Dropping Out of Surgical Training

Philadelphia—For decades, residents have withdrawn from general surgery training programs at far higher rates than other specialties. Now, the first study of its kind shows that women, particularly those from ethnic minorities, are most likely to leave.

In the first longitudinal national cohort study designed to identify surgical residents at risk for dropping out, sex was the most important predictor of withdrawing from a general surgery training program. Overall, 24% of women and 17% of men abandoned general surgery training. Nonwhite women, particularly those in academic programs, had the highest dropout rate (35%).
ADVERTISEMENT

The study was presented at the 2017 annual meeting of the American Surgical Association.

But there are far more factors at play than simply sex and ethnicity, said lead author Heather L. Yeo, MD, MHS, assistant professor of surgery and healthcare policy and research at Weill Cornell Medical College, in New York City.

“This is really the first study to show that there are individual risk factors that together put a person at risk,” she said.

The reasons are complex, and reflect personal and program factors. In fact, some groups of men had high attrition rates while some women, especially those in small community programs, were less likely to drop out than most men, she noted.

“This is the first opportunity to really characterize individuals based on some more in-depth factors that put them at risk. Every woman is not at equal risk.”

Dr. Yeo and her colleagues carried out a nationwide eight-year prospective cohort study of general surgery interns from the class of 2007. The survey results were then linked to data from the American Board of Surgery, including ABS In-Training Examination dates and scores, residency completion, board status, and program characteristics. Nonparametric classification and regression tree analysis identified risk factors for noncompetition of training at the resident level. This type of analysis allows for deeper exploration of the factors affecting individuals for whom attrition was an issue.

Of 1,047 general surgery interns in 2007, 80% completed training. The residents least likely to drop out were non-Hispanic, married, white men studying at smaller nonacademic programs outside the Northeast (5%). White women at small community programs who had a relative in medicine had a similarly low dropout rate (6%).

For women, the most important factor in completing residency training was race, with 30% of nonwhite women leaving compared with 20% of whites. For nonwhite women, the attrition rate was highest in academic programs (35%) versus nonacademic programs (30%). White women in large academic programs were more than twice as likely to drop out than those in smaller programs (25% vs. 11%).

For men, program size was the most important predictor of leaving general surgery residency. Twenty-three percent of men in larger programs dropped out, compared with 11% in smaller programs.

The study findings should not and cannot be used as a screening tool for potential residents, Dr. Yeo noted. “You can’t assume every woman is more likely to drop out or that you should not train women. That’s not a solution, and it’s not an accurate assessment. Many women do very well in training. They are over 50% of the resident workforce, so we now need to figure out how to support them and train them.”

But programs could retain more residents through mentorship of at-risk residents by increasing the numbers of women and ethnic minorities on faculty in teaching programs, she noted. More support structures for residents, including day care and better options for residents who have family or personal emergencies, also would be beneficial.

Mary Hawn, MD, professor of surgery and chair of surgery at Stanford University, in California, said the study highlights the need for diversity among faculty and residents. “It is incumbent on us to create inclusive training environments. Having diversity improves the cultural competency of the program, which has a positive effect on patient care.”

Mary E. Klingensmith, MD, the Mary Culver Distinguished Professor of Surgery and vice chair for education at Washington University School of Medicine in St. Louis, said surgical patients are more diverse than the surgical workforce, and patients may have preferences for physicians with an origin or sex similar to theirs. “A more diverse workforce serves our patients well, but we also become more receptive to patients of different genders and cultures if we are working in more diverse environments where diversity is embraced, accepted, understood and considered an everyday occurrence rather than a rarity.”

The Institute of Medicine’s 2003 report, “Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care,” found that racial and ethnic minorities receive lower-quality health care than nonminorities, even when access-related factors, such as patients’ insurance status and income, are controlled. The authors concluded that the health care workforce and its ability to deliver quality care for racial and ethnic minorities can be improved substantially by increasing the proportion of underrepresented U.S. racial and ethnic minorities among health professionals.

Dr. Klingensmith also suggested that programs consider outreach efforts to residents’ families to raise awareness of the difficulty of residency and the need for support. The study surprisingly showed that residents with family who lived nearby were less likely to complete residency. Female residents with family members living nearby had higher rates of attrition than those without; 39% of female nonwhite residents studying in academic programs with family nearby left their programs before completion. Surgical educators who listened to the presentation said family sometimes may support a resident’s decision to leave due to stress.

For the past 25 years, attrition rates have remained steady in general surgery, with about one in four residents not completing training. That is much higher than other specialties, including surgical specialties. Orthopedic surgery has a dropout rate of less than 1%, OB-GYN is about 4.5%, and medicine is about 5%.

“It was thought that changes in work hours and the kinds of things might lower the attrition rate, but they have not. So we need to look at other solutions,” Dr. Yeo said.

Resident characteristics that did not significantly affect the completion of training included marital status, attending a U.S. or Canadian medical school, and geographic location of a residency program. Attrition rates were higher (24.4%) in larger programs, defined as more than six chief residents graduating each year, than smaller programs (17.7%). Military programs had the highest attrition rates (35.5%) versus academic programs (20.4%) and community programs (16.7%).

Intern age was unavailable for this sample, and there were no data on marriage or children among residents after the start of residency.

There are few reports of interventions designed to prevent surgical resident attrition. In 2010, surgeons from the University of Pennsylvania reported a dramatic reduction in resident attrition after changing the resident selection strategy in 2005. They added an essay and structured interview to the process that allows them to select traits that are aligned with the ethos of their program and assess perseverance and success in difficult situations (Ann Surg 2010;252:537-541). However, critics of this approach say it screens trainees who are likely to complete residency, but does not necessarily identify those who would be good physicians.

The full study is expected to be published in an upcoming edition of the Annals of Surgery. It was presented this spring at the 2017 annual meeting of the American Surgical Association, the oldest surgical association in the United States.