The Schweitzer effect: The fundamental relationship between experience and medical students’ opinions on professional behaviours

Abstract

Purpose

We examined whether medical students’ opinions on the acceptability of a behaviour were influenced by previously encountering a similar professionally challenging situation, assessed the magnitude of effect of ‘experience’ compared to other demographic factors which influence medical students’ opinions, and evaluated whether opinions regarding some situations/behaviours were more susceptible to ‘experience’ bias?

Methods

Confidential, on-line survey for medical students distributed to Australian and New Zealand (AUS/NZ) medical schools. Students submitted de-identified demographic information, provided opinions on the acceptability of a wide range of student behaviours in professionally challenging situations, and whether they had encountered similar situations.

Results

3171 students participated from all 21 Aus/NZ medical schools (16% of registered students). Medical students reported encountering many of the professionally challenging situations, with varying opinions on what was acceptable behaviour. The most significant factor influencing acceptability towards a behaviour was whether the student reported encountering a similar situation. The professional dilemmas most significantly influenced by previous experience typically related to behaviours that students could witness in clinical environments, and often involved breaches of trust.

Conclusions

Our results demonstrate the relationship between experience and medical students’ opinions on professional behaviour- the ‘Schweitzer effect’. When students encounter poor examples of professional behaviour, especially concerning trust breaches, it significantly influences their perception of the behaviour. These results highlight the importance of placing students in healthcare settings with positive professional role modelling/work cultures.

Keywords:

• Professional behaviours
• medical student opinions
• factors influencing
• effect of experience

Example is not the main thing in influencing others; it is the only thing

Albert Schweitzer

Introduction

Medical educators have long recognised the change in identity that characterises the transformation from medical student to health professional, but it is only in the last few decades that terms such as professional identity formation, and hidden/informal curriculum, have been constructed to try to explain and define the nature and types of influences through which this occurs.

Practice points

• Medical students report encountering a wide range of professionally challenging situations, and have varying opinions on acceptable professional behaviours

• Although medical students’ opinions on professional behaviours are influenced by their demography, the most significant factor influencing the acceptability towards a behaviour was whether the student reported encountering a similar professional dilemma

• Students appear to be susceptible to normalising counter productive work behaviours, particularly those related to breaches of trust

• By placing students in toxic work culture environments with poor role models, we may inadvertently enable healthcare systems to perpetuate poor professional behaviour

Although there is a relatively large body of knowledge examining the effects of the formal, informal, and hidden curricula on medical students (Birden et al. 2013; Lawrence et al. 2018; Sarikhani et al. 2020), there have been comparatively few studies (Reddy et al. 2007; Arora et al. 2008; Dyrbye et al. 2010; Ginsburg and Lingard 2011; Kulac et al. 2013; Hendelman and Byszewski 2014; Fargen et al. 2016; Passi and Johnson 2016; Franco et al. 2017; Monrouxe and Rees 2012; Monrouxe et al. 2017a; Ainsworth and Szauter 2018; Mak-van der Vossen et al. 2018; Shaw et al. 2018) that have been designed to assess how students develop their opinions on the norms of professional behaviour. The Professionalism of Medical Students (PoMS) study was designed as a body of research to examine which factors influenced the opinions of Australian/New Zealand based medical students on a wide range of professionalism scenarios. Publications arising from the PoMS have shown that medical students’ opinions on professional behaviours are influenced by their gender, and to a lesser extent their stage in the medical course, type of medical school entry, and country of their medical school (Australia compared to New Zealand) (McGurgan et al. 2020). Medical students’ opinions on professionalism dilemmas related to patient safety appear to show a ‘professional identity effect’; students’ opinions aligned more with qualified doctors compared to the general public as they progressed through the course (McGurgan et al. 2022a). Conversely for alcohol/substance misuse dilemmas, the students’ opinions suggest generational influences rather than health care cultural norms are the dominant influence (McGurgan et al. 2022b).

The PoMS medical student surveys also revealed that students reported encountering a wide variety of the professionally challenging situations covered in the survey (Figure 1), with varying opinions on what was acceptable behaviour for many of the scenarios. A few studies have explored whether medical students previously observing and/or participating in unprofessional behaviours influences the perceived acceptability of the behaviour, however the studies which have looked at this most closely (Reddy et al. 2007; Kulac et al. 2013; Franco et al. 2017) have all been of limited size and scale. In view of this, we decided to analyse the PoMS data to examine whether medical students’ opinions on the acceptability of a behaviour were influenced by previously encountering a similar professionalism dilemma/professionally challenging situation. Secondary aims were to assess the magnitude of effect of experience compared to the other demographic factors which appear to influence medical students’ opinions, and determine whether students’ opinions on some professional behaviours are more susceptible to experience bias?

Figure 1. Ranked bar chart for proportion of medical student respondents (%) who reported encountering scenarios similar to the vignettes’ themes (with adjusted odds ratio for effect this had on modifying their opinion on the acceptability of the behaviour described).

Key: NS = Not significant, aOR = Adjusted odds ratio.

Methods

Design, ethics, setting, and participants

This study was an international, multicentre, prospective, cross-sectional survey. Ethical approval was obtained from University of Western Australia (HREC RA/4/1/8014). We used a convenience sampling approach to survey medical students in Australia and New Zealand (NZ). The methods and results of the medical student data, and validation of the survey instrument have been previously described in the Medical Students’ Opinions on Professional Behaviours paper, referred to henceforth as the PoMS-I study (McGurgan et al. 2020).

Survey instrument

The surveys used in the PoMS research covered a wide variety of professionally challenging situations pertinent to medical students (McGurgan et al. 2020). The themes included honesty, confidentiality, patient safety and respect, professional boundaries, reporting concerns, drug/alcohol misuse, social media, personal integrity, and wellbeing. The survey is included in Supplementary Appendix A; survey scenarios relating to the professionalism dilemmas/themes along with their contextual variables are described in Table 1.

Table 1. Comparison of factors that influence medical students’ responses to the PoMS scenarios.

Scenario	Vignette theme (and variable)	Previously encountered§ [Yes compared to No] aOR (95%CI), p-value	Gender [Male compared to Female] aOR, 95%CI), p-value	Seniority† p-value [Senior compared to Junior] aOR, 95%CI), p-value	Type of entry^ p-value [Postgrad compared to Undergrad] aOR, 95%CI), p-value	LocationΩ p-value [Aus compared to NZ] aOR, 95%CI), p-value
1	Fabricating results (senior student)	10.6 (2.1–53.0), p = 0.004	NS	NS	NS	NS
	Fabricating results (junior student)	NS	NS	NS	NS	NS
2	Forging attendance (student absentee)	2.0 (1.3–2.9), p = 0.001	NS	2.1 (1.4–3.1), p = 0.001	NS	NS
	Forging attendance (student forger)	NS	NS	NS	0.7 (0.5–0.9), p = 0.003	NS
3	Theft/misuse of equipment (theatre scrubs)	3.9 (2.8–5.5), p < 0.001	NS	NS	NS	NS
	Theft/misuse of equipment (suture equip)	3.6 (2.8–4.6), p < 0.001	1.3 (1.1–1.7), p = 0.014	2.0 (1.5–2.6), p < 0.001	NS	NS
4	Infective illness (student not attend)	NS	NS	NS	NS	NS
	Infective illness (student attends)	2.4 (1.7–3.5), p < 0.001	NS	2.0 (1.3–2.9), p = 0.001	0.6 (0.5–0.9), p = 0.014	0.5 (0.3–0.9), p = 0.013
5	Impersonating a doctor (student benefit)	5.7 (2.5–13.1), p < 0.001	2.4 (1.1–5.2), p = 0.031	NS	NS	NS
	Impersonating a doctor (at medical team’s instruction)	3.0 (2.3–4.0), p < 0.001	NS	0.7 (0.5–0.9), p = 0.008	0.7 (0.5–0.8), p = 0.001	NS
6	Told not to use personal protective equipment (student proceeds)	2.1 (1.5–3.0), p < 0.001	2.0 (1.5–2.7), p < 0.001	NS	NS	1.9 (1.3–2.7), 0 = 0.001
	Told not to use personal protective equipment (student refuses to proceed)	NS	2.3 (1.03–5.1), p = 0.043	NS	0.3 (0.1–0.7), p = 0.006	NS
7	Confidentiality breach (photos notes)	4.7 (3.4–6.5), p < 0.001	1.4 (1.1–1.9), p = 0.006	NS	0.6 (0.5–0.8), p = 0.001	NS
	Confidentiality breach (social media photo)	NS	NS	NS	NS	1.5 (1.1–2.2), p = 0.011
8	Student-patient relationship (female student)	NS	1.3 (1.03–1.6), p = 0.031	0.7 (0.5–0.9), p = 0.001	NS	NS
	Student-patient relationship (male student)	NS	2.1 (1.6–2.6), p < 0.001	0.7 (0.6–0.9), p = 0.002	NS	NS
9	Inappropriate social media post (obesity)	1.5 (1.1–2.1), p = 0.010	2.0 (1.5–2.7), p < 0.001	2.0 (1.5–2.7), p < 0.001	NS	NS
	Inappropriate social media post (Indigenous)	2.6 (1.7–3.9), p < 0.001	2.8 (1.9–4.1), p < 0.001	NS	NS	0.5 (0.3–0.9), p = 0.022
10	Boundary patient confidentiality (not busy)	1.7 (1.1–2.5), p = 0.016	1.8 (1.3–2.4), p < 0.001	NS	NS	NS
	Boundary patient confidentiality (busy)	NS	1.5 (1.1–2.0), p = 0.003	NS	NS	0.5 (0.3–0.7), p = 0.001
11	Drunk and disorderly (senior student)	2.1 (1.3–3.4), p = 0.002	NS	NS	NS	NS
	Drunk and disorderly (junior student)	2.2 (1.5–3.3), p < 0.001	NS	1.5 (1.1–2.2), p = 0.017	NS	NS
12	Seeks assistance after distressing event	NS	NS	NS	NS	Not estimable
13	Illicit drug use to assist with study (junior student)	2.1 (1.5–3.1), p < 0.001	2.6 (1.8–3.7), p < 0.001	NS	0.6 (0.4–0.9), p = 0.014	NS
	Illicit drug use to assist with study (senior student)	2.0 (1.4–2.9), p < 0.001	2.0 (1.4–2.8), p < 0.001	NS	NS	NS
14	Illicit drug use for anxiety (senior student)	2.0 (1.3–3.0), p = 0.001	1.5 (1.1–1.9), p = 0.004	NS	NS	NS
	Illicit drug use for anxiety (junior student)	2.5 (1.7–3.9), p < 0.001	NS	1.6 (1.2–2.2), p = 0.002	NS	NS
15	Stripper (male student)	NS	NS	0.7 (0.5–0.9), p = 0.007	NS	NS
	Lap dancer (female student)	NS	0.7 (0.6–0.9), p = 0.009	NS	NS	NS
16	Failure to report drug error (midwife)	NS	2.0 (1.3–3.2), p = 0.002	NS	NS	NS
	Failure to report drug error (junior doctor)	NS	NS	NS	NS	1.8 (1.04–3.0) p = 0.035
17	Failure to attend lecture (dietitian)	2.5 (1.9–3.4), p < 0.001	1.9 (1.5–2.5), p < 0.001	2.1 (1.6–2.8), p < 0.001	NS	NS
	Failure to attend lecture (cardiologist)	2.8 (2.0–3.8), p < 0.001	1.5 (1.1–1.9), p = 0.005	2.5 (1.8–3.4), p < 0.001	NS	NS
18	Using sexuality to gain reference (female student)	1.8 (1.1–3.0), p = 0.028	NS	NS	NS	NS
	Using sexuality to gain reference (male student)	3.9 (2.4–6.2), p < 0.001	1.9 (1.3–2.8), p < 0.001	NS	NS	NS
19	Forge clinical attendance (childcare)	2.7 (1.5–5.1), p = 0.001	NS	1.8 (1.01–3.3), p = 0.045	NS	NS
	Forge clinical attendance (social)	6.8 (2.5–18.7), p < 0.001	2.8 (1.02–7.6), p = 0.047	NS	NS	NS
20	Failure to report potential racism (practice nurse)	NS	NS	2.0 (1.5–2.5), p < 0.001	NS	NS
	Failure to report potential racism (junior doctor)	1.6 (1.2–2.3), p = 0.003	NS	1.9 (1.5–2.5), p < 0.001	NS	NS

Key: p-value calculated for respondents’ combined ‘acceptable’ opinions for the behaviour described in each vignette using logistic regression analysis to compare responses between ^§students who reported ‘Yes’ versus ‘No’ when asked if they had encountered a similar situation, adjusted for demographic characteristics. NS used to denote ‘Not statistically significant’ i.e. p ≥ 0.05 (aOR,95%CI) = Adjusted odds ratio, 95% confidence interval.

Gender defined as survey participants who identified as either ‘Male ’ or ‘Female’.

^† Senior students defined as either undergraduate entry years 4–6 or post-graduate entry years 3–4. Junior students defined as either undergraduate entry years 1–3 or post-graduate entry years 1–2.

^ Medical school entry type defined as undergraduate or post-graduate.

^Ω Location defined as Australian or New Zealand based medical school.

The scenarios were constructed to encourage participants to reflect on the acceptability of behaviours which ranged from serious professionalism breaches to positive examples of professional behaviour. All scenarios ended with the sentence ‘This student’s behaviour is … ’. Respondents gave their opinions on ‘acceptability’ using a four-point Likert scale, with no option for equipoise. This methodology has been used in other large scale national studies (Finnegan) and Gawronski’s highly cited multinomial CNI model, which quantifies sensitivity to consequences, sensitivity to moral norms, and the general preference for inaction over action in responses to moral dilemmas. Participants were also asked if they had encountered a similar real-life situation previously. Students had the opportunity to provide free text comments; analysis of these are not included in this paper.

Although the survey was anonymous, participants were asked some questions about their background, specifically gender, type of course entry (undergraduate or postgraduate), and year level in medical school. Medical students were classified as ‘junior’ if they were in the first 2 years of a postgraduate entry course or in years 1 to 3 for an undergraduate course; by default, senior students were in the latter years of these courses.

With the exception of question 12, each scenario had 2 versions (vignettes) that contained a contextual variable identified a priori using the modified Delphi approach (McGurgan et al. 2020). For example, the protagonist in the scenarios could be either a junior or senior student, or working with different health professionals. Likewise, the context of the scenario could alter from agreeing/not agreeing to perform a procedure or attending/not attending a clinical placement. The online survey (SurveyMonkey Inc., San Mateo, California) was designed such that each participant received a random version of each scenario to allow for the comparison of the contextual variables. This provided a means of analysing whether respondents were influenced by unconscious bias; that is, whether the seniority, gender, or type of health professional involved influenced the students’ opinions on the acceptability of the behaviour.

Data and statistical analysis

Likert scale responses were combined into binary categories of all ‘acceptable’ responses against all ‘unacceptable’ responses and summarized using frequency distributions. The survey design enabled the comparison of 2 types of variables: the first being the participants’ demographic characteristic- gender, medical school entry type (undergraduate versus post-graduate) and country, student background (domestic versus international), student seniority. The second type of variable was the randomized contextual variable within most of the scenarios, for example, the protagonist attending or not attending a clinical placement after a recent illness.

Due to the development and use of a novel survey instrument, the authors analysed the survey’s reliability (internal consistency) in measuring professionalism opinions. Cronbach α was calculated (α = 0.716), consistent with a high (α > 0.7) construct validity and internal consistency (Taber 2018).

The χ2 or Fisher exact test was used to compare responses for the demographic variables. The contextual variables within the scenarios, and whether the student had encountered a similar situation/dilemma were assessed using logistic regression analysis with adjustment for demographic characteristics e.g. gender, medical school entry type (undergraduate versus post-graduate) and country, student background (domestic versus international), student seniority.

Null and blank responses were excluded for all questions, as were ‘prefer not to say’ responses for gender. Adjusted odds ratios (aORs) and 95% confidence intervals (CIs) were calculated. Figures in the tables provided may not add up to 100% because of rounding (nearest decimal place), and totals may vary because of some participants not completing all of the questions or, for example, using option ‘prefer not to specify’ for gender question. SPSS Statistics 24 (IBM Corp.) was used for data analysis, and a p value <0.05 was considered statistically significant.

Results

The participation of all Australian and New Zealand medical schools resulted in a large numbers of student respondents (total 3171). Based on the Medical Student Statistics Annual Tables, this response rate represented 15.2% of enrolled Australian and 21.1% of NZ enrolled medical students (Medical Student Statistics Annual Tables 2017).

Demographic details of the student participants are presented in Table 2. There were lower numbers of responses from students in years 5 and 6 reflecting the number of courses with four-year post-graduate programs. Female students were significantly more likely to participate; 55.9% of respondents in the study were female, compared to the 51.1% of female medical students registered in the Medical Deans of Australia and New Zealand (MDANZ) database (p < 0.001). Respondents from NZ were more likely to be undergraduate entry (75.0 vs 42.4%, p < 0.001) and from more senior years in the medical course (61.1 vs 51.5%, p < 0.001) than their Australian counterparts.

Table 2. Demographic details of the PoMS medical student survey participants.^a

Demographic factor	Numbers (%)
Gender^b	Male = 1342 (42.5)			Female = 1772 (56.1)
Entry into medicine	Undergraduate = 1512 (48.2)			Postgraduate = 1624 (51.8)
Student background	Domestic = 2791 (88.3)			International = 371 (11.7)
Location of medical school	New Zealand = 569 (17.9)			Australia = 2602 (82.1)
Year in the course	Yr 1 = 622 (19.7)	Yr 2 = 586 (18.6)	Yr 3 = 659 (20.9)	Yr 4 = 665 (21.1)	Yr 5 = 376 (11.9)	Yr 6 = 248 (7.9)
Student seniority^c	Junior = 1463 (46.1)			Senior = 1667 (52.6)

Key: ^aFigures may not add up to 100 per cent due to rounding (nearest decimal place) and totals may vary due to some participants not completing all of the questions.

^b 44 students preferred not to answer for gender (1.4%) ^cSenior students defined as either undergraduate entry years 4–6 or post-graduate entry years 3–4. Junior students defined as either undergraduate entry years 1–3 or post-graduate entry years 1–2.

Survey participants were asked if they had encountered real-life situations similar to the vignettes. Figure 1 shows the frequency ranking for students who reported encountering similar situations and illustrates the association between the proportion of students who reported encountering a scenario similar to the vignettes and the effect this had on influencing their overall opinions on the acceptability of the behaviour (calculated as an adjusted odds ratio, aOR). Any scenario which had greater than 10% of survey participants reporting that they had encountered a similar situation had a significant aOR indicating that the participant responses on whether they considered the behaviour described as ‘acceptable’ were influenced by this. The professional dilemmas that were most influenced by whether the participants had encountered a similar situation (aOR >5) typically related to behaviours that students could witness in the clinical environment, and concerningly often related to breaches of trust, e.g. fabricating results, impersonating a doctor, or forging clinical attendance.

Results displayed in Table 1 demonstrate that although survey participant demographics appeared to influence some of their opinions on professionalism-related behaviours, students’ opinions on the acceptability of a behaviour were most significantly influenced by whether they reported previously encountering a similar professionalism dilemma/professionally challenging situation.

It was notable that professionalism dilemmas relating to alcohol/drug misuse (scenarios 11, 13 and 14) were the most consistently influenced by experience, i.e. whether the student respondents had encountered similar situations. In these contexts, students who stated they had previously encountered a similar situation were at least twice as likely to consider the alcohol/drug misuse situation as being ‘acceptable’. No other demographic factor (e.g. gender, student seniority) had a similarly pervasive effect on opinions towards drug/alcohol misuse (Table 1).

Respondents who stated that they had encountered situations related to a senior medical student fabricating patients’ results (scenario 1) had the largest effect on modifying opinions on acceptability; these respondents were at least ten times more likely to consider this sort of behaviour as being acceptable. Likewise, students who stated that they had encountered situations similar to the scenarios which described using social media to make derogatory comments about either obese or Indigenous people (scenario 9) were at least 50% more likely to consider this as acceptable.

For other professionalism themes, the effect on students’ opinions being influenced by whether they had encountered similar situations to the scenarios were less consistent. Survey participants who reported encountering ‘failure to report’ type situations (Scenarios 16 and 20) or looking up a spouse’s laboratory results (scenario 10) had variable effects on influencing opinions regarding whether this behaviour would be acceptable (Table 1).

Students who reported encountering similar situations to where the protagonist did not follow appropriate patient-safety related professional behaviours, such as attending a clinical placement despite a recent illness (scenario 4), or performing a minor surgical procedure without appropriate PPE (scenario 6), were at least twice as likely to consider these actions as being acceptable. Conversely, there was no evidence that students who had previously encountered similar situations to the scenarios that described the protagonist doing the ‘right thing’ had their opinions modified. For example, seeking assistance after a distressing event (scenario 12), not attending a placement when unwell, and not proceeding with a minor surgical procedure if told unable to use PPE.

Discussion

This is the first study to survey medical students from all Australian and New Zealand schools for their opinions on professionally challenging situations, and examine the factors which may influence these, specifically whether previously encountering a professionally challenging situation influences opinions on the acceptability of the behaviour. With over 3000 participants from two countries, covering a wide range of professionalism dilemmas and contextual variables, the PoMS results establish the types of professional behaviours which are most influenced by lived experience, and the magnitude of the effect (Table 1).

Regarding possible limitations, similar to other large surveys, we used a convenience sampling methodology (Dyrbye et al. 2010; Monrouxe et al. 2015; Finnegan and Gauden 2016). Although there were more female respondents (55.9% respondents), we corrected for this and other demographic characteristics by using adjusted regression analysis. The large sample size afforded a high precision; approximately 2% margin of error and 80% power to detect a difference of 3–5% between group proportions when the ‘acceptable’ opinions in the reference group were between 0.1 and 0.4 with an alpha level of 0.05.

The initial aims of PoMS study were to examine the opinions of Australian and NZ medical students on a range of professionalism scenarios, determine the relative frequency of participants encountering similar situations, and explore the role of context and demographic factors in influencing students’ opinions. As such, we cannot determine from this study whether students reporting having encountered a scenario were intrinsically different from students not encountering these. The survey was structured to minimise framing bias (Hodgkinson et al. 1999) by always asking the question ‘have you encountered a scenario like this’ after asking the respondents’ opinions on the acceptability of the protagonist’s behaviour. Analysis of the results from the two scenarios (4 and 6) which contained vignettes with diametrically opposing protagonist behaviours implies that framing effects were negligible. The effect of prior experience on participants’ opinions on acceptability was predicated on whether the behaviour was positive or negative.

The PoMs surveys used a third-party perspective to minimise respondents’ social desirability bias (Lavrakas 2008). Whilst asking participants opinions on the ‘acceptability’ of the protagonist’s behaviour using Likert scales could be considered too positivist/empirical an approach for the complexities of professional behaviours, many psychological studies of human behaviour have used and validated similar approaches (Wallace et al. 2005; Gawronski et al. 2017).

‘Acceptability’ is a fundamental, albeit amorphous principle used in a wide variety of fields from logic to philosophy (Michalos 1971; Gabbay et al. 2010; Sekhon et al. 2017). The caveat to using this criterion is that our results are limited to opinions on the acceptability of certain behaviours, and do not indicate whether the medical students would engage in these behaviours. Previous PoMS research (McGurgan et al. 2022a) found that medical students’ opinions transitioned to align more with the attitudinal norms of qualified doctors’ professional behaviours as they progressed through the course. Monrouxe et al. (2017a) characterised the different models of ethical judgment as two broad categories- the ethics of conduct and the ethics of character. Conduct ethics is influenced by people acting within rules/norms (deontological) or focussing on the outcome (consequentialism). Virtue (character) ethical decision making relies on the individual’s perception of self to set the boundaries for what they do. Greene’s dual-process theory of moral judgment (Gawronski) asserts that both emotional and cognitive processes contribute jointly to moral decision making. Survey participants could have determined the protagonists’ behaviours as being acceptable (or not) using any of the forementioned models. However whichever decision making model that may have been applied, the largest meta-analysis that has attempted to quantify the correlation between attitudes and subsequent behaviour showed a Pearson product-moment correlation coefficient r of 0.41 (CI 0.407–0.413), which is deemed relatively large by social psychology metrics (Wallace et al. 2005).

Analysis of the results of the PoMS surveys demonstrated that the most significant and pervasive factor influencing the perceived acceptability towards a behaviour was whether the student reported encountering a similar professional dilemma. The psychological literature has studied the relationships between environment, attitudes, and behaviour. According to Lewin (Lewin et al. 2019), behaviour (B) is intrinsically linked to the person (P) and their environment (E); B = f(P,E). Large meta-analyses of attitude-behaviour relationships (Wallace et al. 2005; Glasman and Albarracin 2006) demonstrate the importance of ‘experience’ with the attitude object as being pivotal. However, the relationship between medical students’ opinions on professional behaviours and whether these are influenced by previous experience has not been a major focus of medical education research to date (Reddy et al. 2007; Kulac et al. 2013; Franco et al. 2017).

Professional identity formation (PIF) can be viewed as a constructivist, social paradigm. Opinions on professional behaviours will be influenced by the individual’s level of PIF and their environment’s culture (Monson 2023). Authors such as Monrouxe and Bebeau and have demonstrated that cultures (which encompass norms, behaviours, beliefs, and customs) have a major effect on PIF at the individual, local and national level (Monrouxe and Rees 2012; Monrouxe et al. 2015; Monrouxe et al. 2017b; Bebeau and Monson 2012). To promote the formation of a ‘good physician’ Cruess et al. (2014) posit:

• that selection to medical school should target non-academic attributes such as emotional intelligence;

• the concept of PIF should be embedded in all medical education activities; ‘being’ rather than just ‘doing’;

• appreciating the importance of social enculturation into medicine, in particular how students develop their concepts towards the norms of health professional behaviour.

Regarding the latter point, the celebrated polymath doctor Albert Schweitzer stated that ‘Example is not the main thing in influencing others; it is the only thing’ (Byers 1996). Although previous PoMS and other research shows that there are other factors influencing medical students’ opinions on professionalism behaviours, this paper demonstrates that lived experience/example appears to be the most dominant factor.

Medical educators have long recognised the influences that culture and environment have on students and qualified doctors (Lesser et al. 2010; Goldie 2012; Birden et al. 2013; Passi et al. 2013; Fargen et al. 2016; Hodges et al. 2019). In terms of social enculturation to medicine, medical schools are the formative crucibles of professional identity, where students are influenced by a myriad of factors – on a spectrum from extremely positive to negative in terms of professional behaviours (Rees et al. 2013; Carmody et al. 2015; McGurgan et al. 2020). These have been grouped into the formal, informal, and hidden curricula. The hidden/informal curriculum has strong consilience with Bandura’s social cognitive theories (Bandura 1986); students are highly motivated to learn behaviours and ways of being that look successful to them.

As noted in the Introduction, previous surveys have demonstrated attitudinal relationships between medical student observation and/or participation in unprofessional behaviour (Reddy et al. 2007; Kulac et al. 2013; Franco et al. 2017); medical students were more likely to perceive unprofessional behaviour as being ‘acceptable’ if the student had either observed or participated in the activity. However, these studies were characterised by relatively small numbers of participants (<300) and used simple behaviour descriptors which lacked contextual nuances (Reddy et al. 2007; Kulac et al. 2013; Franco et al. 2017). Our results confirm the intrinsic relationship between experience and opinions regarding professional behaviours; which we term the ‘Schweitzer effect’. Compared to the other factors (gender, student seniority in the course, type of medical school entry, and country) that were analysed in the PoMS data, students reporting encountering a similar situation had the largest effect on determining how acceptable they would consider the protagonists’ behaviour.

The implications of the Schweitzer effect are considerable; our results indicate that the professional dilemmas that were most significantly influenced by whether the participants had encountered a similar situation (aOR >5) typically related to behaviours that students could witness in the clinical environment, and concerningly often related to breaches of trust, e.g. fabricating results, impersonating a doctor, or forging clinical attendance. This further underscores the importance of ensuring that students are placed in clinical environments and work with health professionals that model good standards of professional behaviours (Passi and Johnson 2016). The PoMS data also shows that medical students appear to be more heavily influenced by encountering poor examples of professional behaviours than exemplars. This was most clearly demonstrated in the two survey scenarios which were designed such that the protagonist would carry out diametrically opposite behaviours in each vignette; scenario 4 asked respondents their opinions on whether the medical student protagonist should attend/not attend a clinical placement within 24-h of experiencing gastroenteritis symptoms, and scenario 6 explored opinions on the acceptability of the student protagonist either performing/or not performing a minor surgical procedure if their supervisor states that wearing protective equipment (gloves) was unnecessary. The medical students’ opinions were notably consistent; for the survey participants responding to the vignettes which had the protagonist displaying exemplary professional behaviours, for example not attending a clinical placement when recently unwell/refusing to perform surgery without PPE, there was no effect on whether the student respondent had encountered a similar situation. However, if for the same scenario, the protagonist’s behaviour was a negative professional behaviour, for example, attending the placement/performing surgery without PPE, respondents who reported encountering a similar situation were twice as likely to consider this as being ‘acceptable’.

Much of the recent research on human decision making and behaviour has come from the discipline of behavioural economics. Kahneman, Tversky, Thaler and Sunstein’s research findings (Thaler and Sunstein 2008; Kahneman 2011; Tversky et al. 2018) generally support Haidt’s description of human decision making- the emotional tail wags the rational dog (Haidt 2001). There are increasing numbers of studies examining the unconscious bias and heuristics which influence medical decision making (Blumenthal-Barby and Krieger 2015; Klein and McColl 2019). Loewenstein used the term ‘empathy gap’ to describe the bias through which people underestimate how much their emotional states/experience influences their attitudes, preferences, and behaviours (Loewenstein 2005). Tavris and Aronson’s research and review of the literature (Tavris and Aronson 2007) demonstrate that dissonance theory has powerful implications for long-term behaviour. Once we have formed an opinion, we become susceptible to self-justification and confirmation biases; we are more likely to look out for, notice and remember anything which confirms opinions that we already hold (Laurin 2018).

Despite the obvious benefits of placing students with positive role models, working in well-functioning clinical environments, most contemporaneous data demonstrates that this is a difficult goal to achieve. The largest international study to analyse the prevalence of harassment and discrimination in medical training (Fnais et al. 2014) found that the pooled prevalence for harassment and discrimination was 59.4% (n = 51 studies, 38,353 trainees, 95% CI: 52.0%–66.7%;). The Medical Board of Australia’s 2022 annual medical training survey provides contemporaneous data from the Australian health care system (Medical Board Australia 2022). With more than 50% of doctors in training participating (n = 23,000), they found that 30/22% of respondents respectively had either witnessed/experienced bullying, discrimination, or harassment in their workplace that year. More than 30% of respondents stated that the person responsible for these behaviours was their supervisor, and the majority would not report the issue. Given this paper’s findings on how students’ opinions are unfortunately susceptible to normalising unprofessional behaviours, in particular those related to breaches of trust, e.g. fabricating results, impersonating a doctor, or forging clinical attendance, it seems likely that by placing students in the toxic work culture environments described above, we may be unwittingly enabling a system that perpetuates poor professional behaviour.

Westbrook et al.’s and others’ research on the prevalence and impacts of unprofessional behaviour in the health workplace highlight the effects this has on both the individual’s and patients’ health and safety, and acknowledges that organisational cultures can normalise incivility (Cooper et al. 2019; Medisauskaite and Kamau 2019; Westbrook et al. 2021). Brennan’s (2020) systematic review on remediating professionalism lapses in medical students and doctors concludes with the depressing statement that there is a ‘paucity of evidence to guide best practices of remediation of professionalism lapses in medical education at all levels’ (Brennan et al. 2020). However, work by Searle and Rice as part of the UK Centre for Research and Evidence on Security Threats (CREST) on counterproductive work behaviour (CWB, which by definition includes unprofessional behaviours) uses an individual (bad apple), local (bad barrel), and organisational (bad cellar) perspective. They describe a push and pull factors framework to identify factors and potential solutions for dealing with CWBs (Searle and Rice 2018; Searle and Rice 2021). The psychological literature attests to the malleability of human behaviour, and hence organisational behaviours. In order to break the cycle of normalising unprofessional behaviours in healthcare settings, it is imperative that more focus is put on using the CREST and other evidence-based strategies for managing the individual, groups, and the organisation. This paper indicates that medical students encounter a wide variety of professionalism dilemmas and are influenced by them. It seems unlikely that we can fix poor professionalism in the healthcare settings if we continue to expose students to it, given their susceptibility to habituate and normalise counterproductive work behaviours. The ideal paradigm would be that students are not placed in toxic work culture environments. Acknowledging the realpolitik, we propose that those involved in medical education and clinical supervision should reconsider what tools and resources students are provided with to engage them in developing appropriate ethical reasoning and professional behaviours, and equip them to deal with the significant professionalism dilemmas they may encounter in healthcare settings.

Conclusions

This international survey’s results and analysis demonstrate that the optimistic assumption that medical students can be placed in clinical environments which have poor professionalism cultures but will manage to process these experiences in a positive way is fundamentally flawed. Instead, what we have termed the ‘Schweitzer effect’ explains that when students encounter poor examples of professional behaviour, particularly those related to breaching trust, it has a profound effect on how they come to perceive the nature of the behaviour, such that they are five times more likely to normalise these behaviours.

We hope our findings will inform those interested in medical education and/or professionalism of the factors influencing students’ opinions on professionalism behaviours, provide evidence for the importance of placing students in healthcare settings with positive professional role modelling and work cultures, and the need to adequately prepare students to identify and manage the complex group dynamics and CWB they will encounter. We need educational approaches and guidance that are much more sophisticated, and that engage students in strengthening their own ethical reasoning and moral standards, rather than expecting them to learn a theoretical best practice which may not be reflected by their supervisors or role models.

The fact that lived experience has such a significant influence on how medical students view professionalism dilemmas both challenges and provides a key mechanism towards improving the professional culture in healthcare environments.

Acknowledgments

The authors wish to thank the 3171 medical students who participated in the survey, and all of the Aus/NZ medical schools who agreed to distribute the surveys to their students. Dr Laura Fruhen, Senior Lecturer of Applied Psychology, UWA provided some advice on the manuscript.

Disclosure statement

The authors report there are no competing or conflicts of interest to declare. The authors alone are responsible for the content and writing of the article.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.

Notes on contributors

Paul McGurgan

Paul McGurgan, FRCOG, FRANZCOG, SFHEA, is a clinical academic. He is the Associate Dean of UWA Medical School (Student Matters), and theme lead for Personal and Professional Development in the UWA MD course. He chairs the Professional Behaviour Advisory Panel for UWA Faculty of Health Sciences. The PoMs research forms part of his PhD thesis on ‘What factors influence opinions about medical student professional behaviours?’

Katrina Calvert

Katrina Calvert, FRANZCOG, MHPEd, MRCOG, is a consultant obstetrician and director of post-graduate medical education in the state’s largest obstetric and gynaecology hospital, KEMH. Katrina has a special interest in simulation, feedback, and health professional wellbeing. She was an inaugural member, and deputy chair of the RANZCOG Wellbeing Working Group.

Antonio Celenza

Antonio Celenza, MClinEd, FACEM, FRCEM, is an academic emergency physician and was director of the MD program at the University of Western Australia for 8 years with leadership roles in curriculum design, teaching delivery, assessment, and evaluation. He continues to have clinical, educational and research roles as Professor and Head, Division of Emergency Medicine at UWA.

Elizabeth A. Nathan

Elizabeth A. Nathan, RN, B.Sci, M.Biostats (Syd), is a biostatistician and Adjunct Senior Lecturer at UWA Division of Obstetrics and Gynaecology, based at King Edward Memorial Hospital. She has collaborated on many health care research projects applying a broad range of statistical methodologies.

Christine Jorm

Christine Jorm, MBBS(Hons) MD PhD FANZCA, is currently Adjunct Professor, School of Public Health, Sydney University. Christine’s PhD thesis explored why medical (doctor) culture resisted the imperatives of the safety and quality movement. Christine has broad experience as a clinician, policy maker, researcher and medical educator.