Trust, Autonomy, and Teaching Dynamics in Robotic Surgery: A Mixed Methods Study

doi:10.21203/rs.3.rs-5154852/v1

Download PDF

Research Article

Trust, Autonomy, and Teaching Dynamics in Robotic Surgery: A Mixed Methods Study

https://doi.org/10.21203/rs.3.rs-5154852/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Purpose

Compared to open or laparoscopic surgery, the configurational relationship between trainer and trainee in robotic surgery reduces face-to-face interaction and ability to directly co-manipulate the surgical field. To elucidate the impact of this environment on surgical training dynamics, we conducted a mixed-methods study examining dyadic trust, autonomy, teaching, communication, and feedback in robotic surgery.

Methods

Robotic procedures performed on a dual console system at a single academic center were included. Trainee console operative time, representing autonomy, was extracted from the robotic system. Bidirectional trust was measured via a modified Leader Member Exchange (mLMX) questionnaire at the end of each case. Correlation between autonomy and trust was analyzed utilizing Pearson’s coefficient. Procedures were video- and audio-recorded, capturing both endoscopic view and a broad view of the operating consoles with trainer/trainee interactions. Two researchers performed thematic analysis of the synchronized video and transcribed audio.

Results

Nine robotic procedures in colorectal, general surgery, urology, and thoracic surgery were included. Trainee console time was directly correlated with average trainer mLMX trust scores (r=0.54) but not trainee scores (r=-0.19). Average trainer mLMX trust score was 4.08 vs. 3.16 (p=0.038) for trainees with >50% vs. <50% console time; trainee mLMX was not significantly different between the same groups (p=0.74). Thematic analysis revealed major themes of effective teaching techniques, feedback “escalation,” and trust-fostering behavioral strategies.

Conclusion

Higher trainer trust is associated with higher resident autonomy, and trainers/trainees can use specific strategies for teaching, feedback, and fostering trust in the robotic operating room. These findings may improve communication and trainee autonomy in the robotic operating room, and inform future robotic surgical educational curricula.

The traditional educational paradigm in the operating room relies on bidirectional communication involving both verbal and non-verbal cues. In open and laparoscopic surgery, surgeons and trainees can directly observe each other’s behavior and co-manipulate the field during physical task performance. As robotic surgery gains prominence in the operating room, trainees must also be provided opportunities for exposure to this novel technology.^1,2 The shift to robotic surgery poses fresh challenges for surgical educators: the isolated configuration of separate consoles at a distance from the patient leads to an absence of non-verbal cues such as body language, stance, and facial expressions. Additionally, the lack of immediate physical control and haptic feedback further complicate the training environment, requiring explicit verbalization of otherwise tacit knowledge.^3–7

In addition to the pedagogical challenges, robotic surgery presents another psychological challenge for surgical teachers. Control of the robotic console is typically absolute with either the trainer or the trainee having complete control at any given time, making trust a requisite for trainee autonomy. A current absence of systematic training for robotic surgeons may stem from a deeper gap in the understanding of how trainers and trainees interact and communicate in this environment.^8,9 Consequently, this gap could hinder trainee development by decreasing autonomy.⁸

Some studies have begun exploring trainee autonomy in robotic surgery. Recent work by Wang et al. investigated the relationship between console time and various factors such as resident level, surgeon seniority, and case volume, finding that senior residents were given a higher percentage of console time than junior residents, but attending surgeon robotic case volume did not influence how much console time they gave to trainees.⁶ Qualitative interviews with attending surgeons found that they used training level as a surrogate for resident entrustability, and balance this against other concerns such as time pressures and efficiency. However, none of these studies measure the mentor-mentee relationship and bidirectional trust or examine how communication, trust, and autonomy interact.

Understanding the factors that impact trainer / trainee trust and trainee autonomy is essential for mindful robotic surgery instructional design. In this mixed methods study, we first quantitatively examined the relationship between trust and autonomy in the robotic surgery operating room.¹⁰ In contrast to other entrustment related objective measures, we chose to utilize a conceptual framework for dyadic trust from organizational psychology, the Leader-Member Exchange (LMX). LMX captures the bidirectional trust dynamic between trainer and trainee rather than solely one-directional trust from the trainer's perspective. We then qualitatively explored teaching, communication, and feedback factors underscoring trust and trainee autonomy in robotic surgery, aiming to extract insights that could improve teaching and learning in robotic surgery.

Case Selection and Recruitment

This study was approved under Expedited Review by The Ohio State University Institutional Review Board (study ID: 2022H0006). The study was conducted at The Ohio State University Medical Center (OSUMC), a tertiary academic medical center. Robotic procedures were performed using the Da Vinci Xi® dual console system (Intuitive Surgical, Sunnyvale, CA). To optimize specialty representation and generalizability, eligible cases included core robotic surgical procedures from all surgical specialties utilizing robotic surgery (general, gynecology/gynecology oncology, urology, thoracic, and colorectal surgery), with a focus on noncomplex procedures considered essential for trainees. For representation and generalizability reasons, cases from any specialty were eligible. (Table 1) Cases with a single trainer and a single trainee were included.

Once an eligible case was identified on the upcoming robotic surgery schedule, the attending surgeon / trainer and resident surgeon / trainee who were scheduled to participate in these cases were emailed ahead of time. The recruitment email contained a detailed description of the project and a link to an online informed consent form. Trainees and trainers were contacted separately to prevent undue influence or pressure to participate.

Intraoperative recordings

Intraoperative recordings were obtained by a member of the research team, a research assistant (RA), and a videographer. A video camera was positioned at least eight feet behind the robotic consoles to exclusively capture the interactions between the trainer and trainee. Microphones were attached to both the trainer’s and trainee’s surgical masks to ensure clear audio capture. In compliance with IRB stipulations, no patient identifiers were included in the recordings. Audio and video recordings commenced after operative timeout and the video frame excluded the patient on the operative table. At the end of the procedure, the intrabdominal endoscopic view was extracted from the Da Vinci system. Care was taken to minimize obtrusiveness of the research proceedings and data gathering process, in order not to influence the procedure, the operating room dynamics, and patient outcomes.

Surveys

After completion of the case, a link containing a survey was sent to each trainer and trainee. The surveys collected demographic information, training related information, and the modified leader member exchange questionnaire (mLMX). (Appendix 1)

The mLMX is a 9-item questionnaire that measures the quality of relationship and trust between a leader and an individual team member. A high-quality relationship as measured by LMX has been demonstrated to enhance positive outcomes, including task performance, citizenship behaviors, job satisfaction, and commitment, while reducing negative outcomes, such as counterproductive behaviors and turnover intentions.^11–14 As our study sought to investigate communication and behaviors from both trainers and trainees, LMX was a more appropriate instrument than other commonly utilized unidirectional surgical entrustability measures.^10,15,16 We modified the existing LMX questionnaire to fit the robotic surgery training context.

Media Post-processing

Audio and video recordings were securely stored in encrypted online file repositories provided by OSUMC. Videos were verified again for absence of any identifiers. The audio and video streams from the console and endoscopic cameras were then synchronized and combined into an integrated feed. (Fig. 1) This allowed for coordination between console activities, verbal interactions, and the events occurring in the operative field. Audio was transcribed using a transcription service (Rev.com, Inc., Austin TX). Transcriptions were reviewed, deidentified, and edited for accuracy.

Quantitative Analysis

The duration and timestamps of user-controlled activity were extracted from the Da Vinci surgical system. These were then matched to either trainer or trainee and converted into percentage of total console control per case. This is defined as console time, which is the proportion of time which a user acted as the primary surgeon and controlled the operative arms from the robotic console.¹⁷ Console time has been previously established as a reasonable surrogate for trainee operative autonomy.¹⁷

The mLMX survey results were compiled and analyzed. The nine items on each questionnaire were also averaged to create a composite score per participant. Pearson correlation measures were used to determine strength of the linear relationship between console time and mLMX scores. T- test scores were utilized when comparing mLMX means between two groups. All statistical analyses were performed using Prism GraphPad, version 10.2.3 (GraphPad Software, La Jolla CA).

Qualitative Analysis

Two members of the research team, a PGY3 general surgery resident (RJA) and surgeon educator (EH), conducted qualitative analysis utilizing a thematic analysis approach.^18,19 The combined stream, which included both console and endoscopic views, was reviewed alongside the live transcripts of the conversations. Initial codes were generated based on our specific interest in teaching, feedback, communication, trust, and autonomy, and applied independently with this epistemological lens. Analysis was conducted iteratively and inductively; team members met regularly to discuss meanings and interpretations of the data, and to collaboratively collate, review, define, and name themes.

Demographics and Training

A total of nine robotic surgery procedures between May and September 2023 were included. Of the 9 procedures, 4 were Colorectal, 3 General Surgery, 1 Thoracic Surgery, and 1 Urology. (Table 1) 9 trainers and 9 trainees participated and completed surveys. Demographic information is summarized in Table 2. The 9 trainees had an average age of 31.7 years. 44.4% of the trainees identified as men and 55.6% identified as women. 55.6% trainees identified as white, 22.2% as Asian and 22.2% as others. 33.3% of trainees were mid-level residents (PGY3, 4), 33.3% were chief residents (PGY5), and 33.3% were first year fellows. The 9 trainers had an average age of 41 years. 55.6% identified as men and 44.4% identified as women. 77.8% identified as white and 22.2% identified as Asian. The range of combined training and practice was between 11- 25 years.

Quantitative Analysis

The average overall trainee console time was 47.3% ± 16.2, with a range between 31.4% and 74.3%, and notable differences between mid-level residents 49.1%±14.8, chiefs 38.5%± 21.1, and fellows 54.2%± 13.8. The average overall trainer console time was 52.7% ± 16.2 with a range between 25.7% and 68.6%. The average trainee mLMX trust score was 3.83±0.66 with a range between 2.88 and 4.77. The average trainer mLMX trust score was 3.67±0.71 with a range between 2.77 and 4.77.

Correlation analysis between the average trainer mLMX trust score and trainee console time yielded a Pearson coefficient of r=0.547 (Figure 2A). Similarly, correlation between the average trainee mLMX trust score and trainee console time yielded a Pearson correlation coefficient of r=-0.19 (Figure 2B). Additionally, correlation analysis between the average trainer mLMX trust score and average trainee mLMX trust scores yielded a Pearson correlation coefficient of r=0.329 (Figure 2C).

Trainees were then stratified into 2 groups based on their console times; >50% and <50%. 55.6% of trainees had a console time >50% and 44.4% had a console time <50%. Average trainer mLMX trust score was 4.08 for trainees with >50% console time as opposed to 3.16 for trainees with < 50% console time (p=0.038*). Moreover, average trainee mLMX trust score was 3.92 for trainees with >50% console time as opposed to 3.76 for trainees with < 50% console time (p=0.74).

Qualitative Analysis

Thematic analysis revealed a robust educational environment. Despite the use of isolated consoles, there was evidence of dynamic interaction and didactic exchange between trainers and trainees. Three major themes from our qualitative review demonstrated clear interactions between trust-autonomy and training dynamics (teaching, feedback, and communication) in the robotic surgery operating room: 1) effective teaching techniques, 2) feedback escalation, and 3) trust fostering behaviors. (Figure 3)

Effective Teaching Techniques

Surgical trainers employed several specific techniques to enhance teaching effectiveness in robotic surgery, and this theme included sub-themes of clear verbal articulation with agreement, pre-performance face-to-face huddles, active feedback and explanation, and engaging questions.

Most prominent was the impact of clear verbal articulation with concise, simple, and mutually understood vocabulary and concepts, allowing the trainee to perform the action while communicating an important and timely principle. For example, during a low anterior resection (LAR) of the rectum (Case 4), effective communication from the attending surgeon both provides immediate direction and explained the rationale for specific camera horizon changes as the operative field moves from the left lower quadrant to the pelvis. In another LAR (Case 9), the trainer specifically asks their trainee to move, rotate and center the camera view relating to specific visualized structures and plane orientation as they are proceeding with medial to lateral dissection: “pull your camera a bit more…make this [plane] parallel…rotate the other way.” In several other instances the attending instructs their trainee, “push towards the sacrum” or “tip towards the head” communicating with their trainee clear and concise instruction about intended directionality. During a robotic lung wedge resection (Case 6), the attending surgeon directly instructs the trainee on how to proceed with stapling the lung parenchyma, utilizing concise language (e.g. “drop the elbow,” “rotate the stapler counterclockwise,” “advance/retract the stapler”) followed by affirmation when done correctly.

Another effective teaching technique observed in our data was a pre-performance face-to-face huddle. In this behavior, trainer and trainee raise their heads out of their respective consoles for a face-to-face exchange. (Figure 4) The exchange was typically verbal but at times involved demonstrative hand gestures, e.g. physically demonstrating how to align the stapler properly with respect to the colon and mesentery just before resection during an LAR:

Attending: Like this…Colon, [with the] mesentery down…[go] straight across…normally we go anti-mesenteric, but now go straight across. [tapping resident on shoulder to initiate “head out” interaction; gesturing with hands to demonstrate colon and mesentery, holding out two fingers to represent stapler coming straight across vs. “antimesenteric”] (Case 4, Figure 5)

In another example demonstrating how to orient the mesh in an inguinal hernia repair, the attending surgeon similarly initiates a face-to-face interaction and then gestures to demonstrate a teaching point:

Attending: The mesh is laying like this…here is the ML…it needs to [orient] like that… what I want you to do is grab the top and swing it like this. (Case 7, Figure 6)

Pre-performance huddles usually took place just before the start of a case for a preoperative debrief, but might also occur during the procedure, just before a critical step or when a concept is particularly challenging to explain verbally. They were notable in our observations as an important moment for teaching, communication, and reinforcing dyadic trust.

Active continuous feedback and explanation were also observed to be an effective teaching technique. Trainers provided ongoing commentary either verbally or through telestration as the trainee operated, rationalizing a certain technique, emphasizing the importance of a certain structure, or giving recurrent affirmations. One trainer’s utterances during a robotic inguinal hernia repair illustrate this sub-theme:

Attending: Yep. Big spreads. One deeper. Yep. Nice. Yes. Good. And so here you're not going to be able to get all the way to Cooper's because it's a direct, right?

Resident: Right.

Attending: But you can do some more. It's coming. Do what's easy.

Resident: It's so thin here. I'm trying to get a little more tissue.

Attending: You can pop pre-transversalis whenever you feel like you need to, because it doesn't really matter. Because when you get down here, they will become the same plane, right?

Resident: That's true. (Case 3)

In this instance, the attending surgeon provides multiple affirmations and explains why a particular dissection is not suitable, allowing the trainee to adjust their technique in real-time. This strategy maintains engagement for both trainer and trainee, facilitates the progress of the procedure, organically reveals the parameters of the dissection, and reduces the likelihood of the trainer taking control due to ongoing guidance.

The final sub-theme in effective teaching techniques involved utilization of open-ended questions to explore the trainee’s thought process and decision-making. This was frequently utilized when the trainee had control, but next steps were unclear, or the procedure was stalling:

Attending: What do you think?

Trainee: Sure.

Attending: Love your certainty. Are you SURE-sure?

Trainee: I was trying to get a little bit more posterior because I thought there was some more but ...

Attending: Some more what? Mesentery?

Trainee: Yeah. I think this side's fine. The right side I think is okay because I could see it decently well. It's the other side that I was struggling with.

Attending: Left posterior, okay.

Trainee: Yeah, like that fat at the tip, I was struggling to get the tension and stuff. Because I feel like that's lower than where we're going to staple across.

Attending: Could be. [Trainee continues dissecting] Okay, good. (Case 8)

This approach allowed the trainer to understand the trainee's rationale and offer guidance or corrections without taking over the operation.

Feedback Escalation

The second theme noted in our thematic analysis was Feedback Escalation, with stages specific to robotic surgery. (Figure 7) “Escalation” of feedback typically occurred when a trainee was unable to carry out a certain task, and was noted to be correlated with decreased trust, increased urgency, or situations in which a trainer was unable to adequately convey instruction verbally.

Feedback Escalation progressed through stages of 1) verbal feedback, 2) use of surgical telestration and on-screen guidance markers, 3) intraoperative "head-out" huddles, and 4) complete takeover of the console by the trainer, which might be either to demonstrate a specific technique, before returning control, or to complete the surgical task themselves. Generally, feedback escalation followed this stepwise progression. However, during particularly challenging tasks or when the trainer could not effectively explain a concept, it was common to skip steps. Accelerated escalation and skipping steps were observed with low trainer mLMX trust scores, while slower escalation was observed with high trainer mLMX trust scores.

The following examples illustrate Feedback Escalation during a robotic lung wedge resection, in two different contexts (Case 6). In the first context, the attending surgeon asked the trainee to control staple line bleeding. After verbal feedback and instruction, the trainee was unable to control bleeding because of mobile lung parenchyma. The attending took control of the console, demonstrated how to stabilize the lung parenchyma, and partially cauterized the staple line. They then gave control back to their trainee and the trainee was able to complete the task. This example showed accelerated escalation and skipping steps given the situation, demonstrative behavior on the trainer’s part, and then exhibition of “feedback de-escalation” by handing back control. In another context, the attending verbally instructed their trainee to use gauze to retract the lung. When the trainee was unable to retract appropriately, the attending then escalated to using on-screen markers to demonstrate where to apply the gauze and how to position the graspers. After multiple attempts, the attending took control and demonstrated the appropriate technique. The trainer then “de-escalated feedback,” handed control back to their trainee, and used on-screen markers until the trainee was able to emulate their technique.

Trust Fostering Behavior

A third major theme arising in our data described behaviors exhibited by trainees that promote trainer trust. Sub-themes included clarification of intent (verbally or physically) and closed-loop communication. Trust-fostering behaviors were notable in interacting with Feedback Escalation, to de-escalate from console takeover and regain trainee autonomy.

Intent could be verbalized along with physical pointing. For example, in an inguinal hernia repair, the trainee stated to the attending surgeon, “I am going to go here,” as they were fixing the mesh onto Cooper’s ligament during a robotic inguinal repair (Case 7), or “you probably want me to start in here?” while indicating with an instrument where the flap closure suture line might begin (Case 1). Physical gestures to indicate maneuvers before execution were also a common example of trust fostering behaviors that clarify trainee intent. Later in the same procedure, the trainee states, “I see this… and then I am going to do one here…and then I am going to pulley it up” pointing to the proximal edge of the flap and gesturing in a pulley motion before they close it. These statements allow for trainer affirmation and course correction, especially if a trainee is unsure of the appropriate steps.

In the sub-theme of closed loop communication, the trainee signals that they understood the trainer’s instruction by repeating it, sometimes with physical pointing. This method partially compensates for the lack for non-verbal cues typically present in laparoscopic and open surgeries. For instance, during another robotic inguinal hernia repair, the attending and trainee confirm mutual understanding of an anatomical structure:

Attending: See that? [utilizing on-screen pointers to indicate a structure] That’s the pseudosac, right there.

Resident: This edge, right here; I see it. [utilizing tip of instrument to outline the same structure on screen]

Attending: Yep. It's attenuated transversus… So that needs to go up, which is exactly what you're doing. Some people see that and think it's hernia sac and follow it. (Case 3)

This closed-loop communication confirms the trainee's understanding of the trainer’s reference. All the aforementioned actions were generally observed to preempt console takeovers and “de-escalate feedback.” In some instances, closed-loop communication takes place over an extended interaction, integrated with clarification of intent:

Attending: You see that? See your spot in the back where it connects?

Trainee: Yeah.

Attending: Yeah. So, get one jaw back there.

Trainee: Okay. [positions vessel sealer instrument]

Attending: There you go. That's right. Yeah.

Trainee: Ooh… That's firm though. [closing instrument without firing]

Attending: Is it?

Trainee: Yeah. It seems like it is. [continues to open and close instrument to indicate visual cues of tissue firmness]

Attending: I agree with that. So, let's see… So, what else?

Trainee: Either above, I can try taking it higher or ...

Attending: I don't think so. So, try taking… Is that tumor?

Trainee: I'm not sure. I mean… [tapping area of interest with instrument]

Attending: It might be. So cut just the peritoneum. [utilizing on-screen pointers to indicate area to cut]

Trainee: Like this right here? [pointing with instrument]

Attending: Yeah. Try that. (Case 8)

As a result of these trust fostering behaviors, the trainee maintains autonomy as the primary operating surgeon, even in situations of uncertainty.

Training dynamics in robotic surgery diverge from those of open and laparoscopic surgery. Traditionally, trainers could leverage nonverbal cues, such as posture, stance and facial expression; a non-confident stance, a confused frown or handling hesitation might signal a need for intervention. From a trainee perspective, a trainer’s nonverbal cues and physical guidance were likewise traditionally important educational components of their surgical training.^20–22 These dynamics informed the dyadic trust required for training, in addition to the related but incompletely overlapping framework of entrustment, or decisions informing the granting of autonomy.²³ Robotic surgery disrupts this dynamic: control is absolute and non-explicit cues absent. Trainers receive less input from their interpersonal environment to make teaching decisions, and trainees miss out on nonverbal teaching and feedback.

Our study findings suggest a sort of compensatory educational adaptation to the altered environment of robotic surgery: trainee autonomy is entirely contingent upon attending trust, but both parties have adapted their communication behaviors to promote dyadic trust. Trust is fundamental for autonomy (or “entrustment”): in our quantitative analysis, higher levels of trainer trust as measured by mLMX were associated with increased trainee autonomy. But higher levels of trainee trust (in the relationship) did not equate to increased autonomy and were not correlated with increased trainer trust either. This pattern persisted when console time was stratified into > 50% and < 50% groups. Trainer trust was statistically higher in the > 50% console time group compared to the < 50% group, while trainee trust levels showed no significant difference between the two groups. These results imply that trust functions unilaterally in robotic surgery, with the attending surgeon having the ultimate authority in granting autonomy. Trainee trust, even in the high console time cohort, did correlate to autonomy. For trainees who were getting to act as the primary surgeon, it wasn’t because they had built relational trust.

This may represent a relative disruption of the traditional side-by-side partnership model of training.²⁴ In cases with lower trainer trust as measured by mLMX, we noted more instruction, less active teaching, accelerated feedback escalation, skipping feedback steps, and increased console takeover, although our methodology was not specifically designed to measure or quantify these events. However, our qualitative analysis revealed several ways in which trainers and trainees work around the robotic platform, or use the new tools and affordances provided by the system, to reconnect teaching and communication to dyadic trust. Attending surgeons specified verbally, pointed on screen, and physically invited conversations outside the console to convey teaching points and make sure trainees completed tasks safely. Trainees described in detail, gestured, confirmed, and reiterated to foster trust and maintain autonomy. When the dynamic was insufficient, feedback was escalated to console handover—and in the best situations, console return.

To this end, effective teaching and feedback remain of utmost importance for the trainer / trainee relationship in robotic surgery. We found examples of effective teaching techniques in robotic cases involving continuous active verbalization and explanation from the trainer. These behaviors may partly compensate for the absence of nonverbal cues and physical guidance from the trainee’s perspective but were most effective when utilizing mutually understood terms. For instance, there was evidence of agreement in inguinal hernia cases that “up” meant “towards the anterior abdominal wall,” an agreement requiring mutual tacit knowledge which is a component of expertise in surgery.²⁵ We posit that explicitly codifying the teaching vocabulary used in robotic surgery could not only improve teaching effectiveness, but help trainers and trainees build trust. Preperformance huddles and open-ended questions were also found to be effective teaching techniques. These methods facilitated the explanation of difficult concepts that typically require traditional face-to-face communication and kept trainees engaged.

On their part, trainees employed trust fostering behaviors and signals to compensate for the lack of non-verbal cues traditionally demonstrated to their trainers, many of which aligned with behaviors known to promote entrustment.¹⁵ These include increased explicit communication, mirroring the compensation strategy used by their trainers. As trainees progress through a case, they signal their intentions and next steps to the trainer, describing what they observe, perceive, and anticipate. These verbal communications are supplemented by physical indications of maneuvers prior to execution. Closed loop communication signals good understanding of concepts and fosters increased trust with the trainer. Although trainee trust was not found to be directly correlated with autonomy in our quantitative data, our study may have been unable to capture nuances of this trust direction that impact how much a trainee is willing to exhibit trust fostering behaviors. In other words, while a trainee who demonstrates they can do the job well is typically given more opportunities to operate, it may also be the case that a teacher who demonstrates that they can teach well is given more opportunities to teach. Open-ended questioning and demonstration of trust fostering behaviors may be bidirectional trust markers of a high-quality teaching relationship.²⁶ While our mixed methodology allowed us to note some broad themes and notice some trends between our quantitative and qualitative data, our small sample size prohibited us from making connections and subsequent broad conclusions.

Limitations of our study include a small sample size. A larger cohort might help further delineate the observed correlational relationships. It may have also offered a more comprehensive understanding of differences related to specialty, case type, case urgency, or trainee level. Our qualitative data collection was effortful, but represented a broad variety of case types, teachers, and learners. However, expanding the sample beyond a single institution might provide additional cultural insights not seen in our data. Another potential limitation is the introduction of additional recording equipment into the operating room. Although the data collection strategy was designed in a thoughtful manner and video taken in as unobtrusive a manner as possible, any type of recording may result in a small amount of participant reactivity (changes in behavior when participants know they are being observed).²⁷ We utilize console time as a reasonable surrogate measure of trainee autonomy.¹⁷ Primary operating time is difficult to objectively measure in open and laparoscopic surgery, but easily extracted from the DaVinci Xi® system. However, console time may not provide an accurate picture of whether the trainee is performing a substantive portion of the case, even if they are performing a substantial portion. Future work aimed at segmentation of robotic procedures may help improve objective measures of trainee autonomy. Finally, at our institution, consoles are arranged in a side-by-side configuration, which facilitates effective “head-out” huddles. In contrast, other institutions with consoles facing each other may find it challenging to engage in this type of feedback behavior but develop different ways of engaging with the affordances of their own operating room configuration.

The robotic console has introduced unique configurational challenges to the operative educational environment. Our mixed methods study sheds light on the shift in communication between a trainer and a trainee in this new isolated arrangement. To mitigate these challenges, we examined how new training dynamics in robotic surgery intersect with established notions of autonomy and trust. Our findings highlight effective strategies that promote not only good teaching practices, but also trainee autonomy and trust fostering behaviors. We also highlight the need for increased bidirectional clear and simple verbal communication to compensate for the absence of traditional nonverbal communication in robotic surgeries. As this technology gains increasing prominence in the operating room and surgical training programs, our findings serve as an important resource in building more robust educational curricula to train the next generation of robotic surgeons.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Nwaelugo NS, Goldblatt MI, Gould JC, Higgins RM. The evolution of the general surgery resident operative case experience in the era of robotic surgery. Surgical Endoscopy. 2022;36(9):6679-6687.
Sheetz KH, Claflin J, Dimick JB. Trends in the adoption of robotic surgery for common surgical procedures. JAMA network open. 2020;3(1):e1918911-e1918911.
Okamura AM. Haptic feedback in robot-assisted minimally invasive surgery. Current opinion in urology. 2009;19(1):102-107.
Green CA, Chu SN, Huang E, Chern H, O’Sullivan P. Teaching in the robotic environment: use of alternative approaches to guide operative instruction. The American Journal of Surgery. 2020;219(1):191-196.
Green CA, Lin JA, Huang E, O’Sullivan P, Higgins RM. Enhancing robotic efficiency through the eyes of robotic surgeons: sub-analysis of the expertise in perception during robotic surgery (ExPeRtS) study. Surgical Endoscopy. 2023;37(1):571-579.
Wang TN, Woelfel IA, Huang E, Pieper H, Meara MP. Behind the Pattern: General Surgery Resident Autonomy in Robotic Surgery. Heliyon. Published online 2024.
Green CA, Lin J, Higgins R, O’Sullivan PS, Huang E. Expertise in perception during robotic surgery (ExPeRtS): What we see and what we say. The American Journal of Surgery. 2022;224(3):908-913.
Carpenter BT, Sundaram CP. Training the next generation of surgeons in robotic surgery. Robotic Surgery: Research and Reviews. Published online 2017:39-44.
Puliatti S, Mazzone E, Dell’Oglio P. Training in robot-assisted surgery. Current Opinion in Urology. 2020;30(1):65-72.
Graen GB, Uhl-Bien M. Relationship-based approach to leadership: Development of leader-member exchange (LMX) theory of leadership over 25 years: Applying a multi-level multi-domain perspective. The leadership quarterly. 1995;6(2):219-247.
Dulebohn JH, Bommer WH, Liden RC, Brouer RL, Ferris GR. A meta-analysis of antecedents and consequences of leader-member exchange: Integrating the past with an eye toward the future. Journal of management. 2012;38(6):1715-1759.
Ilies R, Nahrgang JD, Morgeson FP. Leader-member exchange and citizenship behaviors: a meta-analysis. Journal of applied psychology. 2007;92(1):269.
Martin R, Guillaume Y, Thomas G, Lee A, Epitropaki O. Leader–member exchange (LMX) and performance: A meta‐analytic review. Personnel psychology. 2016;69(1):67-121.
Rockstuhl T, Dulebohn JH, Ang S, Shore LM. Leader–member exchange (LMX) and culture: A meta-analysis of correlates of LMX across 23 countries. Journal of applied psychology. 2012;97(6):1097.
Teman NR, Gauger PG, Mullan PB, Tarpley JL, Minter RM. Entrustment of general surgery residents in the operating room: factors contributing to provision of resident autonomy. Journal of the American College of Surgeons. 2014;219(4):778-787.
Wagner JP, Lewis CE, Tillou A, et al. Use of entrustable professional activities in the assessment of surgical resident competency. JAMA surgery. 2018;153(4):335-343.
Wang TN, Woelfel IA, Pieper H, Haisley KR, Meara MP. Is Robotic Console Time a Surrogate for Resident Operative Autonomy? Journal of Surgical Education. 2023;80(11):1711-1716.
Riger S, Sigurvinsdottir R. Thematic analysis. Handbook of methodological approaches to community-based research: Qualitative, quantitative, and mixed methods. Published online 2016:33-41.
Braun V, Clarke V. Using thematic analysis in psychology. Qualitative research in psychology. 2006;3(2):77-101.
Sutkin G, Littleton EB, Kanter SL. How surgical mentors teach: a classification of in vivo teaching behaviors part 2: physical teaching guidance. Journal of Surgical Education. 2015;72(2):251-257.
Moore A, Butt D, Ellis‐Clarke J, Cartmill J. Linguistic analysis of verbal and non‐verbal communication in the operating room. ANZ journal of surgery. 2010;80(12):925-929.
Bezemer J, Cope A, Faiz O, Kneebone R. Participation of surgical residents in operations: challenging a common classification. World journal of surgery. 2012;36:2011-2014.
Schumacher DJ, Cate O ten, Damodaran A, et al. Clarifying essential terminology in entrustment. Medical Teacher. 2021;43(7):737-744.
Torbeck L, Wilson A, Choi J, Dunnington GL. Identification of behaviors and techniques for promoting autonomy in the operating room. Surgery. 2015;158(4):1102-1112.
Alderson D. Developing expertise in surgery. Medical teacher. 2010;32(10):830-836.
Sandhu G, Thompson-Burdine J, Nikolian VC, et al. Association of faculty entrustment with resident autonomy in the operating room. JAMA surgery. 2018;153(6):518-524.
Paradis E, Sutkin G. Beyond a good story: from Hawthorne Effect to reactivity in health professions education research. Medical education. 2017;51(1):31-39.

Tables 1 and 2 are available in the Supplementary Files section.

Table1.png
Table 1: Distribution of Cases
Table2.png
Table 2: Demographics of Trainers and Trainees
Appendix1Survey.docx

Download PDF

Reviewers agreed at journal
24 Oct, 2024
Reviewers invited by journal
23 Oct, 2024
Editor invited by journal
23 Oct, 2024
Editor assigned by journal
23 Oct, 2024
First submitted to journal
17 Oct, 2024

You are reading this latest preprint version

Trust, Autonomy, and Teaching Dynamics in Robotic Surgery: A Mixed Methods Study

Status:

Version 1

Abstract

Figures

Introduction

Methods

Case Selection and Recruitment

Intraoperative recordings

Surveys

Media Post-processing

Quantitative Analysis

Qualitative Analysis

Results

Discussion

Conclusion

Declarations

References

Tables

Supplementary Files

Status:

Version 1