Secondary Logo

Journal Logo

Special Theme Article

Addressing the Urban Pipeline Challenge for the Physician Workforce: The Sophie Davis Model

Roman, Stanford A. Jr. MD, MPH

Author Information


There had been periodic concerns that a physician surplus would emerge by the year 2000. But this did not occur; moreover, the trend may be going in the other direction. Specifically, beginning with a report from the Graduate Medical Education Advisory Committee in 1980 and reaffirmed by several reports from the Council on Graduate Medical Education between 1982 and 1998, it became clear that U.S. allopathic medical schools had stabilized the number of medical graduates. During the same period, the number of graduates from osteopathic schools of medicine doubled. Despite these trends, the combined increase in medical school graduates was 12%, while the U.S. population increased by more than 24%. Today several observers predict a physician shortage by 2020 if these trends continue.1 Further, the projected demographic profile of the U.S. population suggests that within 20 years, our urban communities will be increasingly diverse, creating a need for a health care force that is culturally competent.

The Challenges

To respond to this projected shortage, U.S. medical schools will need to consider increasing the number of their graduates. To achieve any goals to increase that number, a larger pool of applicants is needed. This effort will, however, be hindered by new challenges in the pipeline of medical school applicants. The educational conduit for these applicants has increasingly become constricted for many minorities and students from families with limited financial means. Yet, as we look forward, these groups will represent a greater portion of the college-age population. Between 2000 and 2020, the number of African Americans and Latinos alone will increase from 30% to 37% of the college-age population. During the same period, whites will decrease in the college-age population from 72% to 66%, continuing a decline over the past decade. Asian Americans will increase from 4% to 6% of 18-year-olds. In the aggregate there will be only a 5% increase in the total number of 18-year-olds.2 These trends suggest that any goal to increase the number of U.S. medical school graduates will depend on the ability of medical schools to attract applicants and enrollees more broadly from the college-age population than they have in the past.

This ability will be challenged by differences in the persistence in their college educations of students from various population groups, which in turn reflect, among other factors, certain disparities in the quality of precollege education based on family income and race. Even when students’ persistence and academic achievement are exhibited, differences exist in the availability of highly complex learning, advisement, assessment, achievement, and educational resources, particularly in science and math, which build the foundation for advanced learning in medicine and other skilled professions.3–6

Ultimately these differences cannot be corrected without significant investment in K-12 education to more widely foster educational opportunities, as we saw in the 1970s and early 1980s, a period during which disparities among income and racial groups lessened. We must consider models in medical education that can enhance the academic skills and discipline among talented youths, who, despite a documented will to excel, have not been sufficiently challenged from early ages to support professional pursuits successfully. While high-school enrichment programs make a valuable contribution, they are rarely of sufficient duration and intensity to support the development and maintenance of the academic skills that are immediately needed in the premedical curriculum in postsecondary education.

The Model Program

Since 1973, the Sophie Davis School of Biomedical Education has operated a combined BS/MD medical education program in The City University of New York at The City College. Like most programs of its type, the school has selectively recruited and admitted talented high school seniors who have expressed a definite interest in a medical career and given them assurance that if they successfully complete the baccalaureate part of the program leading to the BS degree, they will gain admission to a related or cooperating medical school.

The Sophie Davis School has, however, been the only such program that has maintained a mission to expand access to medical careers among talented inner-city youths and youths who may have experienced educational disadvantages despite demonstrated evidence of high levels of academic achievements. To achieve the mission, we have introduced an innovative curriculum (see Table 1) that integrates the biomedical and sociomedical sciences into the baccalaureate program to permit successful students to transfer into the third clinical year in cooperating medical schools. This practice substitutes actual performance in medical studies rather than results from following a traditional premedical curriculum and taking the Medical College Aptitude Test (MCAT). The cooperating medical schools—Albany Medical College, New York Medical College, New York University School of Medicine, SUNY Health Science Center at Brooklyn, and SUNY Health Science Center at Stony Brook—have permitted advanced transfer almost seamlessly for more than 30 years. Additionally, we have developed and collaborated with a continuum of high-school enrichment programs, including Gateway to Higher Education, Project Explore, and the Bridge to Medicine, each of which has different levels of intensity and duration but in the aggregate contribute 10% of the applicants to the Sophie Davis School and 15% of the enrollees.

Table 1
Table 1:
BS/MD Curriculum—The Sophie Davis School of Biomedical Education, 2004

Second, we have maintained a mission to develop among our students the knowledge and skills requisite to encourage entry into a primary care specialty (internal medicine, pediatrics, family practice, and obstetrics–gynecology) and also to provide for students who choose other specialties an enriched foundation in those classroom-based and experiential areas that better prepare them to function more effectively in the diverse urban environment.

Third, we have embraced a mission to increase the availability of medical services for traditionally underserved communities in New York State. This mission has been reinforced since 1977 by the New York State Legislature, which has required two years of service in a medically underserved community in New York State in a primary care specialty upon completion of residency as compensation for the substantial savings realized by students pursuing their baccalaureate and medical education in this combined program. While this model does not eliminate the substantial costs of a medical education, the increased affordability permits those students who would perceive the prevailing costs of a medical education as an absolute barrier, to stay the course. Almost 27% of our students are from families whose incomes are below federal poverty levels. Seventy percent of the students come from families whose incomes make them eligible to receive support from the New York State-sponsored Tuition Assistant Program.


Our entering class of 70 each year is selected by assessing eight factors: high-school grade-point averages, the New York State Regents Examinations scores, American College Test (ACT) scores, Scholastic Aptitude Test (SAT) scores, personal statement and writing sample, high-school references, extracurricular and community activities, and two interviews, at least one of which is with a member of the admission committee. One-quarter of the applicant pool is invited for interviews, and approximately half of those interviewed are finally accepted. Table 2 shows the profile of applicants, acceptances and enrollees over five years (1999–2003). Ultimately, approximately 80% of the admitted applicants enroll.

Table 2
Table 2:
Comparative Profiles of First-Year Classes, the Sophie Davis School of Biomedical Education, 2004–2008

Applicants and enrollees reside in all five boroughs of New York City; smaller numbers reside in Nassau, Suffolk, Westchester, Rockland, Putnam, and Rensselaer Counties. Although 45% of enrollees attended public high schools that between 1999 and 2003 either accepted students by examination or had honors programs, 48% of enrollees attended average public schools. The common threads are academic achievement, motivation, and demonstrated concerns for others.

Our structure permits the introduction of a rigorous curriculum in the biomedical and sociomedical sciences with coordinated academic and student supports to offer tutorials to develop academic discipline and skills and to handle the personal challenges of maturation and professional development. The system of advisors and mentors increases the relevance of professionalism to our students and encourages them to seriously consider their academic and career goals. More than half of the entering students are the first in their families to enter higher education. Our decision to eschew reliance on premedical courses and the MCAT encourages students to learn for the purpose of increasing their fund of knowledge to support their becoming effective physicians. Additionally, the early identification as “medical students” enhances motivation and dedication to develop the skills, values, and characteristics to support their goal. While these characteristics and values are also sought in traditional medical school applicants, these applicants acquire them via relevant associations and exposures that many of our students don't have.

The integrated curriculum also facilitates longitudinal instruction and experiences. Our students develop an appreciation for the organization and function of the health care system. They learn to understand the contextual relationship between the patient, the family, and the community and to recognize the influence of culture, ethnicity, and race on the transaction among patient, physician, and health care institutions. Our students are exposed to the value of population-based and evidence-based tools in medical decision making. These subjects of instruction, experience, and inquiry enable all students to engage more effectively in providing medical services in the diverse urban environment—the environment from which most of the students have come or have adopted.

Since 1978 we have collaborated with community-based health and human service organizations within New York City through our Department of Community Health and Social Medicine to provide a range of required, longitudinal experiences for students. These experiences begin in the second year of the five-year program when our students apply their acquired tools of epidemiology to diagnose community health status from the reference point of 61 health and human services organizations directly serving local communities. In the fourth year, they have clinical experiences in eight community health centers for an introduction to primary care and patient interviewing.

Because these centers serve local communities, students immerse themselves in the cultural and socioeconomic issues that influence health and illness behavior as they begin to develop their own clinical skills. The affiliated community health centers are located in culturally and racially different communities in New York City including Chinatown (in Manhattan), the South Bronx, Bedford Stuyvesant, Sunset Park (in Brooklyn), the Lower East Side and Upper West Side of Manhattan, and East Harlem. Students develop a high degree of cultural competence, increasingly required by the physician in the urban center. The diverse cohorts of students share among themselves, in formal and informal group settings, their growing awareness of the subtle and sometimes not-so-subtle influences of different cultural and social factors on health and illness behavior and the importance of fostering the trust so necessary in the doctor–patient relationship.

Although Sophie Davis students do not participate in the MCAT, they must take the United States Medical Licensing Examination (USMLE™) Step 1 in the spring of the fifth year, as do most second-year students in U.S. medical schools. Passage of the USMLE Step 1 is required for transfer with advanced standing in all but one of the cooperating medical schools. Over the past five years the first-time passing rates have been at least 90%, which is the national average.

Several studies have assessed the predictive value of the many factors suggested to influence performance in the preclinical years of medical school, performance on the USMLE Steps 1 and 2, and performance in the clinical years and beyond. In many of these studies there is a strong correlation between SAT and ACT scores and MCAT scores with preclinical performance and performance on the USMLE Step 1.7–12 These predictive indicators weaken in the clinical years and in performance in residency training and in medical careers. Our selection of students in high school with combined SAT scores above 1,100 may indicate that once a good test taker always a good test taker, with the appropriate instruction.


In the baccalaureate environment, particularly in the large urban public commuter college or university, the concept of student cohort is most clearly defined at matriculation and at graduation. Rarely is the cohort an integral part of the longitudinal educational process. Such cohesiveness is more important in the medical school setting to assure that students acquire the appropriate knowledge and skills, as well as exposure to experiences based on common and time-limited experiences. With this in mind, in the baccalaureate environment there ideally should be an academic and administrative structure similar to that of the traditional medical school to prepare students for the required comprehensive and interdependent nature of a medical education. The structure of Sophie Davis achieves this and includes a dean and three associate deans, representing academic, student, and administrative affairs. The 34 full-time faculty are organized into five academic departments: Behavioral Medicine, Cell Biology and Anatomical Science, Community Health and Social Medicine, Microbiology and Immunology, Physiology and Pharmacology, each led by an appointed chairperson. Additionally, 70 clinical adjunct faculty offer a range of preceptor and other services in the biomedical and sociomedical sciences; some participate in teaching contracts for instruction and skills development in Pathology, Introduction to Clinical Medicine, and Physical Diagnosis. Our academic policies are developed, maintained, and monitored through standing committees: The Executive Faculty, Curriculum, Student Academic Progress, Ethics and Professional Behavior, Student Appeals, Faculty Promotion and Tenure, and Admissions. Necessary academic, student, and administrative functions are provided through 68 professional, technical, and support staff. Although all full-time faculty are engaged in instruction and are therefore the core faculty, they are also expected to actively participate in research and in the pursuit of competitive outside funding. The teaching demands limit the ability of faculty to devote most of their effort solely to biomedical research. Nonetheless, most of the 34 full-time faculty successfully compete for funding to support their research initiatives.


In the last 31 years, we have had 1,400 graduates. Less than 1% have not subsequently earned their MD degrees. The reasons for not obtaining the MD degree have been varied, including medical illness, academic failure, career change, and failure to pass Part 1 or Step 1 of the National Boards or the USMLE, respectively. Twenty-five percent of the graduates are African American, 8% are Latino, 28% are Asian American, and 39% are white. Forty percent of graduates are men and 60% are women. The racial composition among graduates as well as applicants to the school has changed over the past 20 years, reflecting the change in public and parochial high-school enrollment in New York City, as may be noted in the admission profiles for the past five years (Table 2).

The attrition rate within the school has remained in the range of 15% to 18% over the last 15 years. This is comparable to that of other combined BS/MD programs nationwide. Students dismissed from the school usually remain in good academic standing in The City College and may complete their baccalaureate degrees. Some have subsequently entered a medical school in the first year. Fifteen percent of the entering students remain in the program for six years before graduating because of leaves of absence or the need to repeat a year. This period is called a “prescription year,” during which students repeat no more than two deficient courses and receive focused supports and courses to enhance academic skills. No student may have more than one prescription year. Students transfer to the cooperating medical schools by entering into a matching program at the end of the third year; therefore, all fourth- and fifth-year students know which one of the cooperating schools they will enter as an advanced transfer upon graduation and passing Step 1 of the USMLE. Three students who have been eligible to enter the matching program have not been matched to a cooperating medical school in 15 years. Two of these entered other U.S. medical schools as advanced transfers and subsequently earned their MD degrees and are currently completing or have completed their graduate medical education. One student who was not matched has chosen another career.

During the past five years, 6% of graduates have been elected to Alpha Omega Alpha in the cooperating medical schools that do not require four years of attendance in the same school. Another indication of the effectiveness of the student transition is the residency placement after earning their MD degrees. Figure 1 shows the distribution of residency placements over five recent years (1999–2003). Over the same period, 82% of the graduates entered residencies in New York State; 54% of these entered residencies in New York City.

Figure 1.
Figure 1.:
The distribution of the placements of Sophie Davis residents over five recent years (1999–2003). Over the same period, 82% of the graduates entered residencies in New York State; 54% of these entered residencies in New York City.

Primary care has been an emphasis of the school based not on market forces, but on patient and community health needs. Over the five-year period of 1999–2003, 83% of the graduates entered residencies in internal medicine, pediatrics, family medicine, and obstetrics–gynecology. In an earlier review, an estimated 65% of graduates between 1977 and 1987 were actually engaged in primary care practice.13 Finally, 13% of graduates between 1977 and 1990 are currently members of U.S. medical school facilities.

Importance of the BS/MD Model

The development of a physician workforce for the future faces many new challenges. On the macro level, the U.S. population is projected to increase by 50% by the year 2050 with an increase of 169% in the total minority population and an increase of only 7% in the nonminority population.14 Associated with these trends will be an increase in the proportion of minorities and students from families who are less affluent. The college-age population of African Americans and Latinos alone will increase from 30% to 37% between 2000 and 2020. These projections have particular implications for medical education and for medical care.

The demographic changes suggest a population that is far more diverse both racially and culturally, particularly in the nation's cities. U.S. medical schools have appropriately begun to stress the acquisition of cultural competence among medical students and residents. Studies over the past two decades suggest, however, that at the end of the day, the concordance between physician and patient by race and culture seems to positively influence the availability of services, patients’ perceptions of the quality of care, the development of trust, and possibly, in some cases, the actual quality of care.15–17 Whether instruction in cultural competence alone can bridge these challenges without a greater diversity in the medical student and physician supply must be documented.

U.S. medical schools have maintained efforts since 1970 to increase the number of underrepresented minorities entering medicine. Most notable was the initiative of the Association of American Medical Colleges (AAMC) to reach a goal of 3,000 underrepresented minorities entering U.S. medical schools by the year 2000. Despite considerable effort and some progress, the AAMC's goal was not realized. Admissions have stagnated and declined since 1995 due to pipeline challenges. By the year 2000, 12% of U.S. medical school enrollees were members of groups underrepresented in medicine, while these groups had increased to 25% of the U.S. population.1

The pipeline for underrepresented minorities and students from families of limited financial means is challenged by several factors:

  • The increasing cost of higher education and particularly medical education. The average indebtedness today upon receipt of the MD degree was a little over $109,000 in 2003.18
  • Competition from other careers in which financial stability is perceived to be more attainable.
  • As mentioned earlier, an increasing disparity in the quality and rigor in precollege science education that prepares fewer students of all races—but particularly students from underrepresented groups and those from less affluent families—to successfully negotiate the hurdles of the traditional premedical pathway. This is apparent even when these students have demonstrated evidence of high levels of achievement before entering higher education.

The increasing cost required to provide a high-school science program of quality and the limited availability of skilled science teachers at all levels has resulted in fewer students’ being challenged in the sciences to develop the skills requisite to pursue rigorous advanced education. These factors have increased the differences among school districts, schools, and within schools as to how many students are truly prepared for advanced education in the sciences and particularly medicine. For example, the life sciences use technology as a valuable tool—a means to an end, not the end in itself. This approach requires that the appropriate technology skills be developed in high school students. In that way, those who may have the ability and interest to learn life sciences will be able to apply those technology skills in the ways that will allow them to pursue medical careers. Differences in who is taught what determine the potential pipeline to medicine. Yet, if we are to educate the number of medical graduates needed in the future, these differences in the educational conduit must be addressed. We have focused on the necessary support for biomedical research and support for those who pursue graduate studies in the biomedical and, to a lesser degree, the sociomedical sciences. We must now forcefully question from whence the future biomedical scientists and physicians are going to come, given the challenges in the precollege conduit and the projected changes in the college-age population.

The Sophie Davis model suggests that those students who excel in mastering even average complexities of precollegiate sciences can rise to the challenge of our school's rigorous medical school biomedical and sociomedical science curriculum when appropriate academic and personal supports are offered. The curriculum is presented not as a hurdle, but as part of the goal to become an effective physician. Our students perform comparably to those peers who successfully negotiate the traditional premedical and preclinical pathway. Sophie Davis's approach to curriculum shows how important it is for students to improve their academic skills. Even students whose high-school preparation might be considered strong are challenged to explore and improve areas of weakness. The shared goal of becoming an effective physician also facilitates students’ supporting each other's growth and development.

Additionally, this approach has been effective in disproportionately contributing to the number of MD graduates in New York State from underrepresented groups in medicine. During the past five years, 211 Sophie Davis graduates have earned their MD from cooperating schools, which is 3% of all MD recipients in New York State during that period. During the same period, 57 graduates of Sophie Davis from underrepresented groups represented 8% of the 684 persons from underrepresented groups in medicine receiving their MDs in New York State from 13 medical schools. Sophie Davis graduates from underrepresented groups in medicine have represented a range of 6% to 45% of the underrepresented MD graduates in the five cooperating medical schools.

Second, the substantial cost-saving of pursuing one's undergraduate and preclinical education at an undergraduate rate of tuition in a public university permits students who would not have pursued a medical career to perceive that it is possible and that they can shoulder the burden of costs and the associated indebtedness of the two years after transfer, with advanced standing, to the cooperating schools.

Third, the integrated preclinical and baccalaureate curriculum facilitates greater attention to the development of knowledge, skills and values that will enhance the ability of graduates to address the needs of a diverse multicultural urban community.

Albanese et al.19 identified the “buffering” effect of a combined BA/MD program when the medical school applicant pool declines. Identifying students interested in medicine from high school gives talented students a “path” to medicine that permits them to keep on the trajectory even if general interest in medicine declines or interest is not sufficiently high to support an increasing demand for medical graduates.

A combined BS/MD degree that integrates the basic biomedical and sociomedical sciences in the baccalaureate program can compliment goals to increase the number of medical school graduates. This is particularly relevant when the partnering medical school(s) can more easily expand its clinical capacity than its preclinical capacity for students. Additionally, this model can be effective in increasing the success of talented students who have faced the disparities of academic preparedness increasingly being observed among high schools despite the students’ demonstrated will to excel among peers. In so doing, such an approach can draw more broadly both racially and economically from the projected college-age population of the future.

It will be important, however, for this model, whether applied at Sophie Davis or elsewhere, to be recognized and monitored by the Liaison Committee on Medical Education to ensure that the longer-term standards of quality are maintained and that these standards are supported by the structure, process, resources, and outcomes requisite for the education of the physician. The control of the curriculum to the MD degree can be achieved through a collaborative partnership between the MD granting institution and the baccalaureate phase that takes full advantage of the value of additional time to prepare the physician. In the urban environment, such partnerships with strong public universities can be beneficial to increase the access to medicine among talented youths who may otherwise lose the trajectory, but who are additionally motivated and determined by the relevant “path” provided.

Helping Talented Students Rise to the Challenge

Pursuit of a medical career depends on a continuum of educational and experiential milestones that build academic skills, motivation, and personal characteristics. The quality of education, experiences, and expectations along that continuum can determine who succeeds. As our urban high schools struggle with large numbers of students and the lesser availability of skilled teachers in science and math, and the goal to provide a basic educational foundation, the richness, rigor, and depth of preparation for the skilled professions is diluted for many who are intellectually capable of more challenge. The lack of challenge for many of these students is masked because they do achieve good grades and may even perform well on standardized tests and may gain admission to selective colleges and universities. They are therefore highly successful students. However, the dilution of their educational challenge makes it much less likely that they will have the tools to easily enter medicine.

The Sophie Davis School has demonstrated that many of these students can rise to the challenge if their goal to enter medicine is made more relevant in their premedical school education and if they receive the academic and personal supports necessary for them to become effective physicians.1–17


1.Salsberg ES, Forte GJ. Trends in the physician workforce, 1980-2000. Health Aff (Millwood). 2002;21:165–73.
2.Cooper RA. Impact of trends in primary, secondary, and postsecondary education on applications to medical school. II: considerations of race, ethnicity, and income. Acad Med. 2003;78:864–76.
3.Miller L, Thomson WA, Smith QW. Perceived barriers to careers involving math and science: the perspective of medical school admissions officials. Teach Learn Med. 1992;4:9–14.
4.Attewell P. The winner-take-all high school: organizational adaptations to educational stratification. Sociol Educ. 2001;74:267–95.
5.New York Association of Community Organizations for Reform Now (ACORN) Schools Office. Secret Apartheid III: Follow Up to Failure [report]. Brooklyn: ACORN, 1998:1-65.
6.Gamoran A. American schooling and educational inequality: a forecast for the 21st Century. Sociol Educ. 2001;extra issue:135-53.
7.Koenig JA, Sireci SG, Wiley A. Evaluating the predictive validity of MCAT scores across diverse applicant groups. Acad Med. 1998;73:1095–106.
8.Roth KS, Riley WT, Brandt RB, Seibel HR. Prediction of students’ USMLE Step 2 performances based on premedical credentials related to verbal skills. Acad Med. Feb 1996;71:176–80.
9.Reede JY. Predictors of success in medicine. Clin Orthop. 1999;362:72–7.
10.Fadem B, Schuchman M, Simring SS. The relationship between parental income and academic performance of medical students. Acad Med. 1995;70:1142–4.
11.Way DP, Biagi B, Clausen K, Hudson A. The effects of basic science pathway on USMLE Step 1 scores. Acad Med. 1999;74(10 suppl):S7–9.
12.Veloski JJ, Callahan CA, Xu G, Hojat M, Nash DB. Prediction of students’ performances on licensing examinations using age, race, sex, undergraduate GPAs, and MCAT scores. Acad Med. 2000;75(10 suppl):S28–30.
13.Roman SA Jr, McGanney ML. The Sophie Davis School of Biomedical Education: the first 20 years of a unique BS-MD program. Acad Med. 1994;69:224–30.
14.Minority Business Development Agency. Minority Population Growth: 1995 to 2050 [report]. Washington, DC: USDoC, 1999:1–8.
15.Saha S, Komaromy M, Koepsell TD, Bindman AB. Patient-physician racial concordance and the perceived quality and use of health care. Arch Intern Med. 1999;159:997–1004.
16.Schneider EC, Zaslavsky AM, Epstein AM. Racial disparities in the quality of care for enrollees in medicare managed care. JAMA. 2002;287:1288–94.
17.Komaromy M, Grumbach K, Drake M, et al. The role of black and Hispanic physicians in providing health care for underserved populations. N Engl J Med. 1996;334:1305–10.
18.Association of American Medical Colleges. Medical student education: cost, debt, and resident stipend facts (October 2003) 〈〉. Accessed 23 August 2004.
19.Albanese M, Vaneyck S, Huggett K, Barnet JH. Academic performances of early-admission students to a BA/MD program compared with regular-admission students in relation to applicant pool fluctuations. Acad Med. 1997;72(10 suppl 1):S66–8.
© 2004 Association of American Medical Colleges