Medicine in the United States has been hampered by a lack of diversity for centuries. Over the past few decades, greater attention and efforts have been focused on tackling this problem. A diverse workforce is important for many reasons, including a commitment toward a representative workforce in the name of social justice, improved patient care in the form of increased access and patient satisfaction,1 and service to communities for whom we provide care.2 Programs aimed at increasing diversity exist at every level of medical training, from premedical programs to the subspecialty match. Programs throughout are important because structural and individual racism impede minority success at every level, but the earliest interventions are the most likely to succeed. Efforts aimed at increasing diversity will not succeed by simply targeting subspecialty trainees alone. In essence, at the subspecialty level, this is a zero-sum game. The gains made by any one program are exactly balanced by the losses borne by others. A significant impact on increasing the pool of underrepresented in medicine (URiM) candidates for fellowship positions requires that efforts begin early and continue throughout the entire educational journey.
The Association of American Medical Colleges defines URiM as “racial and ethnic populations that are underrepresented in the medical profession relative to their numbers in the general population.” This is defined further as individuals who self-identify as African American or Black, Hispanic or Latino, American Indian or Alaskan Native, or Native Hawaiian or Pacific Islander.3
Many programs exist that aim to increase diversity within the medical profession. Some initiatives are focused on children and youth as early as middle school (Science Students Together Reaching Instructional Diversity & Excellence through the Florida State University College of Medicine4 ) and high school (Doctors of Tomorrow at the University of Michigan5 ) to increase exposure to medicine. The University of Illinois at Chicago’s Urban Health Program, which encompasses programming from preschool through graduate education, has long-standing partnerships with community colleges that serve many URiM students.6 At the medical school level, recruitment initiatives have included increasing URiM student representation on admissions committees, holistic review used to assess applicants by using experiences and attributes in addition to purely academic metrics, and exposing and combating implicit bias. Combating racism itself at all levels is more difficult but absolutely essential to create a meaningful path forward.7 Residency programs have increased outreach via conferences targeted to URiM applicants as well as URiM-specific second look programs. For fellows, multiple institutions have travel awards for interviews as well as targeted recruitment.
However, with specific regard to pediatric subspecialty fellows, few data exist on which, if any, of these efforts have made an impact at the end of the pipeline. Analysis of pediatric subspecialty data at a national level is required to appreciate the ultimate effects, what challenges remain, and the further steps required. Although recruitment efforts at an institutional level may be locally successful in increasing diversity, a potential outcome may be to provisionally enhance specific programs at the expense of competing programs, without increasing diversity overall.
We analyzed published data from the National Graduate Medical Education Census for the years 2006–20188–20 and extracted data for pediatric residents and the 13 American Board of Pediatrics recognized subspecialties in existence during that time period. We used the Association of American Medical Colleges definition of URiM above. Trainees who self-identified as multiracial or other or unknown were excluded because we were unable to determine if they met the URiM definition. Data from 97 047 residents and 41 096 fellows were analyzed.
From 2006 to 2018, there was a 42% increase in the annual overall number of fellows (from 2503 to 3542) but only a 37% increase in the number of URiM fellows (from 382 to 524) leading to a decrease in the overall proportion of URiM pediatric fellows (15.3% in 2006 to 14.8% in 2018; P = .01; Fig 1). Pediatric fellowship applicants primarily come from the pool of pediatric residency graduates. From 2006 to 2018, there was a 12% increase in the annual total number of pediatric residents (from 6965 to 7794) and 16% increase in the number of URiM residents (from 382 to 524). Although the proportional increase of URiM pediatric residents was greater than the proportional increase in total pediatric residents over this time period, the United States proportion of URiM pediatric residents overall was statistically unchanged (18.3% in 2006 vs 19% in 2018; P = .13).
Efforts at individual training programs to increase URiM fellows do not exist in a vacuum. There are a finite number of trainees at every level. Pediatric fellowship programs compete with each other to attract candidates from a defined pool. Programs designed to “increase” diversity at the subspecialty level are limited to those URiM trainees completing residency. Efforts to “increase” diversity in any given residency program are limited to the number of URiM graduating medical students, and so forth. The proportion and absolute number of URiM residents or fellows completing training in any given year is a product of the educational pipeline set in motion in the previous 10 to 15 years. This number is divided over the multitude of training programs. Because this is a zero-sum game, any program designed to increase the number of URiM medical students going into a specific field or residents going into subspecialty training, if successful, will decrease the number entering other fields. Sadly, we are collectively failing in addressing this fundamental problem.
It follows that programs designed to increase diversity in one pediatric subspecialty will result in less diversity in other subspecialties. Trying to increase URiM physician representation at higher levels of training forces each individual pediatric subspecialty to compete against other subspecialties and primary care for a limited pool of applicants. The efforts put forth in this competition does nothing to change the overall landscape of URiM physicians caring for children. To truly make a difference in the proportion of URiM physicians overall and, specifically, pediatrics it is imperative to examine and address problems in the pipeline before and during medical education. Further efforts at every stage are necessary to determine at what point disparities arise and at what points and why URiM students diverge from the path. Only by evaluation of the pipeline will we be able to make educated, data-driven steps to truly increase diversity.
Efforts need to begin long before medical training. Although small programs that focus on the pipeline exist, marked expansion of endeavors that connect with and provide mentorship for URiM students in middle and high school and even earlier is needed. The implementation of additional longitudinal early engagement programs, which increase exposure and pique interest in medicine and provide critical mentorship early on, is essential. The FACES for the Future Coalition, founded in Oakland but now a national program, focuses on high school students and has demonstrated success with 100% graduation rates in participants, compared with 40% of their peers. In addition, 90% of participants at graduation intend to pursue a health care career.21 Historically Black colleges and universities have been successful in preparing students to succeed and serve as an example of what type of programs are needed. Xavier University has dedicated peer and instructor led drill and tutoring programs for premedical studies, an early alert system for struggling students, support from top leadership, and an environment rich in same race role models.22 The success of these programs is validated by the fact that more Black students from Xavier University attend medical school than almost any other university.23 Participation in activities that increase the likelihood of college and medical school application success and interaction with physician mentors will help URiM students envision medicine as a desirable and achievable career path. Programmatic changes in the admissions process for undergraduate and medical school are also needed, including reevaluation of commonly used admissions metrics. Once the hurdle of admissions is crossed, further investment in support of organizations focused on URiM students; formal commitment to antiracism, diversity, and inclusion; and targeted mentorship to facilitate graduation success will be key to retention.
Such efforts will require bidirectional engagement between medical institutions and our communities as well as a coordinated national effort. Programs such as the HealthCare Anchor Network provide a platform for medical institutions to tackle social determinants of health and invest in building local, sustainable, inclusive, and equitable communities.24 Our academic medical centers and professional associations should take a leadership role in both arenas across developmental stages of the next generation of URiM physicians. In addition, mere implementation of programs is insufficient; ongoing evaluation is essential so that ineffective interventions can be discontinued and resources can be focused on those that work. Broader benefits to an academic medical center could include a framework for teaching cultural humility and communication skills for both trainees and practicing physicians.
Unfortunately, those looking for a “quick fix” for diversity will be disappointed. The reality is that the road to medicine is long, with significant lag time between early educational programs and increased URiM student representation. Success will require significant effort, far beyond merely recruitment at postgraduate levels. Academic medical center leaders with a serious commitment to diversity will need to ensure programming at every educational and developmental level ensures that URiM students interested in medicine have a fair chance at success. Investment at any one level is unlikely to be successful without continued efforts at all levels: from prenatal care to science enrichment classes in high school to mentorship in medical schools.25
If we want to see meaningful change in our lifetimes, we must place increasing emphasis on an early and sustained pipeline effort. We believe our institutions, professional societies, and profession must rise and give leadership to the challenge.
Dr Weyand conceptualized and designed the study, collected data, conducted analysis, drafted the initial manuscript, and reviewed and revised the manuscript; Dr Nichols conceptualized the study, was involved in interpretation of data, critically reviewed the manuscript for important intellectual content, and revised the manuscript; Dr Freed conceptualized and designed the study, was involved in interpretation of data, and reviewed and revised the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: Dr Nichols is employed by the American Board of Pediatrics and sits on the board of the Annie E. Casey Foundation. Drs Weyand and Freed have indicated they have no potential conflicts of interest to disclose.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.