Attrition during surgical training is a critical issue, as the loss of a training registrar represents a waste of resources and effort as well as having implications for future workforce planning. Workforce modelling has predicted that unless there is a substantial increase in the number of graduating surgeons, Australia will face a surgical workforce crisis within the next 15 years. It has been conservatively estimated that at least 264 new surgeons will be needed each year between now and 2025, which equates to at least an additional 80 per annum in addition to the current 184 already graduating.1 It is therefore critical to ensure that we retain as many surgical trainees and fellows as possible, and prevent attrition.The literature regarding the rate of attrition from surgical training is limited, and there have been no studies in Australasia concerning either the incidence of attrition or the reasons contributing to it. Attrition is defined as a decrease in the surgical workforce due to all causes, including voluntary withdrawal, transfer to other programmes or dismissal. In the US, several studies have reported trainee attrition rates (including all trainees who have left or been dismissed for any reason), with fairly consistent figures of 17-30%.2-7 In Australia and New Zealand, the withdrawal rate for general surgical trainees during the 5 years from 2010-2014 was 5.9% (with 5.8% for male trainees, and 6.2% for female trainees) due to personal reasons or transfer to another specialty.8 These figures do not include trainees who were dismissed from training.Few studies have considered the factors contributing to a trainees decision to leave. It is perceived that attrition is the result of the highly demanding and stressful climate that trainees and their families endure in the postgraduate years. It has been demonstrated that dissatisfied physicians are 2 to 3 times more likely to leave medicine than those that are satisfied.9 A 2005 study by Dodson et al2 considered reasons for attrition and found that 65% of residents who left their programme withdrew from training for life-style reasons (other reasons included opportunity for early specialisation, emotional issues or performance difficulties and desire to leave medicine altogether).It has been reported previously that gender may play a role in attrition. This is certainly an important consideration in the context of increasing feminisation of the surgical workforce, with 33% of current general surgical trainees in Australasia being female.8 In the US, Bergen et al10 reported that women were 2.26 times more likely to voluntarily withdraw from general surgical training than men, and this was predominantly due to family commitments or health reasons. Similarly, the study by Dodson et al reported attrition rates for females as being double that for males (27% vs 13%). However, a comprehensive study of 6,303 US trainees by Yeo et al found no gender difference in attrition rates.3 This same study found that the only independent predictor of attrition was post-graduate year level, with the highest rates in the first and second years of specialty training.The aim of this study was to examine both the satisfaction and attitudes of Australasian general surgical trainees toward their training programme and to assess the number and characteristics of trainees at risk of withdrawal.MethodsEthics approval for the project was obtained from the ethics committee of the Sydney Local Health District and approval to distribute the survey was obtained from General Surgeons Australia (GSA) and the New Zealand Association of General Surgeons (NZAGS). An invitation to participate in a survey was extended to all Australian and New Zealand General Surgical trainees. Trainees were emailed a link by the Royal Australasian College of Surgeons to an online survey delivered via the Survey Monkey website. The survey comprised socio-demographic questions and a 44-item questionnaire implementing a 5-point Likert scale response (modified version of survey developed by Yeo et al at the Yale University School of medicine11). Trainees were given 8 weeks to respond with a reminder letter e-mailed four weeks following initial invitation. The voluntary responses were submitted online via Survey Monkey.OutcomesThe primary outcome was the prevalence of risk of withdrawal from general surgical training. This was established by determining how many participants responded in the affirmative to the question, Have you considered leaving surgical training? The secondary outcomes were to investigate reasons for leaving surgical training and to explore satisfaction and attitudes towards surgical training. Firstly, respondents who identified themselves as being at risk of withdrawal were asked whether or not 12 factors influenced their decision to leave surgical training (n=77 respondents identified themselves as being as at risk of attrition, see Table 2). These 12 factors were presented as a list of common concerns formulated by the authors based on previous experience with surgical trainees who had left or thought about leaving surgical training. Secondly, all respondents were asked to rate 44 statements regarding attitudes towards surgical training on a five-point Likert scale (strongly agree-strongly disagree). Analysis was performed to determine statistically significant differences between at risk and not at risk groups (a difference was found for 21 of the 44 statements). These answers are presented in Table 3, stratified by whether or not trainees identified themselves at risk of withdrawal from training.Statistical analysisStatistical analysis was conducted using SPSS 20.0 (IBM, US). The level of significance for all tests was p<0.05. Differences in outcomes between categorical groups were analysed using Chi-square test or Fishers exact test if cell counts were <5.Table 1: Characteristics of survey respondents Frequency (%) Participant Characteristics Total\u2020 Considered leaving training\u2021 No Yes Age 25-29 31 (22) 15 (48) 16 (52) 30-34 70 (50) 31 (44) 39 (56) 35-40 40 (28) 18 (45) 22 (55) Gender Male 79 (56) 42 (53) 37 (47) Female 63 (44) 22 (36) 40 (65) Year of SET 1 40 (29) 19 (48) 21 (52) 2 31 (22) 14 (45) 17 (55) 3 24 (17) 16 (67) 8 (33) 4 19 (14) 5 (26) 14 (8) 5 26 (18) 10 (39) 16 (62) Medical degree completed Australia/NZ 115 (81) 47 (41) 68 (59) Overseas 27 (19) 17 (65) 9 (35) State/Country of SET training NSW 56 (40) 25 (45) 31 (55) VIC 35 (25) 14 (40) 21 (60) QLD 16 (11) 6 (38) 10 (63) WA 10 (7) 5 (50) 5 (50) SA 9 (6) 5 (56) 4 (44) NZ 15 (11) 9 (60) 6 (40) Marital Status Single 41 (29) 15 (37) 26 (63) Married 73 (52) 38 (52) 35 (48) De Facto 24 (17) 10 (42) 14 (58) Divorced/separated 3 (2) 1 (33) 2 (67) Children Yes 54 (39) 29 (54) 25 (46) No 86 (61) 35 (41) 50 (46) \u2020 Column %, \u2021 Row %, SET - Surgical Education and Training ProgrammeResultsA total of 142 survey responses were received of 550 general surgical registrars in Australia and New Zealand (26% response rate) in 2013. Of the respondents, 56% were male and 44% female, with females relatively over represented, given that in 2013 only 35% of general surgical trainees were female. Half the responses received were from trainees in their first or second of five years of training. The socio-demographic characteristics of survey participants are summarised in Table 1.Overall, 54% (n=77) of respondents had considered leaving surgical training, with 43% (n=61) considering this within the previous year. Females were more than twice as likely to consider leaving surgical training compared to males; OR=2.1(95% CI 1.0-4.1), p=0.036 (Table 1). Respondents who studied in Australia or New Zealand were also significantly more likely to report considering leaving surgical training compared to those who studied in another country (p=0.023, OR=2.7 [95% CI 1.1-6.7]); 59% vs 35% respectively. Other factors\u2014including age, gender, year of training, location of training, marital status and whether respondents had children\u2014were not found to be significantly associated with attrition risk (Table 1).Reasons for considering leaving surgical training and the response frequency are listed in Table 2. The most common reason for considering leaving was poor lifestyle and quality of life during surgical training (47% respondents). The next most cited reason was a perceived lack of support, either at work or home (37%), and this was followed by a concern over excessive working hours (35%). Males were significantly more likely to consider leaving surgical training because of poor lifestyle and quality of life as a consultant compared to females (p=0.001). There were no other significant differences between genders in reasons why respondents had considered leaving surgical training.Table 2: Reasons for leaving surgical training (n=81 respondents) Reason Frequency. (%) of Respondents agreeing or strongly agreeing Poor lifestyle and quality of life during training 38 (47) Lack of support (home or work) 30 (37) Excessive working hours 28 (35) Job dissatisfaction 27 (33) High levels stress/anxiety/pressure 26 (32) Domestic or social reasons/family commitments 19 (24) Excessive workload/work intensity 18 (22) Poor lifestyle and quality of life as a consultant 15 (19) Desire to travel 10 (12) Financial/lack of job security 3 (4) Lack of flexible training/work opportunities 3 (4) To pursue training in another field/specialty 3 (4) Table 3 shows differences in satisfaction and attitudes towards surgical training between trainees at risk of withdrawal and those not at risk. Respondents who identified themselves at risk of withdrawal from training were significantly less likely: to be satisfied with their training; to think their opinions were important; to agree that the programme had support structures in place; to feel like they could turn to senior colleagues; to look forward to going to work; to be happy at work; to feel their operating skills were appropriate; to feel they fitted well into their training programme; and to feel they could count on other registrars to help them. Respondents who reported risk of withdrawal were also significantly more likely to: worry they were not confident enough; agree that the hours they were working were causing strain on their personal life; agree that the personal cost of training was not worth it; and agree that surgeons do not make as much money as they used to.Table 3: Satisfaction and attitudes towards surgical training (n=142 respondents) Attitudes Those that agree or strongly agree n* (%) Total (n=142) Thought of leaving surgical training P value between groups \u2020 No (n=64) Yes (n=77) I am satisfied with my training programme 74 (52) 40 (63) 33 (43) 0.020 As a surgical registrar, my opinions are important 90 (63) 51 (80) 38 (49) <0.001 My training programme has support structures in place which provide me with someone to turn to when I am struggling 51 (36) 34 (53) 16 (21) <0.001 I feel I can turn to my consultants and senior colleagues when I have difficulties in my training 69 (49) 39 (62) 29 (38) 0.005 I look forward to coming to work every day 73 (51) 41 (64) 31 (40) 0.005 I am satisfied with the teaching in my training programmeme 53 (37) 35 (55) 17 (22) <0.001 I am satisfied with the operative experience in my training programme 63 (44) 37 (58) 26 (34) 0.004 I have considered leaving my training programme in the last year 61 (43) 5 (8) 56 (73) <0.001 I am happy when I am at work 89 (63) 48 (75) 41 (54) 0.010 I often feel that I am in over my head 20 (14) 5 (8) 15 (20) 0.045 I feel that my operating skill is level appropriate 71 (50) 41 (64) 30 (39) 0.003 I worry that I will not feel confident enough to perform procedures by myself before I finish training 89 (63) 32 (51) 57 (74) 0.004 The hours I am working are causing a strain on my personal and family life 88 (62) 32 (51) 55 (71) 0.012 My consultants will think worse of me if I ask for help when I do not know how to do a procedure 35 (25) 11 (17) 24 (32) 0.045 I really care about my patients 133 (94) 61 (95) 71 (92) 0.453 The personal cost of surgical training is not worth it to me 26 (18) 3 (5) 23 (30) <0.001 I feel that I fit in well in my training programme 88 (63) 49 (78) 38 (50) 0.001 I am committed to completing my general surgical training 123 (87) 61 (95) 61 (79) 0.005 My operative experience so far has helped me develop my skills well 92 (65) 52 (81) 39 (51) < 0.001 If I have a problem, I feel I can count on other registrars to help me out 88 (62) 46 (72) 41(53) 0.023 Surgeons do not make as much money now as they used to 52 (37) 17 (27) 34 (44) 0.030 * Where n \u2260 total data is missing \u2020 Only significantly different results included in this table DiscussionThis study, the first of its kind in Australasia, found that 54% (n=77) of respondents had considered leaving their surgical training and demonstrates that thoughts of discontinuing surgical training are prevalent amongst trainees. While this study captured responses from only 26% of the cohort of general surgical trainees, this number still represents a high absolute number of trainees who had considered leaving. The available data on attrition in Australian general surgical trainees shows a current attrition rate of 5.9%, and whilst this is much lower than the potential attrition rate of trainees, it remains concerning that 1 in every 17 general surgical registrars will not complete their training programme.The most common reason given when considering withdrawal is poor lifestyle and quality of life during surgical training. This is in agreement with previous American studies by Morris et al12 and Kelz et al,13 where poor lifestyle was given as the most common reason for voluntary resignation form surgical training. Interestingly, in the US, the introduction of restricted working hours has not been associated with any improvement in the wellbeing of surgical residents.14We found that female trainees were significantly more likely to consider leaving, and this is in agreement with previous findings in the literature.2,10,15 The reasons women are more likely to withdraw are multi-factorial, but include pregnancy and childcare responsibilities. There are currently steadily increasing numbers of female doctors entering surgical training, and it is imperative that these trainees are retained as they represent a valuable resource and ever increasing percentage of the surgical workforce. Interestingly, despite female trainees being at higher risk of withdrawal, the reasons that trainees chose to leave were not significantly different between genders other than more men attributing poor lifestyle as a consultant as a factor in potential attrition.This study found that locally-educated registrars (those completing their medical education in Australia or New Zealand) were at a significantly higher risk of attrition than their counterparts who underwent medical training overseas. The reasons for this trend are not entirely clear, but may be related to different expectations and perceptions, as well as the fact that international trainees have often chosen to move to Australasia (particularly from more disadvantaged nations) and therefore may be more committed to completing training and more willing to accept the challenges that training presents. They have usually made a significant financial sacrifice to go through the Royal Australasian College of Surgeons International Medical Graduate pathway, and this, coupled with the associated sequelae, is likely a significant driving force against consideration of withdrawal.In addition to establishing withdrawal risk, we also questioned attitudes toward and satisfaction with training and significant differences were identified between trainees at risk of withdrawal and those who had not considered it (Table 3). Overall, we found that 52% of trainees agreed or strongly agreed that they were satisfied with their training programme. This is significantly less than the 85.2% of US general surgical trainees who felt satisfied with their training in the survey by Yeo et al.9 We do, however, acknowledge that our sample may be inherently biased by the fact that a relatively high proportion of dissatisfied trainees or those at withdrawal risk may have responded to the survey.While it is beyond the scope of this paper to consider every difference in detail, there were broad themes of dissatisfaction that emerged amongst the cohort of at risk trainees. Registrars contemplating withdrawal felt unsupported and undervalued by their colleagues and seniors, with only 21% of at risk trainees vs 53% of other trainees agreeing or strongly agreeing their programme had structures in place to provide them with someone to turn to when they felt they were struggling (p<0.001). This is in comparison to 71.6% of US residents in the survey from Yeo et al. Overall, only 22% of at risk trainees felt satisfied with teaching in their training programme, compared to 55% of remaining respondents (p<0.001). Trainees who identified themselves at risk of leaving also felt that their operating skills were not appropriate for their training level, and 74% were worried they would not feel confident performing procedures independently before they complete their training programme. Similar studies in the US showed higher rates of confidence in operating skills, with a recent study by Fonseca et al of 653 final year US general surgical trainees reporting that approximately 25% of respondents had a significant lack of confidence in performing a variety of open surgeries16.While the largest American study by Yeo et al found year of training to be significantly associated with attrition risk (most withdrawals occurring in the first or second year), this was not the case in our study. To the contrary, we found the highest percentages of trainees considering withdrawal were in their penultimate or final training year (74% and 61% respectively), although this was not statistically significant. We postulate that this may be due to the very high levels of stress and sacrifice related to study and completion of the general surgical fellowship exam at the beginning of the final year of training.This study has several limitations, most of which are due to the nature of survey-based research. Our response rate was low (26%) and we acknowledge that we only have received data from a minority of trainees. Whilst the rate of withdrawal risk in this study population was found to be 54%, the true rate amongst trainees may be much higher or lower. It is also possible that there was a selection bias in the type of trainees who replied, with those who are inherently less satisfied or struggling being overrepresented. Our sample also had a higher rate of female trainees than the overall training cohort, and as female trainees are at higher risk of withdrawal this may have had an influence on the high attrition risk rate. Despite these limitations, this study is the first to consider this issue in Australian and New Zealand general surgery trainees, and we hope the results will be used as impetus for further research in this area.It would be of great value to survey trainees who have actually withdrawn to elucidate the reasons they did so, as this would be helpful in further determining the factors that place individuals at attrition risk. In addition, it would be interesting to know what paths these trainees took following leaving surgery\u2014whether they transferred to another specialty, took a position in a non-clinical field or left medicine altogether. If a negative experience in surgical training is resulting in abandonment of any medical career, this would be a cause of significant concern. There is clearly potential for further research in this field, and results will be of great use in terms of workforce planning and development, surgical educational reform and in the broader context of gender and generational cultural shifts in medicine.ConclusionAttrition in surgical trainees presents a significant concern for all those involved in surgical education and training as well as workforce planning. The current 5 year training programme entails commitment and sacrifice on behalf of the trainee, and there is no doubt that the stressors and difficulties encountered during these years will leave some trainees questioning whether they can complete their surgical education. This study is the first of its kind in Australasia, and has allowed estimation of withdrawal risk compared to actual attrition rate and the factors contributing to it. Whilst this study may have been affected by a responder bias, with dissatisfied trainees being more likely to reply, it has identified that female trainees are significantly more at risk of withdrawal than their male counterparts and that thoughts of attrition are prevalent amongst a proportion of trainees. Overall, the majority of trainees we surveyed were satisfied, but problems relating to lack of support, perceived poor teaching, lack of appropriate operative skills and excessive working hours were expressed by all trainees, regardless of whether or not they had considered withdrawal from training. Pleasingly, despite these issues, the vast majority of trainees felt committed to completing their training. From here, we must focus on identifying the group of at risk trainees, particular female locally-trained registrars, and start considering how the modifiable risk factors for attrition can be addressed. We need to listen to the concerns of current general surgical registrars and find ways to improve the quality of their training to ensure we have not only adequate numbers for our future surgical workforce, but a cohort of surgeons who have enjoyed their training experience and will strive to create a similar one for their own trainees in the future.

Attrition during surgical training is a critical issue, as the loss of a training registrar represents a waste of resources and effort as well as having implications for future workforce planning. Workforce modelling has predicted that unless there is a substantial increase in the number of graduating surgeons, Australia will face a surgical workforce crisis within the next 15 years. It has been conservatively estimated that at least 264 new surgeons will be needed each year between now and 2025, which equates to at least an additional 80 per annum in addition to the current 184 already graduating.1 It is therefore critical to ensure that we retain as many surgical trainees and fellows as possible, and prevent attrition.The literature regarding the rate of attrition from surgical training is limited, and there have been no studies in Australasia concerning either the incidence of attrition or the reasons contributing to it. Attrition is defined as a decrease in the surgical workforce due to all causes, including voluntary withdrawal, transfer to other programmes or dismissal. In the US, several studies have reported trainee attrition rates (including all trainees who have left or been dismissed for any reason), with fairly consistent figures of 17-30%.2-7 In Australia and New Zealand, the withdrawal rate for general surgical trainees during the 5 years from 2010-2014 was 5.9% (with 5.8% for male trainees, and 6.2% for female trainees) due to personal reasons or transfer to another specialty.8 These figures do not include trainees who were dismissed from training.Few studies have considered the factors contributing to a trainees decision to leave. It is perceived that attrition is the result of the highly demanding and stressful climate that trainees and their families endure in the postgraduate years. It has been demonstrated that dissatisfied physicians are 2 to 3 times more likely to leave medicine than those that are satisfied.9 A 2005 study by Dodson et al2 considered reasons for attrition and found that 65% of residents who left their programme withdrew from training for life-style reasons (other reasons included opportunity for early specialisation, emotional issues or performance difficulties and desire to leave medicine altogether).It has been reported previously that gender may play a role in attrition. This is certainly an important consideration in the context of increasing feminisation of the surgical workforce, with 33% of current general surgical trainees in Australasia being female.8 In the US, Bergen et al10 reported that women were 2.26 times more likely to voluntarily withdraw from general surgical training than men, and this was predominantly due to family commitments or health reasons. Similarly, the study by Dodson et al reported attrition rates for females as being double that for males (27% vs 13%). However, a comprehensive study of 6,303 US trainees by Yeo et al found no gender difference in attrition rates.3 This same study found that the only independent predictor of attrition was post-graduate year level, with the highest rates in the first and second years of specialty training.The aim of this study was to examine both the satisfaction and attitudes of Australasian general surgical trainees toward their training programme and to assess the number and characteristics of trainees at risk of withdrawal.MethodsEthics approval for the project was obtained from the ethics committee of the Sydney Local Health District and approval to distribute the survey was obtained from General Surgeons Australia (GSA) and the New Zealand Association of General Surgeons (NZAGS). An invitation to participate in a survey was extended to all Australian and New Zealand General Surgical trainees. Trainees were emailed a link by the Royal Australasian College of Surgeons to an online survey delivered via the Survey Monkey website. The survey comprised socio-demographic questions and a 44-item questionnaire implementing a 5-point Likert scale response (modified version of survey developed by Yeo et al at the Yale University School of medicine11). Trainees were given 8 weeks to respond with a reminder letter e-mailed four weeks following initial invitation. The voluntary responses were submitted online via Survey Monkey.OutcomesThe primary outcome was the prevalence of risk of withdrawal from general surgical training. This was established by determining how many participants responded in the affirmative to the question, Have you considered leaving surgical training? The secondary outcomes were to investigate reasons for leaving surgical training and to explore satisfaction and attitudes towards surgical training. Firstly, respondents who identified themselves as being at risk of withdrawal were asked whether or not 12 factors influenced their decision to leave surgical training (n=77 respondents identified themselves as being as at risk of attrition, see Table 2). These 12 factors were presented as a list of common concerns formulated by the authors based on previous experience with surgical trainees who had left or thought about leaving surgical training. Secondly, all respondents were asked to rate 44 statements regarding attitudes towards surgical training on a five-point Likert scale (strongly agree-strongly disagree). Analysis was performed to determine statistically significant differences between at risk and not at risk groups (a difference was found for 21 of the 44 statements). These answers are presented in Table 3, stratified by whether or not trainees identified themselves at risk of withdrawal from training.Statistical analysisStatistical analysis was conducted using SPSS 20.0 (IBM, US). The level of significance for all tests was p<0.05. Differences in outcomes between categorical groups were analysed using Chi-square test or Fishers exact test if cell counts were <5.Table 1: Characteristics of survey respondents Frequency (%) Participant Characteristics Total\u2020 Considered leaving training\u2021 No Yes Age 25-29 31 (22) 15 (48) 16 (52) 30-34 70 (50) 31 (44) 39 (56) 35-40 40 (28) 18 (45) 22 (55) Gender Male 79 (56) 42 (53) 37 (47) Female 63 (44) 22 (36) 40 (65) Year of SET 1 40 (29) 19 (48) 21 (52) 2 31 (22) 14 (45) 17 (55) 3 24 (17) 16 (67) 8 (33) 4 19 (14) 5 (26) 14 (8) 5 26 (18) 10 (39) 16 (62) Medical degree completed Australia/NZ 115 (81) 47 (41) 68 (59) Overseas 27 (19) 17 (65) 9 (35) State/Country of SET training NSW 56 (40) 25 (45) 31 (55) VIC 35 (25) 14 (40) 21 (60) QLD 16 (11) 6 (38) 10 (63) WA 10 (7) 5 (50) 5 (50) SA 9 (6) 5 (56) 4 (44) NZ 15 (11) 9 (60) 6 (40) Marital Status Single 41 (29) 15 (37) 26 (63) Married 73 (52) 38 (52) 35 (48) De Facto 24 (17) 10 (42) 14 (58) Divorced/separated 3 (2) 1 (33) 2 (67) Children Yes 54 (39) 29 (54) 25 (46) No 86 (61) 35 (41) 50 (46) \u2020 Column %, \u2021 Row %, SET - Surgical Education and Training ProgrammeResultsA total of 142 survey responses were received of 550 general surgical registrars in Australia and New Zealand (26% response rate) in 2013. Of the respondents, 56% were male and 44% female, with females relatively over represented, given that in 2013 only 35% of general surgical trainees were female. Half the responses received were from trainees in their first or second of five years of training. The socio-demographic characteristics of survey participants are summarised in Table 1.Overall, 54% (n=77) of respondents had considered leaving surgical training, with 43% (n=61) considering this within the previous year. Females were more than twice as likely to consider leaving surgical training compared to males; OR=2.1(95% CI 1.0-4.1), p=0.036 (Table 1). Respondents who studied in Australia or New Zealand were also significantly more likely to report considering leaving surgical training compared to those who studied in another country (p=0.023, OR=2.7 [95% CI 1.1-6.7]); 59% vs 35% respectively. Other factors\u2014including age, gender, year of training, location of training, marital status and whether respondents had children\u2014were not found to be significantly associated with attrition risk (Table 1).Reasons for considering leaving surgical training and the response frequency are listed in Table 2. The most common reason for considering leaving was poor lifestyle and quality of life during surgical training (47% respondents). The next most cited reason was a perceived lack of support, either at work or home (37%), and this was followed by a concern over excessive working hours (35%). Males were significantly more likely to consider leaving surgical training because of poor lifestyle and quality of life as a consultant compared to females (p=0.001). There were no other significant differences between genders in reasons why respondents had considered leaving surgical training.Table 2: Reasons for leaving surgical training (n=81 respondents) Reason Frequency. (%) of Respondents agreeing or strongly agreeing Poor lifestyle and quality of life during training 38 (47) Lack of support (home or work) 30 (37) Excessive working hours 28 (35) Job dissatisfaction 27 (33) High levels stress/anxiety/pressure 26 (32) Domestic or social reasons/family commitments 19 (24) Excessive workload/work intensity 18 (22) Poor lifestyle and quality of life as a consultant 15 (19) Desire to travel 10 (12) Financial/lack of job security 3 (4) Lack of flexible training/work opportunities 3 (4) To pursue training in another field/specialty 3 (4) Table 3 shows differences in satisfaction and attitudes towards surgical training between trainees at risk of withdrawal and those not at risk. Respondents who identified themselves at risk of withdrawal from training were significantly less likely: to be satisfied with their training; to think their opinions were important; to agree that the programme had support structures in place; to feel like they could turn to senior colleagues; to look forward to going to work; to be happy at work; to feel their operating skills were appropriate; to feel they fitted well into their training programme; and to feel they could count on other registrars to help them. Respondents who reported risk of withdrawal were also significantly more likely to: worry they were not confident enough; agree that the hours they were working were causing strain on their personal life; agree that the personal cost of training was not worth it; and agree that surgeons do not make as much money as they used to.Table 3: Satisfaction and attitudes towards surgical training (n=142 respondents) Attitudes Those that agree or strongly agree n* (%) Total (n=142) Thought of leaving surgical training P value between groups \u2020 No (n=64) Yes (n=77) I am satisfied with my training programme 74 (52) 40 (63) 33 (43) 0.020 As a surgical registrar, my opinions are important 90 (63) 51 (80) 38 (49) <0.001 My training programme has support structures in place which provide me with someone to turn to when I am struggling 51 (36) 34 (53) 16 (21) <0.001 I feel I can turn to my consultants and senior colleagues when I have difficulties in my training 69 (49) 39 (62) 29 (38) 0.005 I look forward to coming to work every day 73 (51) 41 (64) 31 (40) 0.005 I am satisfied with the teaching in my training programmeme 53 (37) 35 (55) 17 (22) <0.001 I am satisfied with the operative experience in my training programme 63 (44) 37 (58) 26 (34) 0.004 I have considered leaving my training programme in the last year 61 (43) 5 (8) 56 (73) <0.001 I am happy when I am at work 89 (63) 48 (75) 41 (54) 0.010 I often feel that I am in over my head 20 (14) 5 (8) 15 (20) 0.045 I feel that my operating skill is level appropriate 71 (50) 41 (64) 30 (39) 0.003 I worry that I will not feel confident enough to perform procedures by myself before I finish training 89 (63) 32 (51) 57 (74) 0.004 The hours I am working are causing a strain on my personal and family life 88 (62) 32 (51) 55 (71) 0.012 My consultants will think worse of me if I ask for help when I do not know how to do a procedure 35 (25) 11 (17) 24 (32) 0.045 I really care about my patients 133 (94) 61 (95) 71 (92) 0.453 The personal cost of surgical training is not worth it to me 26 (18) 3 (5) 23 (30) <0.001 I feel that I fit in well in my training programme 88 (63) 49 (78) 38 (50) 0.001 I am committed to completing my general surgical training 123 (87) 61 (95) 61 (79) 0.005 My operative experience so far has helped me develop my skills well 92 (65) 52 (81) 39 (51) < 0.001 If I have a problem, I feel I can count on other registrars to help me out 88 (62) 46 (72) 41(53) 0.023 Surgeons do not make as much money now as they used to 52 (37) 17 (27) 34 (44) 0.030 * Where n \u2260 total data is missing \u2020 Only significantly different results included in this table DiscussionThis study, the first of its kind in Australasia, found that 54% (n=77) of respondents had considered leaving their surgical training and demonstrates that thoughts of discontinuing surgical training are prevalent amongst trainees. While this study captured responses from only 26% of the cohort of general surgical trainees, this number still represents a high absolute number of trainees who had considered leaving. The available data on attrition in Australian general surgical trainees shows a current attrition rate of 5.9%, and whilst this is much lower than the potential attrition rate of trainees, it remains concerning that 1 in every 17 general surgical registrars will not complete their training programme.The most common reason given when considering withdrawal is poor lifestyle and quality of life during surgical training. This is in agreement with previous American studies by Morris et al12 and Kelz et al,13 where poor lifestyle was given as the most common reason for voluntary resignation form surgical training. Interestingly, in the US, the introduction of restricted working hours has not been associated with any improvement in the wellbeing of surgical residents.14We found that female trainees were significantly more likely to consider leaving, and this is in agreement with previous findings in the literature.2,10,15 The reasons women are more likely to withdraw are multi-factorial, but include pregnancy and childcare responsibilities. There are currently steadily increasing numbers of female doctors entering surgical training, and it is imperative that these trainees are retained as they represent a valuable resource and ever increasing percentage of the surgical workforce. Interestingly, despite female trainees being at higher risk of withdrawal, the reasons that trainees chose to leave were not significantly different between genders other than more men attributing poor lifestyle as a consultant as a factor in potential attrition.This study found that locally-educated registrars (those completing their medical education in Australia or New Zealand) were at a significantly higher risk of attrition than their counterparts who underwent medical training overseas. The reasons for this trend are not entirely clear, but may be related to different expectations and perceptions, as well as the fact that international trainees have often chosen to move to Australasia (particularly from more disadvantaged nations) and therefore may be more committed to completing training and more willing to accept the challenges that training presents. They have usually made a significant financial sacrifice to go through the Royal Australasian College of Surgeons International Medical Graduate pathway, and this, coupled with the associated sequelae, is likely a significant driving force against consideration of withdrawal.In addition to establishing withdrawal risk, we also questioned attitudes toward and satisfaction with training and significant differences were identified between trainees at risk of withdrawal and those who had not considered it (Table 3). Overall, we found that 52% of trainees agreed or strongly agreed that they were satisfied with their training programme. This is significantly less than the 85.2% of US general surgical trainees who felt satisfied with their training in the survey by Yeo et al.9 We do, however, acknowledge that our sample may be inherently biased by the fact that a relatively high proportion of dissatisfied trainees or those at withdrawal risk may have responded to the survey.While it is beyond the scope of this paper to consider every difference in detail, there were broad themes of dissatisfaction that emerged amongst the cohort of at risk trainees. Registrars contemplating withdrawal felt unsupported and undervalued by their colleagues and seniors, with only 21% of at risk trainees vs 53% of other trainees agreeing or strongly agreeing their programme had structures in place to provide them with someone to turn to when they felt they were struggling (p<0.001). This is in comparison to 71.6% of US residents in the survey from Yeo et al. Overall, only 22% of at risk trainees felt satisfied with teaching in their training programme, compared to 55% of remaining respondents (p<0.001). Trainees who identified themselves at risk of leaving also felt that their operating skills were not appropriate for their training level, and 74% were worried they would not feel confident performing procedures independently before they complete their training programme. Similar studies in the US showed higher rates of confidence in operating skills, with a recent study by Fonseca et al of 653 final year US general surgical trainees reporting that approximately 25% of respondents had a significant lack of confidence in performing a variety of open surgeries16.While the largest American study by Yeo et al found year of training to be significantly associated with attrition risk (most withdrawals occurring in the first or second year), this was not the case in our study. To the contrary, we found the highest percentages of trainees considering withdrawal were in their penultimate or final training year (74% and 61% respectively), although this was not statistically significant. We postulate that this may be due to the very high levels of stress and sacrifice related to study and completion of the general surgical fellowship exam at the beginning of the final year of training.This study has several limitations, most of which are due to the nature of survey-based research. Our response rate was low (26%) and we acknowledge that we only have received data from a minority of trainees. Whilst the rate of withdrawal risk in this study population was found to be 54%, the true rate amongst trainees may be much higher or lower. It is also possible that there was a selection bias in the type of trainees who replied, with those who are inherently less satisfied or struggling being overrepresented. Our sample also had a higher rate of female trainees than the overall training cohort, and as female trainees are at higher risk of withdrawal this may have had an influence on the high attrition risk rate. Despite these limitations, this study is the first to consider this issue in Australian and New Zealand general surgery trainees, and we hope the results will be used as impetus for further research in this area.It would be of great value to survey trainees who have actually withdrawn to elucidate the reasons they did so, as this would be helpful in further determining the factors that place individuals at attrition risk. In addition, it would be interesting to know what paths these trainees took following leaving surgery\u2014whether they transferred to another specialty, took a position in a non-clinical field or left medicine altogether. If a negative experience in surgical training is resulting in abandonment of any medical career, this would be a cause of significant concern. There is clearly potential for further research in this field, and results will be of great use in terms of workforce planning and development, surgical educational reform and in the broader context of gender and generational cultural shifts in medicine.ConclusionAttrition in surgical trainees presents a significant concern for all those involved in surgical education and training as well as workforce planning. The current 5 year training programme entails commitment and sacrifice on behalf of the trainee, and there is no doubt that the stressors and difficulties encountered during these years will leave some trainees questioning whether they can complete their surgical education. This study is the first of its kind in Australasia, and has allowed estimation of withdrawal risk compared to actual attrition rate and the factors contributing to it. Whilst this study may have been affected by a responder bias, with dissatisfied trainees being more likely to reply, it has identified that female trainees are significantly more at risk of withdrawal than their male counterparts and that thoughts of attrition are prevalent amongst a proportion of trainees. Overall, the majority of trainees we surveyed were satisfied, but problems relating to lack of support, perceived poor teaching, lack of appropriate operative skills and excessive working hours were expressed by all trainees, regardless of whether or not they had considered withdrawal from training. Pleasingly, despite these issues, the vast majority of trainees felt committed to completing their training. From here, we must focus on identifying the group of at risk trainees, particular female locally-trained registrars, and start considering how the modifiable risk factors for attrition can be addressed. We need to listen to the concerns of current general surgical registrars and find ways to improve the quality of their training to ensure we have not only adequate numbers for our future surgical workforce, but a cohort of surgeons who have enjoyed their training experience and will strive to create a similar one for their own trainees in the future.