Making Surgery as Safe as It Should Be: A Qualitative Study

Abstract

Existing literature supports the view that adverse outcomes from surgical interventions are more likely to be the result of degraded nontechnical skills (NTS) rather than the technical skills of surgeons. In the present context, NTS comprise the behaviors and cognitions deployed by surgeons to make decisions, maintain awareness of the operating environment, communicate with and lead supporting personnel. A contemporary safety thesis suggests that focusing on what makes things go right (eg, NTS) is far more productive than retrospective analysis (learning from mistakes). The present qualitative research explored how surgeons deploy NTS to facilitate safe and effective outcomes from surgical interventions. The thematic analysis revealed that this surgical cohort engaged specific self-regulated NTS along an intervention construct consisting of planning, implementation, monitoring progress, and deliberate learning through reflection during the preoperative, operative, and postoperative phases of care. Behaviors supporting these contentions were identified in the data and used to amplify use of the construct.

Introduction

In the present context, nontechnical skills (NTS) comprise the behaviors and cognitions deployed by surgeons to make decisions, maintain awareness of the operating environment, communicate with and lead supporting personnel with the view to producing reliably safe outcomes.

Weick and Sutcliffe¹ contend that high reliability requires continual self-assessment and reinvention. Accordingly, highly reliable organizations are, by definition, highly mindful of the potential for error. They recognize that most errors are complex and try to tease out the root causes. This reactive activity is the hallmark of organizations, which have adopted a traditional approach to performance improvement.² However, a contemporary thesis² suggests that focusing on what makes things go right will be far more productive. In surgery, the contention is that, on a day-to-day basis, the operating room functions in the same way except that on some occasions adverse outcomes emerge because of a confluence of disruptive factors.²

When translating this thinking to the health care sector, Hollnagel et al³ amplify the concept that reliability results from the flexibility and adaptiveness demonstrated by practitioners when accommodating operational variability. They further propose that, unlike aviation, safety cannot be managed by the imposition of workplace constraints. Hollnagel et al³ propose that “safety is defined by what happens when it is present” (p. 25). To wit:

• Operators thoroughly and effectively anticipate events.

• Most things go right but operators remain alert to possible failure.

• Operators devise flexible solutions to issues that threaten the production cycle (ie, an intervention in the surgical context).

• Operators have a tacit understanding of how to make things go right.

• Some performance variation is expected, recognized, and managed.

Dekker⁴ supports this notion when he reports that the difference between interventions that went well and those that went badly did not pivot on whether the usual suspects of human and system shortcomings were present. Rather, when things turned out well there were more of the following attitudes and behaviors present:

• more speaking up than shutting up

• not taking things for granted and being ready to adapt

• deferring decision to the real experts

• moderating team hierarchies

• not relying on audit to prompt improvement (ie, being engaged in rather than prodded to self-improve).

The types of activities proposed^3,4 can be seen as generative. That is, preconsidered and intentional attitudes and behaviors deployed to enable success.

In summary, Hollnagel et al³ regard the combination of proactive generative activities and reactive (responding to events) processes as the most desirable safety posture for health care; that is, a proactive culture that is informed by reactive processes. Dekker⁴ supports the safety framework proposed by Hollnagel² in that things go right because of generative operator attitudes and behaviors.

These generative attitudes and behaviors^3,4 generally describe NTS.^5–8 Behavioral tools such as the Human Factors Analysis and Classification System for Healthcare have been used to identify impending failures in the application of NTS and other inhibiting human factors.⁹ While the nontechnical skills for surgeons system of assessing the NTS of surgeons in the operating room has been shown to be reliable,⁵ the underlying supporting framework of NTS application by surgeons has not been identified. Knowledge of this supporting system will make it easier to select, train, and assess surgeons.

Methods

The objective of this research was to investigate and develop a novel approach to support nontechnical skills training and feedback to surgeons, in particular trainee surgeons. Semistructured interviews with extensive follow-up questioning were used to gather qualitative data.^10,11 Each recorded session lasted between 45 and 60 minutes. Consideration of an unfamiliar procedure (to reduce the limitations resulting from automaticity) was used as the stimulus for participant thinking (ie, “Due to circumstances beyond your control, you have to complete a procedure which you would rather not do because you lack both recent and in-depth experience with the procedure.”). Participants were then asked to verbalize their thinking about activities that they would engage to assure the desired surgical outcome in this less than desirable circumstance. Example of the intent of questions used to initiate/progress discussion on the proposal are as follows:

Is this situation a problem for you?

How would you approach this scenario?

What are the aspects of any proposed intervention that need the most attention?

What would you do to ensure a successful outcome?

How would you progress the intervention?

How would you know that the intervention was progressing in the correct direction?

What is the most difficult aspect of this situation?

At what point would you declare any intervention successful?

In keeping with qualitative methodology, only volunteer participants were interviewed. These participants were representative of specialty, gender, and experience. Data collection was planned to continue until saturation was verified.^12–14 Saturation verification was facilitated by a validated methodology, which nominates a process of base interviews, run interviews, and a new information threshold.¹⁴

In the present research, a base size of 4 interviews was initiated. After coding data from the first 2 interviews, a dual tier coding structure was adopted. This parent/child classification was used by the authors to code primary and secondary themes (ie, themes and subthemes) for the remainder of the data coding. Discrepancies were resolved by reanalyzing the source text and arriving at a consensual understanding. The base interviews were followed by a run of 12 interviews to enable a 0% new information saturation level to be reached and confirmed. The authors’ previous research and vocational coding experience of case narratives to identify areas of concern were instrumental in the recognition of saturation.

Following approval of the study by the facility Ethics Committee, surgeon volunteers were accepted principally from the deidentified cohort of Staff Specialists and Visiting Medical Officers. However, ≈20% of respondents with different affiliations were accepted to expose any possibility of site bias. All participants provided informed consent verbally.

The contributing cohort of surgeons (n = 16) comprised approximately one-third with <10 years of independent practice, one-third with between 10 and 20 years of independent practice, and one-third with >20 years of independent practice. The participants were accepted to establish a sample of 2 from each of 8 specialties and to maintain a representative gender ratio of female to male surgeons (1:2).

Results

In response to the initial and follow-up questions, each participant verbalized activities they engaged into assure positive outcomes. These activities were coded and categorized into recurring themes and subthemes (2 level coding). Sample size adequacy was evident since data collected during the base interviews (total of 4 comprising 2 initial and 2 subsequent dual level coded) and the run interviews yielded no new recurring themes (0% new information) after a total of 13 interviews.¹⁴ Three (3) extra run interviews were conducted to confirm the 0% new information saturation level.

The depth of the data that facilitated saturation resulted from the use of open questions and probing follow-up questions.^12–14 In excess of 400 quotation, paraphrase, and summary exemplars of behaviors and cognitions contributed to the authoring of the results narrative. These data all contain references to the NTS subject matter areas of decision-making, communications, situational awareness, and leadership^5–8 used during an intervention. Interventions were found to require preparation, planning, implementation, progress monitoring, and reflective learning to achieve the desired outcome.

Examples of all identified themes and subthemes are given below and in Table 2. Throughout this study, the years of independent practice of the surgeons from Table 1 is indicated by closed brackets around a number [eg, (29)].

Table 2.

Dominant Self-Regulating Activity Themes/Subthemes.

The dominant self-regulating activity themes/subthemes extracted from the data are detailed below as reflective questions. The originating themes/subthemes are included in brackets after each activity. Accordingly, did I:

Arrive at work not fatigued or stressed with adequate time to prepare? (preparation and planning/motivation) Exclude inappropriate pressure from others, bias, inappropriate personal motivation, and Group Think from the DTO and the Consenting processes? (preparation and planning/decision-making) Complete the DTO after developing a thorough understanding of the case pathology and only then conduct an ethical consenting process? (preparation and planning/decision-making) Develop and rehearse detailed normal and contingent plans for the procedure beforehand? (preparation and planning/mental modeling) Treat fellow clinicians, other staff, and patients with respect and courtesy? (implementation/team coordination) Show interest in other team members to elevate cohesiveness? (implementation/team coordination) Give other team members permission to speak up? (implementation/team coordination) State plans and check understanding by others? (implementation/team coordination) Generally, influence others and only direct when necessary? (implementation/team coordination) Prioritize tasks according to the must/should/could hierarchy? (implementation/maintaining focus) Manage all distractions and complete all required activities? (implementation/maintaining focus) Continuously compare what was happening against what should have been happening to identify discrepancies? (monitoring progress/detecting discrepancy) Resolve all discrepancies to avoid adverse outcomes? (monitoring progress/maintaining direction) Maintain sufficient spare cognitive capacity to make timely decisions and act decisively in time critical situations? (monitoring progress/maintaining direction) Escalate assertive language when certain that proposed actions are not in the best interests of the patient? (monitoring progress/maintaining direction) Actively listen for input from others? (monitoring progress/receiving assertive input) Provide suitable feedback and coaching to others? (monitoring progress/reflective practice) Admit to any lack of knowledge and own my errors? (monitoring progress/reflective practice) First consider that I might be the problem if a conflict arises? (monitoring progress/reflective practice) Critique my actions and decide on what, if anything, to do differently next time? (monitoring progress/reflective practice/deliberate learning)

Abbreviation: DTO, decision to operate.

Table 1.

YOIP of the Surgeons Who Engaged in Discussions.

Surgeon	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16
YOIP	4	34	11	29	3	4	10	6	18	33	6	10	20	24	17	31

Abbreviation: YIOP, years of independent practice.

• Preparation and planning—“It’s all about pre-empting what could go wrong.” (33).

  • Motivation—“If you do a lot of surgery, then the patient is going to get a surgical solution. If your practice is largely noninvasive, then the patients will get a nonsurgical solution.” (24).

  • Decision-making and judgement—“Decision-making itself is not a skill that we are particularly taught.” (17).

  • Preoperative decision-making—“The most difficult decisions in surgery are whether to operate, not how to do it. Especially when the decision is whether to reoperate.” (29).

• Implementation—“I chat to the whole team beforehand so that pretty much everybody knows what the plan is.” (11).

  • Team coordination—“If we have the luxury of time, I’ll talk to the nurses the week before and it’s usually the senior nurse that’s consistent because the scrub and scout team can change.” (6).

  • Maintaining focus—“It’s good for a theater for people to be orderly and focused without being repressed into silence. It can get out of hand if people are too social and not focusing on their job. A lot of noise in the theater is counterproductive.” (29)

  • Emotional considerations—“The only thing that I need to do is to ensure that I am calm and in control of the situation.” (11).

  • The procedure—“You’re looking at the steps that are coming.” (10) and “It’s one thing to get the patient off the table but you don’t want to cause a situation where they could be sick for years and years.” (31).

• Monitoring progress—“There is a low level of irritation and distress that you get because something is not quite right. When it’s gross, it’s easy to spot.” (4).

  • Detecting discrepancy—“As you proceed along you need to understand your own thought processes as well as what’s going on around you.” (4).

  • Maintaining direction—“What should be happening might change because you haven’t been able to do what you thought you would be able to do.” (10).

  • Receiving assertive input—“Taking offence at considered input can be a lack of confidence. You need a more level playing field for responsibility to contribute to good outcomes. But surgeons need to feel supported.” (34).

  • Reflective practice—“Self-critique is very important to identify your discrepancies and deal with them through life and surgery.” (11).

  • Postoperative decision-making—“Post-operatively, it gets a lot more complicated because you not only have to account for the preoperative and operative factors, but you have all these other factors coming in from everywhere.” (6).

After further analysis of the narrative, it was possible to compile groups of paraphrases and summaries from the data that were preoperative, operative, and postoperative activities. This in-depth analysis did not produce a sequential checklist. Rather activities were completed/revisited as deemed necessary by the surgeon. The participants referred to the threat of becoming paralyzed by repeating gathered information analysis ad infinitum (ie, paralysis by analysis). Participants also contended that they would seek help early if they became uncertain of how to proceed at any stage.

Preoperatively, they proposed the following:

• Gather data and information (colleagues, experiences, the patient, the family, tests, and scans can all contribute).

• Built robust case awareness (ie, situational awareness) by synthesizing all intelligence to develop alternatives (including recommending an alternate therapy).

• Decide with patient participation after reviewing alternatives and exploring material risks.

• Plan for best case scenario and worst-case scenario by developing a comprehensive plan that caters for all foreseeable contingencies.

• Build a robust mental model (which is populated with actions and requirements) to free up intraoperative cognitive space so that discrepancies can be detected and surprises can be managed.

• Rehearse by visualizing mental and physical activities required for normal and abnormal facets of the procedure.

Intraoperatively these participants proposed the following:

• Build team cohesion by showing interest in others and giving permission to team members to assert opinions as they thought appropriate.

• Brief normal and exceptional issues (and planned coping strategies and actions) during time out.

• Upgrade plans by incorporating any new data or information to assure an accurate and progressive working model.

• Establish decision action points whereby uncertainty must be resolved before proceeding.

• Apply technical skills as taught, approved, and expected by the team.

• Monitor progress so that when what is happening is not what should be happening, identified threats and discrepancies can be managed.

• Manage workload using techniques (such as progressive modeling) to generate surplus cognitive capacity.

• Complete time out procedure before operating and share learnings with team members (giving credit when due) postoperatively.

• Write up notes ensuring that they are comprehensive and clearly detail postoperative care requirements.

Postoperatively, participants concluded that patients normally exhibit some awareness of the care that they are receiving. The safety and quality of surgical interventions can rise or fall on the standard of this phase (24). Participants contended the following during this important and sometimes uncertain phase:

• Avoid denial since the perfect operation does not necessarily beget the perfect outcome.

• Visit patients frequently because postoperative care quality is increased by frequent physical reviews.

• Interpret clues and cues by synthesizing vital signs (nb respiratory rate), wound management, and general patient demeanor.

• Give clear instructions to ward manager and staff on upgraded/changed management and contemporaneous notes.

• Time critical actions such as drain removal and discharge carefully.

• Reflect to deliberately learn by using a meaningful framework to determine what went well and what could be done better next time. Learnings would be shared with team members and colleagues.

Discussion

The behaviors attributed to the preparation, planning, implementation, and monitoring progress phases of surgical interventions as described above and referenced in this discussion amplify the propositions and findings of Hollnagel et al³ and Dekker.⁴ These behaviors are clinician driven (rather than system protocols) and purposed to enable individuals to secure safe outcomes of high quality. As NTS, they facilitate the use of the necessary technical skills.^5–8 These NTS are deployed by surgeons to meet the hierarchical decision-making challenges of the preoperative phase, the postoperative phase, and the operative phase of surgical interventions:

1. “The real art of surgery is decision-making. That is the most difficult thing. If you have the choice between a surgeon with good judgement and a surgeon with good hands you would go with the surgeon with good judgement every time.” (29).

2. “Postoperative decision-making has very similar issues to the preoperative phase.” (6).

3. “People tend to underestimate and recognize big time problems and they are too slow to act because there is a form of surgical denial. This is a problem because of this incapacity to acknowledge critical deviations from what should be happening.” (24).

4. “The surgery side of things is not where the stress comes from necessarily other than those extreme cases where I’m thinking about the problems that I am not sure how to solve until I’m in there.” (10).

When viewed holistically, these NTS themes and subthemes mimic constructs of self-regulation.^16–19 The notion of self-regulation has long been associated with education.¹⁶ Some researchers have previously linked production and safe outcomes with the ability of operators to self-regulate.^18,19 In the production (surgical intervention) context, self-regulation is said to be the ability to track awareness of one’s own mental and physical activities selected to safely produce the desired outcome.¹⁹ In the surgical setting, participants are expected to self-regulate their approach to care²⁰ and to produce safe outcomes. From this research, it seems that safe surgical outcomes are dependent on self-regulation^18–21 of process and self-regulation of attitude to task as stipulated in the surgeons’ Code of Conduct.²¹

Accordingly, any activity that engages NTS to produce self-regulated safety is worthy of consideration since this is proactive. Self-regulation generates cognitive and behavioral activities, which enable (as opposed to inhibiting) preparation and planning, the implementation of plans, monitoring task progress, discrepancy recognition, and the devising of oft times unique interventions^3,4 to control outcomes (see preoperative, operative, and postoperative activities above). From the data, participant surgeons believe that the activities noted above reduce the chances of unfavorable outcomes. The translation of these surgical experiences into expertise was seen to be accomplished by deliberate learning through reflection.

This present study adopted a domain centric grounded approach. Further, this study met qualitative criteria in that the resulting narrative was derived solely from the quotes, paraphrases, and summaries of recorded interactions with participants. As these participants of representative gender and ages were drawn from 8 specialties, this cohort comprised a rich cross-section of experience in independent surgical practice. Apart from the tendency of less experienced surgeons to talk with peers before proceeding, no other substantial NTS differences could be identified across the participant cohort.

As a result of this robust approach, it can be reported that the recurring themes with subthemes saturated the data. Further, as proposed by academics,¹² these data demonstrated the capacity to speak for themselves. The resulting narrative demonstrated low inconsistency, high certainty, and, at worst, a low-to-moderate risk of bias because the goal of all participants was to avoid adverse outcomes.

This research was conducted in a jurisdiction where surgical outpatient clinics are not normally conducted. Consequently, trainees do not get practice or coaching in the decision to operate until relatively late in their traineeships. This may have rendered a degree of overemphasis on the decision to operate difficulty by the less experienced surgeons. More data will be required to clarify this situation. However, it is clear that the responding surgeons completed setting specific activities designed to establish what should happen so that corrective or alternative actions could be initiated upon recognition of a discrepancy (ie, development and maintenance of situational awareness).^6,7

The results of this research might be considered to develop a surgical approach to the teaching and practice of NTS. This research could also be adapted to research the teaching and practice of NTS in other health care settings. The usefulness of this research to the trainee selection process has yet to be explored. However, any research that speaks to a positive outcome enabling process^18–21 should be able to inform selection processes. Notwithstanding, more data may be required to ensure that this application is robust.

Since this study is limited in scope, further research might also be undertaken to assess the following:

• The prevalence of these NTS application in routine interventions.

• The ability of these data to inform NTS needs analysis tools.

• The usefulness of these data to construct an NTS feedback tool for surgeons.

• The contribution of support staff to the successful deployment of NTS by surgeons.

Notwithstanding these limitations, trainees and surgeons in independent practice might find it useful to engage the deliberate learning (reflective) activities sourced from the data gathered during this research.

These types of deliberate task activities enable surgeons to stay in touch with their own thinking throughout procedures to assess task requirement progress.²² Activities with this intent may well become automated²¹ with experience and render expertise. Further, if resilience^2–4 requires that competence be maintained under stress, detailed but flexible mental models constructed by completing activities of this nature can contribute positively.²³ Typically, a highly self-regulated surgeon would automatically use these activities to enhance their expertise after every surgical intervention not just when something does not go right. Deliberate (reflective) learning is an example of what happens when safety happens.^2,4

Given the historical professional independence claimed by the surgical profession,²⁴ successfully understanding and engaging NTS is potentially more proactively useful to surgeons than feedback from more invasive techniques used by some approaches to safe operator assurance (eg, convincing peer review²⁵). Certainly, research has shown that deficient NTS are instrumental in many of the adverse outcomes, which result from surgical interventions.^26,27 and only limited training (usually voluntary) is available.^28,29 As discussed in the introduction, NTS can be seen as generative behaviors^2,4 which attach to the Safety II approach³ (behavior focused) rather than the Safety I approach³ (outcomes focused) as is presently widely engaged by the health care sector.^30,31

Conflicts of Interest

The authors have no conflict of interest to disclose.

ORCID ID

David Robinson: 0000-0001-6301-7780