying more space with equipment and isolating console surgeons away from the patients and their team. We aimed to evaluate how anticipation of surgical steps and familiarity between team members impacted efficiency.

Methods We analysed recordings (video and audio) of 12 robot-assisted radical prostatectomies. Any requests between surgeon and the team members were documented and classified by personnel, equipment type, mode of communication, level of inconvenience in fulfilling the request and anticipation. Surgical team members completed questionnaires assessing team familiarity and cognitive load (National Aeronautics and Space Administration – Task Load Index). Predictors of team efficiency were assessed using Pearson correlation and stepwise linear regression.

Results 1330 requests were documented, of which 413 (31%) were anticipated. Anticipation correlated negatively with operative time, resulting in overall 8% reduction of OR time. Team familiarity negatively correlated with inconveniences. Anticipation ratio, per cent of requests that were non-verbal and total request duration were significantly correlated with the console surgeons’ cognitive load (r=0.77, p=0.006; r=0.63, p=0.04; and r=0.70, p=0.02, respectively).

Conclusions Anticipation and active engagement by the surgical team resulted in shorter operative time, and higher familiarity scores were associated with fewer inconveniences. Less anticipation and non-verbal requests were also associated with lower cognitive load for the console surgeon. Training efforts to increase anticipation and team familiarity can improve team efficiency during RAS.

  • Audit and feedback
  • Communication
  • Human factors
  • Performance measures
  • Teamwork
View Full Text

Statistics from


Communication, team coordination and other non-technical skills are crucial for patient safety in a high-stress environment like the operating room (OR).1 In 2014, The Joint Commission on Accreditation of Healthcare Organizations identified human factors and communication errors as the leading root causes for operative and postoperative sentinel events.2 New technology, including robot-assisted surgery (RAS), is constantly being introduced in the OR to refine surgical technique, and therefore improve patient outcomes. RAS has been associated with reduced blood loss, transfusions, quicker recovery and enhanced convalescence.3 However, introduction of relatively new technology into the surgical armamentarium may be associated with newer forms of errors or near-miss events.4 Additionally, compared with the traditional open approach, RAS may be more demanding in terms of OR staff technical and non-technical skills.5 The OR layout is modified to accommodate the robot and ancillary equipment obstructing views and physically isolating members of the team — most notably the surgeon, who is no longer located beside the patient and the bedside assistants.6 This arrangement of the team may inhibit the interpersonal cues and microcommunications that a traditional arrangement allows, which may increase the potential for minor incidents and susceptibility to errors that have been associated with sentinel events.7

Another notable factor related to surgical performance and team effectiveness is cognitive workload. Individuals have a finite amount of mental resources that can be devoted to a task at any given time.8 9 Higher complexity of a given task correspondingly increases the amount of mental resources used. When complexity and dedicated mental resources are high, there are fewer resources available to address unexpected events and the error rate can increase.10 Procedure duration and surgical errors have been found to be directly related to cognitive workload in surgery.11 12

In the RAS setting, where the technical demands and the surgical environment can be obstacles to effective communication, it may be advantageous to address the effectiveness of the team’s communication and coordination strategies for this purpose. Previous research has suggested that almost half of adverse events during surgery are related to provider error and are avoidable.13 In this context, the Department of Defense developed teamwork training programme — evaluated across 68 facilities nationwide — that addressed efficiency and performance in healthcare through team training and emphasis on a wilfully coordinated approach to patient care. The results of this study and others have demonstrated that team training improved communication and created a safer environment for patients.14–16 Additional studies have shown improved patient outcomes with increased team familiarity.17

Anticipation has been previously established as behavioural indicator of the quality of teamwork.18 19 Human factors research has indicated that highly effective teams anticipate each other’s actions and needs, which allows for more efficient coordination.19 These skills allow the team to adjust their communication strategies in response to high-pressure situations.18 19 While other studies have looked at team dynamics in the surgical setting and demonstrated a connection to patient outcomes, this has not been undertaken during RAS.

In this study, we aimed to investigate the impact of anticipation as a measure of efficiency in the setting of RAS. We hypothesise that higher team anticipation would be associated with improved OR efficiency.


The Applied Technology Laboratory for Advanced Surgery programme at Roswell Park Cancer Institute initiated the ‘Techno-Fields’ Project in 2013 (I 244113) to analyse and optimise team interactions in the OR environment. Data were collected from (1) three video cameras positioned for optimal coverage of the OR, (2) the video feed from the surgical console and (3) up to eight individual microphones assigned to members of the team. The three video cameras were positioned to record (1) the OR table recording interaction between the physician’s assistant (PA) on the right and the scrub nurse on the left of the patient, (2) the console surgeon and the nurses station, and (3) the left bedside assistant, anaesthesia station and the OR door. Recording began once the patient’s face was covered and time out ended (to ensure anonymity of patients) and stopped after the undocking of the robot. Recordings were stopped temporarily if OR staff who had not yet consented were in the OR. The video and audio data from each surgery were synchronised using Adobe Premier Pro CS6. Custom-programmed, multivideo software was used to play the synchronised files simultaneously and permit isolation of individual microphone channels. This permitted observational analysis of the actions and interactions of the team within the surgical environment.5 Data were collected by two trained medical students supervised by a urology fellow and a human factors engineering expert. Our methodology has been previously detailed.5 All patients and team members who were recorded gave informed consent to participate in the study.

OR efficiency was defined in terms of anticipation index, proportion of surgery duration for requests and inconveniences (table 1). All requests between the console surgeon and other team members during each surgery were documented. Requests were classified based on personnel involved, equipment type and mode of communication (table 2). Detailed time analysis of each request was preformed, including start, end, duration and anticipation. In human factors analysis, the anticipation ratio describes the ratio of exchanges of information (without verbal request) to exchanges where information was requested.20 We adapted the term for our study purpose to describe the ratio of anticipated requests (where execution occurred without or before enquiry) to total requests (whether enquired by the console surgeon or not). The cognitive load was quantified using the National Aeronautics and Space Administration – Task Load Index (NASA-TLX) administered at the end of each procedure.21 Multiple studies in the surgical domain have established NASA-TLX as a sensitive and effective tool for evaluating cognitive workload in the OR.21–23 This self-assessment incorporates six variables: mental demand, physical demand, temporal demand, perceived performance, effort and frustration. NASA-TLX scores from individuals were combined to calculate composite scores for the entire team. The inconvenience index was defined to reflect disruptions in team efficiency within the context of catalogued requests. It is a sum total of the catalogued events when one of the following criteria was met: communication breakdowns, multiple requests were made for an item, or an item was not readily available and had to be retrieved. In addition to NASA-TLX, participants completed questionnaires about their demographics and team familiarity at the end of each procedure. Composite familiarity scores were calculated for the entire team and for the ‘key triad’ consisting of the console surgeon, bedside assistant and scrub nurse (online supplementary appendix 1).5 24

Supplementary Material

Supplementary Appendix 1
Table 1

Definitions of operating room efficiency measures: anticipation index, proportion of requests’ time and inconveniences

Table 2

Classification and frequency of requests

Descriptive statistics consisted of relative frequencies for the categorical variables and means and SD for the continuous variables. Inter-rater reliability (IRR) between the two raters was assessed using intraclass correlation for the measured durations of each action during one surgery. The relationships between predictors of team efficiency were assessed using Pearson correlation. Forward stepwise linear regression was then used to identify which factors were significant predictors of the clinical outcome of surgical duration. All correlations were two-sided, with statistical significance defined as p value ≤0.05. The entry criterion for the stepwise regression was p=0.05 and the exit criterion was p=0.15. Data were analysed using the Statistical Package for the Social Sciences (SPSS V.24).


Our study included data from 12 consecutive robot-assisted radical prostatectomies enrolled in Techno-Fields between November 2014 and April 2015 (table 3). Participants across the 12 surgeries included three lead surgeons, three assistant surgeons, three PAs, seven scrub nurses and eleven circulating nurses. Acceptable agreement has been achieved between both raters (IRR 0.74, p<0.001). A total of 1335 requests were documented during 36 hours (2147 min) of console time across the 12 procedures. The mean console time was 179 min (range 111–330) and the mean requests’ duration was 36 min (20% of overall operative time). On average, 111 requests were documented per procedure, of which 37 were anticipated and 18 were associated with inconvenience. Procedures 5–7 were performed by different surgeons, and procedure 6 was significantly longer because of an adverse event (accidental clipping of the obturator nerve), which consequently led to more requests and discussions among the team. For this reason, console time was controlled for in the correlation analysis and the anticipation ratio considers the anticipated requests relative to total requests for that procedure (table 4). Anticipated requests averaged 5 s to complete, while non-anticipated requests took 26 s. This was translated into an overall 7% reduction of OR time.

Table 3

Preoperative, operative and postoperative characteristics of 12 robot-assisted radical prostatectomies included in the study

Table 4

Analysis of requests (duration, anticipation and workload)

Several significant correlations were observed when comparing measures of team efficiency with familiarity assessments completed by the team members and cognitive load assessments. There was a significant positive correlation between anticipation ratio and surgeon NASA-TLX (r=0.74, p=0.01), overall team NASA-TLX (r=0.66, p=0.03) and non-verbal requests (r=0.74, p≤0.001). There was a significant positive correlation between requests’ duration and surgeon NASA-TLX (r=0.70, p=0.02). Team familiarity was inversely correlated with inconveniences (r=−0.67, p=0.02) (table 5).

Table 5

Partial correlations for National Aeronautics and Space Administration – Task Load Index (NASA-TLX) and familiarity scores when controlling for console time (p value is provided in parentheses, bold values indicate a significant correlation)

On stepwise linear regression analysis, larger number of requests was significantly associated with longer operative time (r=0.79, p<0.001). Higher anticipation ratio significantly correlated with shorter operative time (r=−0.44, p=0.01) (figure 1) (table 6).

Figure 1

Regression model showing association between console time and anticipation ratio.

Table 6

Stepwise linear regression analysis looking at predictors of team efficiency


Teamwork, communication and familiarity have been associated with better team performance and improved clinical outcomes in surgery.15 17 Inversely, deficiencies in these areas have been implicated in surgical error, increased surgery times and less favourable clinical outcomes.25 There has been growing body of literature looking at leadership and communication in open and laparoscopic surgery.1 26 27 However, there is paucity of literature that investigated the impact of these factors on teamwork and clinical outcomes in the setting of RAS. Prior studies were limited by the utilisation of questionnaires that may be subject to recall and subjective biases, or simulated surgical scenarios rather than the real operative environment. While simulations can provide an insight about the possible consequences of such factors, studying the real operative environment would offer more accurate depictions of the workload and challenges that face the OR staff during RAS.28 Our methodology allowed for detailed analysis of anticipation and inconvenience events in a real RAS setting and assessed the impact on team performance.

The results of the current study support the impact of these factors on measures of team efficiency. Anticipation has been found to be related to team effectiveness and the ability for a team to respond to high-stress situations.18–20 This study demonstrated that anticipation in RAS is both measurable and quantifiable with the use of the anticipation ratio. It also establishes the importance of anticipation in the RAS setting where more non-verbal communication may be necessary due to the unique environment of the RAS OR. Anticipated requests were five times shorter than non-anticipated ones. The significant difference in average time to complete anticipated versus non-anticipated requests and the corresponding reduction in console time illustrates the potential importance of this finding.

Another example was the influence of team familiarity on inconveniences, where higher familiarity scores were associated with fewer inconvenience events. Inconvenience events lead to prolongation of surgery times, which is associated with higher costs and less favourable clinical outcomes.5 25 29 30 This goes in hand with prior studies that demonstrated the benefit of both surgical experience and team familiarity.17 This establishes the potential importance of considering these factors when the aim is to improve surgical efficiency and patient safety, and also opens new opportunities and creates new challenges for quality improvement strategies for RAS. While it seems beneficial to increase the frequency that a surgical team, whose members are familiar with each other, work together, this may not be feasible as both scheduling and staff retention in the OR can be a challenge.31 Another implication is to consider team familiarity during scheduling procedures to cross-train new staff without jeopardising outcomes, and to maintain a versatile workforce in hospitals staff covering across different subspecialties.

Higher task load of the console surgeon was associated with both increased time spent by other team members fulfilling requests, whereas higher task load of the whole team (driven by increased load on the assistant surgeon) was associated with higher anticipation ratios. More complicated surgeries require more focus, and in turn are likely related to an increased cognitive workload on the surgeon and team. This increased workload may require more non-verbal communication and anticipation from the team. Complicated surgeries are often unavoidable, and this finding highlights the importance of efficient communication and teamwork during surgery. Improvements in teamwork and communication may potentially help to reduce the increased surgical times associated with complex procedures and improve patient safety. High cognitive and physical demands have been linked to occurrence of errors, inadvertent injuries and adverse patient outcomes.32 33 Consequences that relate to staff stress, safety and career satisfaction have been also reported.34

Better understanding of the surgical team interactions and behaviour during RAS can lead to substantial improvement in team ‘culture’, help minimise errors and improve efficiency.35 Interventions targeting team interactions were associated with reduction in errors, job stress, sick leaves, better job satisfaction and patient outcomes.36 37 The findings of this study highlight the importance of team familiarity and that it should be taken into account during staff scheduling, especially with newer team members. Our group is currently developing a training curriculum to ensure comprehensive non-technical skills training for RAS staff.

This study has several limitations. First, the nature of observational studies may cause some participants to change their behaviours. However, we believe this effect to be small as video data have been collected in this OR for an extended period of time and the novelty has diminished. As the data were collected via video, there were technical challenges in addition to issues related to ‘blind spots’ in the videos. These were minimised by having multiple cameras that were strategically positioned. The nature of video and audio data required time-consuming data analyses but provided very rich data. Participants were asked to complete the questionnaires at the end of the procedure and therefore may be subject to recall bias. However, this was inevitable to minimise interference with patient care. Also, we used a composite TLX and familiarity scores, which may not have been previously validated.


This study used an effective means to measure anticipation in the OR. Anticipation and team familiarity correlated with requests and surgical times, as well as workload. It highlights the importance of non-technical factors in improving OR efficiency and ultimately patient safety.


View Abstract


  • Contributors None.

  • Funding This research was supported in part by funding from the National Cancer Institute of the National Institutes of Health under award number: R25CA181003, and Roswell Park Alliance Foundation.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval Roswell Park Cancer Institute IRB.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles