Is team‐based perception of safety in the operating room associated with self‐reported wrong‐site surgery? An exploratory cross‐sectional survey among physicians

Abstract Aims Participation in wrong‐site surgery may negatively influence the perception of safety by the health care professionals in the operating room (OR). The objective was to explore if perception of safety in the OR was seen as a team‐based or individualist concern and whether having participated in wrong‐site surgery was associated with perception of safety. Method and Results Cross‐sectional survey at 2 annual meetings of surgery, in Switzerland, 2010. We used multivariate generalized models to assess the association of perception of safety in the OR (1 item) with self‐reported participation in wrong‐site surgery—overall, past (more than 3 y ago), or recent (last 3 y) participations—controlling for sociodemographic characteristics and opinion of the surgical safety checklist. One hundred ninety respondents answered the questionnaire (participation rate of 22.6%). Respondents mostly had a team‐based, rather than an individualistic, perception of safety in the OR. In multivariate analyses, the influence of ever participation in wrong‐site surgery was not significant. However, past participation in wrong‐site surgery (more than 3 y ago) was associated with perception of safety as team based, whereas recent participation (last 3 y) was associated—despite not significant at α ≤ 5%—with perception of safety as individualistic. Conclusion In this sample, safety in the OR is most often seen as team based rather than individualistic. Perceiving safety in the OR as team based varies according to recent or past participation in wrong‐site surgery. Longitudinal research is needed to assess causality between participation in wrong‐site surgery and change in perception of safety.

Safety culture is one important dimension of patient safety 1 and reflects values, attitudes, and behaviors that health care professionals have in common when administering care to patients and includes their perception of safety as an individual and/or team-based responsibility. Health care professionals are trained to be diligent in their care of patients and to follow the guiding principle "first do no harm." Over the past decades, the development of a safety culture in health care -and, specifically, in the operating rooms (ORs)-has become a significant issue, and efforts have been made to enhance the importance of safety as a collective, rather than an individual, process. 2 Yet changing attitudes towards safety perceptions has been difficult, mostly because of cultural reasons. 3,4 Furthermore, an excessive emphasis on individual responsibility can discourage team-based safety procedures, such as the implementation of the surgical safety checklist. 5 There has been significant progress in safety culture in health care settings over the past decade. The monitoring of safety indicators shows that health care professionals are more inclined to report medical errors, adverse events, or near-miss errors. 6 The worldwide implementation of the surgical safety checklist 5

by the World Health
Organization's initiative "Safe Surgery Saves Lives" 7 and its frequent use by OR teams [8][9][10][11][12] suggest that OR health care professionals have endorsed this team-based safety procedure in their daily routine.
However, in the OR, a persisting perception of safety uniquely based on individual competency may be problematic, because it can impede effective communication within the team and appropriate handling of team-based safety procedures. 13 Both individual and team-based competencies are required to insure good safety of care in the OR.
However, for historical reasons, many surgeons believe that OR safety is their responsibility. Currently, the prevalence of an individual-based perception of safety in the OR is unknown.
Wrong-site surgery is potentially devastating for the patient, but also for the health care professionals, who can become the "second victim" of these errors. [14][15][16] Participation in errors (whether the health care professionals is responsible or not) can result in psychological regrets 17 and mental health problems 18,19 and affect psychological well-being. 20 It may also negatively influence the perception of safety in the OR as a process that should be team based, but knowledge of this influence is limited. A study among otolaryngologists investigated the corrective actions to errors and adverse events and showed that actions included not only patients (error disclosure and mitigating the consequences of the errors) but also care practices within their units and departments, like time-out, cross-checking of patient identity, medications labelling, and other surgical protocols. 21 Defensive changes, such as keeping the error for oneself, distancing, and escaping or avoidance, can affect clinical practice. 16 However, it is unknown if defensive changes can modify safety attitude and, in particular, can affect the perception of safety as an individual responsibility. After participation in wrong-site error surgery, the perception of colleagues can influence the feeling of guilt. 16 The fears of loss trust and/or reputation by their colleagues and of making another error can change the attitude towards more individualistic safety attitudes. Conversely, constructive changes such as accepting responsibility, problem solving, seeking advices, learning from mistakes, and promoting changes could increase the perception of safety as a team-based process. The impact of wrong-site surgery on the "second victim" and how he/she responds is also influenced by the support provided by the health care organization in which the professional works. Unfortunately, this support is all too often lacking. 22,23 To evaluate the link between wrong-site surgery participation and perception of safety of care in the OR as either an individual or teambased process, we conducted an exploratory study using a self-administered questionnaire distributed in 2 annual congresses of surgery, thereby allowing for its use in an anonymous and neutral environment.
The primary objective of this exploratory study was to examine if safety in the OR was perceived as a team-based process or as an individualist process. Secondarily, the objective was to assess the factors associated with the perception of safety in the OR and, in particular, to explore if self-reported participation in wrong-site surgery was associated with it. In other words, we tested the hypothesis that self-reported participation in wrong-site surgery is associated with a more individualistic perception of safety in the OR.

| Study design and settings
We conducted a cross-sectional survey with a self-administered questionnaire. The questionnaire-available in German, French, Italian, and English (see Appendix S1)-was distributed during 2 surgery meetings in Switzerland: the 97th Annual Meeting of the Swiss Society of Surgery, 24

| Dependent variable
Perception of safety as team based or individualistic in the OR was assessed by a single item: "Regarding safety of care in the operating theatre, to what extent are you in agreement with the following opinion: Safety is an individual concern above all, and a team concern to a lesser extent" (hereafter, perception of safety). Answers were given on a scale of 1 (do not agree at all) to 5 (fully agree). The variable was reverse coded to figure a high score for team-based safety perception (and a low score of an individual-based safety perception). To assess validity of this single item, we reported a convergent validity analysis (see Appendix S2).

| Opinion of the surgical safety checklist
Respondents answered questions about their perceptions of the surgical safety checklist (see Appendix S3). We calculated a global score (between 0 and 100, high score indicating positive perceptions) when at least 3 of the 8 items were answered (N = 179), after reverse coding 3 items (waste of time, no extra value, and efficacy). The opinion of the surgical safety checklist has been examined in more detail elsewhere. 26

| Statistical analysis
We examined univariable associations with perception of safety in the OR using analysis of variance when the independent variable was categorical and simple linear regression when the independent variable was continuous. Then, we examined 5 multivariate linear models:  Respondents were mostly men, surgeons, employed, without postgraduate training in another country, and with a median age of 42.5 years (range: 20-79) (see Table 1 Participation in wrong-site surgery (irrespective of the type of errors) was frequent, with 36.2% of respondents reporting at least one error over their whole practice (mean number of 0.7 errors) and 16.6% of respondents reporting a recent participation in wrong-site surgery, ie, in the last 3 years ( Table 1). Types of errors are reported in Appendix S4.

| Perception of safety in the OR
Respondents mostly rejected the idea that "Safety is an individual concern above all, and a team concern to a lesser extent": the mean score In the multivariable analysis (Table 3)     participation in wrong-site surgery was associated with safety perception as a team process was mostly supported among anesthesiologists.

| DISCUSSION
The results of this exploratory study showed that most respondents All models are adjusted with language of the questionnaire (German, English, and French), significant in univariable analysis.
b Single item "Regarding safety of care in the operating theatre, to what extent are you in agreement with the following opinion: Safety is an individual concern above all, and a team concern to a lesser extent" answers were given on a scale of 1 (do not agree at all) to 5 (fully agree), reverse coded to figure a high score synonym of team-based safety perception (low score of an individual-based safety perception). c Respondents having participated in wrong-site surgery during both periods (past and recent) are excluded.
surgery. To our knowledge, this is a novel study. Our results showed that participation in itself (yes versus no), or the number of participation in wrong-site surgery during the whole clinical practice, was not related to perception of safety in the OR. However, when splitting participation in wrong-site surgery according to time of participation (recent versus past participation), we found a differential impact: For recent participation, the effect was negatively-but not significantlyassociated with perception of safety, signifying that recent experience of wrong-site surgery favors individualistic perception of safety. In contrast, past participation (more than 3 y ago) was significantly positively associated, ie, favored a team-based perception of safety. This result was similar when treating participation in wrong-site surgery as a dichotomous variable (model 3) or a count variable (model 5).
To our knowledge, this is a novel finding and could be interpreted in terms of psychological coping processes with errors. 30 17,19,32,35,36 and a lack of support of the hospital or perceived barriers to seeking counselling. 23 In contrast, when the error was com- Potentially, one explanation could be that anesthesiologists often receive more training on teamwork for clinical safety than surgeons.
The time lag of 3 years after the error was committed used in this study was arbitrary and did not relate to any natural or psychological process occurring over time. We did not ask the date of the wrongsite surgery events, as we thought this information was too sensitive. Therefore, it was not possible to determine when the effect reversed.
We simply asked respondents to distinguish between errors that occurred within the last 3 years and prior.
Our results also showed that the number of overall participation in wrong-site surgery was not associated with perception of safety in the OR. Three possible interpretations could be put forward. First, the number of self-reported participation may be too small and the present study may have missed the association. Second, we can hypothesize that doctors that have had multiple participations may have used psychological coping strategies like medical errors acceptance, 17,37 in which doctors think that medical errors are expected and known to occur during routine care and, therefore, perceive the learning process of errors as futile. Third, we do not know if the participants who declared wrong-site surgery were directly involved in this or not (ie, responsible), as the direct responsibility of the respondent was not questioned. Possibly, most respondents reported participation in which they were not directly implicated or responsible, which could explain this lack of association.

| Limitations and strengths
Nine points have to be considered. First, the participation rate was low (22.6%), rising concerns about selection bias in the prevalence of team safety perception. Low participation rates are not uncommon when surveying doctors 38,39 and are not automatically synonymous with nonresponse bias 40 and do not systematically imply an important bias. 41,42 Second, this cross-sectional study was not designed to be representative of professions working in the OR. Consequently, its findings cannot be generalized. Third, because of the cross-sectional design of this survey, we cannot determine the causality of the association between perception of safety in the OR and self-reported participation in wrong-site surgery. It is worth noting that the reverse association-higher individualistic perceptions of safety in the OR are associated with increased number of self-reported participation in wrong-site surgery-is theoretically sound and plausible and may be examined in future studies. 43 Fourth, this study may be affected by information bias. Wrong-site surgery is a sensitive topic and many respondents may have underreported their errors. Fifth, we compiled data from 2 congresses of surgery into one sample. We replicated the multivariate models in the sample of Interlaken only and found no differences with the overall sample (data not shown), with the exception of opinion of the surgical safety checklist, which was not significantly associated with perception of safety in all 5 models. Sixth, we assessed safety culture in the OR by administering a survey in 2 surgical congresses. Thus, it limits our capacity to understand safety culture as an interprofessional concept. Moreover, past studies have shown differences in safety perceptions between physicians and nurses. [44][45][46][47] Further research will be needed to replicate this survey among nonphysicians professions working in the OR, such as scrub nurses, anesthetist nurses, auxiliary nurses, or other professionals involved in OR safety. Seventh, we cannot avoid the possibility of duplicate responses between the 2 congresses, although we consider this risk to be reasonably low because the congresses were organized by different surgical societies and in different cities. Eighth, the data collection of this study was conducted in 2010. Considering both evolving 48 and persistent 49 trends in the safety culture of acute health care institutions in high-income countries over the recent decades, the relevance of these findings need to be updated. Ninth, an analysis of convergent validity (Appendix S2) showed that the outcome, perception of safety in the OR, was weakly correlated with other items or scales related to safety. Despite suggesting weak association with similar concepts, this result is also in line with reviews suggesting that the concept of safety (safety culture and safety climate), imported from other industries, varies considerably across scales (high heterogeneity of face validity), 50 and theoretical underpinnings are often lacking. 51

| CONCLUSION
Respondents mostly had a team-based perception of safety in the OR.
Safety in the OR was perceived differently (team based or individualistic) depending on respondents having had past or recent participation in wrong-site surgery. Further research should be conducted using random and larger samples to confirm these results and to determine if and when the reversal impact of wrong-site surgery occurs on perception of safety in the OR.