The Right Thing to Do, The Smart Thing to Do: Enhancing Diversity in the Health Professions -- Summary of the Symposium on Diversity in Health Professions in Honor of Herbert W. Nickens, M.D. (2001)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

How Do We Retain Minority Health Professions Students? Michael Larimer Rainey SUNY Stony Brook School of Medicine “In the competition to recruit minority students, most medical schools relaxed their admissions standards . . . On the other hand, no school relaxed its gradua- tion requirements. Even as affirmative action spread, schools remained bound by their fiduciary duty to society to graduate only competent physicians. Ac- cordingly, schools accepted the fact that some students would require extra help and additional time.” (Ludmerer K. 1999) There is a wealth of published information on efforts to increase the prepa- ration for, admission of, and education of historically underrepresented minority (URM) students in allopathic medical education. This paper will focus on reten- tion of Blacks, American Indian, Mexican-American, and Mainland Puerto Ri- can (URM) students compared with non-minority (white) medical students. A third group, “other minorities” (Asian/ Pacific Islander, other Hispanic, and Commonwealth Puerto Rican), will not be included in this analysis. In the ensuing discussion, it is hoped that representatives from other health professions can contribute, since little data is published for these health profes- sions schools. It is also recognized that students in the other minority categories as well as foreign students and non-traditional students may also experience problems with retention similar to the problems discussed in this paper. THE CHALLENGE In 1999, 4,181 underrepresented minority students applied to 126 allopathic medical schools. A total of 2,041 (49%) matriculated. URM students represented 328 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 329 10.9% of the applicant pool and 12.1% of the entering class for the 1999–2000 academic year (AAMC, 1997). Using the best available retention data, by No- vember 2000, we can predict that about 3% (61) of this group of URM students are already off schedule for graduation with their class in 2004. Ultimately, at graduation day in May/June 2004, the model predicts that 38.9% (814) of URM students admitted in 1999–2000 will not be on stage with their original classmates receiving the M.D. degree, being hooded by the faculty and shaking hands with the dean. In contrast, 14.6% of non-minority (white) students would be predicted not to graduate with their class that day. This differ- ence in graduation rates reflects very different patterns of attrition, promotion, remedial strategies, intervention, and retention between these two groups of medical students. Historically, the primary strategy to increase URM representation in medi- cal education has been to increase the number of minority students admitted to medical schools. It is time to take a closer look at the retention rates of accepted URM applicants. Why are URM medical students three times more likely to experience academic problems that result in changes in academic status and de- layed graduation than their non-minority classmates? Either URM students are not as well prepared as non-minority students to succeed in medical school or medical schools do not provide a learning envi- ronment conducive to the success of URM students. Or both! What are the bar- riers and what can be done to lower, or better yet remove, these barriers and maximize URM retention and on-time graduation rates? EFFORTS TO INCREASE THE “URM PIPELINE” TO THE MEDICAL PROFESSION In 1970 when the enrollment of underrepresented minority (URM) students in U.S. medical schools was 2.8%, the AAMC initiated a task force to expand educational opportunities in medicine for Blacks, Hispanics, and American In- dian/Native Alaskan students in the medical profession. The stated goal was an enrollment of 12% URM students by 1975 or 1976 (AAMC, 1970). One strategy employed by some medical schools was to start post baccalau- reate or special reinforcement programs using medical school resources. Early examples of these programs were Wayne State (1965), the University of Illinois (1969), SIU (1972), and New Jersey (1972). In the late 1980’s the AAMC concluded that there needed to be a renewed effort directed at the fundamental cause of minority underrepresentation. “…Too few minority young people are both academically prepared for and interested in the health professions…” (Nickens & Ready, 1999). Medical schools also began to admit URM students with academic credentials that were lower than the school’s usual cut-off levels. Non-academic factors, such as extracurricular ac- tivities, leadership, and inter-personal skills which could be discerned from the 

330 THE RIGHT THING TO DO, THE SMART THING TO DO application and through a personal interview, were also used to make admissions decisions. The focus was on non-cognitive factors which would help schools to identify URM students who had the potential to be successful medical students even if their grades and MCAT scores were lower than non-minority students. These same factors were also used to accept women, non-science majors, rural applicants, and older applicants. This was a period of great diversification of the medical school student body, especially women and non-traditional students, but URM students did not benefit as much through these efforts as did other groups. Despite good intentions and considerable effort, first-time enrollment of URM students did not reach 12% until the 1994–1995 academic year, and total URM enrollment did not reach 12% until the 1996–1997 academic year. From 1975 to 1989, the proportion of minorities in the population increased by 22%, while the proportion in medical school increased by only 12% (AAMC, 1997). In 1991, Dr. Robert Petersdorf, in his presidential address at the Annual Meeting of the Association of American Medical Colleges, challenged United States allopathic medical schools to matriculate 3,000 underrepresented minority students by the year 2000 (Petersdorf, Turner, Nickens, & Ready, 1990). This highly promoted initiative highlighted a renewed interest on the part of the AAMC and the medical schools to increase the number of medical students from historically underrepresented minority groups—Black, Hispanic, and American Indian/Alaska Native. At the time Project 3000 by 2000 was announced in 1991, the total number of first-time URM applicants to medical school was 2,854 and 1,584 of these URM applicants joined the 1991–1992 class of medical students. Medical schools, historically the passive benefactors of the college pre- medical applicants, began to explore ways to directly increase the size and qual- ity of the URM applicant pool. In contrast to the short-term strategy of post bac- calaureate programs, a long-term strategy was also explored. Many medical schools joined in educational partnerships with elementary and secondary schools and community groups at the start of the pre-medical pipeline and then at various later stages with colleges and universities. These initiatives involved medical school faculty and administrators di- rectly interacting with potential applicants before and during the application process which made it possible for these minority youths to have an opportunity to have first-hand exposure to the medical school culture, medical students, fac- ulty, and administrators. Using a variety of different templates, collaborative efforts were made to increase the draw, flow, and output of the pipeline to maximize the quality and quantity of URM students who applied to and were accepted by medical schools. The April 1999 issue of Academic Medicine is devoted to descriptions of 12 K–12 programs and 14 College and Medical School Programs (Nickens, & Ready, 1999). The pipeline approach focused on convincing young minority students and their parents that medicine was a desirable and realistic career goal. Once stu- dents entered the pipeline, the focus was on improving their overall science edu- 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 331 cation and retention, and reinforcing their motivation to seek out careers in the health professions, especially medicine. It was assumed that if more qualified and motivated minority applicants applied to medical school, more URM students would be accepted at predominately white medical schools. Once accepted, pre- sumably the hardest part, they would progress satisfactorily through the medical school curriculum, graduate in four years, complete residency training, and enter the practice of medicine, hopefully helping the poor in underserved areas. In summary, starting in the 1960s the AAMC and most medical schools be- gan serious efforts to increase the number of enrolled medical students from historically underrepresented minority groups. In 1960, black, Hispanic, and American Indian/Alaska Native students represented 1% of the graduating sen- iors. For many medical schools, that represented none or one person of color on stage receiving the M.D. degree. In 1970 the percent of URM students had in- creased slightly to 1.3%. In the 1970s the numbers began to noticeably increase and by 1980, 8.4% of the graduating class were members of a URM group. In 1990 the percentage had increased modestly to 11.7% (AAMC, 1997). During this 40-year period, the number of medical schools increased and the overall enrollment increased from 5,553 to 15,398 students. In other words, the almost 12-fold increase in minority enrollment did improve both in absolute numbers and compared with the overall 3-fold increase in medical student total enrollment. Meanwhile, the minority population from which these students were drawn and the minority patient population continued to grow at a significantly higher rate. By 1999 there were 4,181 URM applicants to medical schools, a decline of 6.8% from the previous year. Of these, 2,041, or 49%, matriculated. URM ap- plicants represented 10.9% of the total applicants in 1999 and 12.1% of the ac- cepted students starting in the 1999–2000 academic year. For purposes of comparison, the URM applicant pool for the 1998 entering dental school class included 9.5% URM applicants and the entering class con- sisted of 8.7% URM students. In the 1996/1997 academic year, URM students comprised 11.1% of the entering dental class (American Dental Association, 1999. Dental Practice. [Online]. Available: www.ada.org/prof/ed/careers/ factsheets/dentistry.html). In baccalaureate nursing schools in fall 1999, black students represented 10.8% of the enrolled students, American Indian or Alaska Native represented 0.7% and Hispanic 4.5%. In graduate nursing programs URM students represented 12.4% of students in masters programs and 8.1% in doctorate programs. (American Association of Colleges of Nursing and the Na- tional Organization of Nurse Practitioner Faculties, 2000. 1999–2000 Enroll- ment in Baccalaureate and Graduate Programs in Nursing. [Online]. Available: www.aacn.nche.edu and www.nonpf.com). In clinical laboratory education pro- grams, a cohort study published in 1999 reported URM enrollment of 14.7%. This was the only health professions study outside of medicine that I was able to find which reported attrition data. Of the 272 URM students in the study, 25 

332 THE RIGHT THING TO DO, THE SMART THING TO DO voluntarily withdrew and 61 were dismissed, for an overall attrition rate of 31.7% (Laudicina, 1999). In fall 1998 a total of 33,090 students enrolled in schools and colleges of pharmacy in pursuit of their initial professional phar- macy degree. Of these, 12.3% were described as minority students (no definition of “minority” was provided.). In 1999, in a study of first professional degrees conferred in pharmacy, 9.8% of the graduates were black, Hispanic, and Native American. (American Association of Colleges of Pharmacy, 2001, Pharmacy Education Facts and Figures. [Online] Available: www.aacp.org/students /pharmacyeducation.html). Recent anti-affirmative action initiatives and judicial decisions in the latter part of the decade have had a negative impact on the number of URM students who applied to and were accepted by medical schools. The medical school class that matriculated in 1999–2000 contained 1,923 URM students. The breakdown of URM students in this cohort is 7.9% black, 0.7% Native American, and 2.8% Mexican American. Clearly, much more work needs to be done to expand the draw, flow, and output of the pipeline, both in terms of the quantity and quality of URM appli- cants applying to medical schools. But this is only part of the equation. The other part is to decrease the “leakage” of URM students once admitted to medi- cal school. As we will see, too many URM students, deemed to have potential to succeed in medical school, are not being retained or graduated on time at the same rate as non-minority students. Why? URM ATTRITION AND RETENTION Starting in 1992, the AAMC conducted a cohort study of all students ad- mitted to U.S. medical schools. Data is presented, year by year for URM stu- dents (Black, Mainland Puerto Rican, American Indian/Native Alaskan), non- minority (white), and Asian Pacific and other Hispanic. Data is presented on withdrawals, leaves of absence, dismissals, graduation, and still in school. For this paper I have chosen only to look at the URM and non-minority (white) stu- dent cohorts (AAMC, 1998). 

TABLE 1 Academic Progression of 1992 Underrepresented Minority (URM) Matriculants1 1992 1993 1994 1995 1996 1997 Status Number Percent Number Percent Number Percent Number Percent Number Percent Number Percent Withdrawal 4 0.2 10 0.5 26 1.4 41 2.2 46 2.5 51 2.8 Dismissal 13 0.7 28 1.5 49 2.7 57 3.1 62 3.4 On Official 13 0.7 28 1.5 61 3.3 69 3.8 46 2.5 38 2.1 Leave In School 1,804* 99.1 1,772 97.2 1,708 93.7 1,664 91.3 560 30.7 228 12.5 Graduated 1,114 61.1 1,444 79.2 *All remaining Number columns add up to 1823. This column adds up to 1821. SOURCE: AAMC (1998). Minority Students in Medical Education: Facts and Figures XI, 1998. Washington, DC: AAMC. TABLE 2 Academic Progression of 1992 Non-Minority Matriculants1 1992 1993 1994 1995 1996 1997 Status Number Percent Number Percent Number Percent Number Percent Number Percent Number Percent Withdrawal 17 0.2 66 0.6 110 1.0 129 1.2 137 1.2 142 1.3 Dismissal 18 0.2 28 0.3 41 0.4 52 0.5 60 3.2* On Official 53 0.5 120 1.1 337 3.0 304 2.7 215 1.9 178 9.5* Leave In School 11,083 99.4 10,949 98.4 10,678 95.7 10,666 95.6 1,226 11.0 396 21.2* Graduated 13 0.1 9,523 85.4 10,377 93.0 1 Based on enrollment status as of November of each year. Note: Racial/ethnic categories do not include foreign students. *These figures would seem to be in error. They should be corrected as follows: 3.2 should be 5; 9.5 should be 1.6; 21.2 should be 3.6. SOURCE: AAMC (1998). Minority Students in Medical Education: Facts and Figures XI, 1998. Washington, DC: AAMC. 

334 THE RIGHT THING TO DO, THE SMART THING TO DO MEDICAL STUDENT RETENTION As shown in Table 1, in 1992 1,821 URM students were admitted to medi- cal school, and 11,157 non-minority students were admitted: • 1,114 URM students graduated in four consecutive years. This is a four- year on-time graduation rate of 61.2%. This compares with a four-year on-time graduation rate of 85.4% for non-minority (white) medical students. • A total of 707 URM students were either dismissed, withdrew, or were placed on an extended educational program. If the non-minority graduation rate is applied to the URM cohort, only 273 URM students would have failed to graduate on time. This means that there was an “excess” of 434 URM students who did not graduate on schedule. • The overall four-year graduation rate for this cohort of students (includ- ing other minorities) is 18%. Using this figure we would expect 328 minority students not to graduate in four consecutive years, which gives us an “excess” of 399 students if the same non-graduation rate was used for all students. • In November of the second year 97.2% of URM and 98.4% of non- minority students were still enrolled although not necessarily promoted. • A few months into the third year, 93.7% of URM and 95.7% of non- minority students were still enrolled. • The percent of URM students who have withdrawn or been dismissed by the beginning of the junior year is 4.3%, twice the rate of non-minority students at 2.1%. In addition, 5.6% of the URM students have been granted a leave of absence (LOA) compared to 4.6% of the white students. • At the start of the fourth year the percentages of students still enrolled were 91.3% (URM) and 95.6% non-minority. At this juncture, 8.7% of the en- rolled URM students are not likely to graduate on time, compared with 4.5% of the non-minority students. • In November of the fifth year following a May/June graduation, 30.7% of the URM students were still enrolled and had not graduated with their original classes. This compares dramatically with 11% of the non-minority students. In other words, on graduation day URM students are roughly three times more likely to find themselves not receiving the M.D. degree than their non-minority counterparts. • By the end of the cohort study, the URM students had accumulated a total of 242 leaves of absence, or about 1 instance for every 7 students. The non- minority students accumulated a total of 1,207 LOAs or 1 per every 9 students. Leaves can occur multiple times and may extend from one academic year to the next. • By the start of the sixth year, 79.2 % of URM students had graduated compared with 93% of non-minority students, and 14.6% of the URM students 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 335 were still enrolled or on an official leave of absence. In contrast, 5.2% of the non-minority students were still enrolled or were on an official leave of absence. Put another way, after more than five years in medical school, 379 minority stu- dents, 20.8% of the original cohort who matriculated in 1992, either had not received their M.D. degree or were at high risk of never receiving it. This com- pares with 776 white students (7%) who also matriculated in 1992. • Overall, URM students were dismissed six times more frequently than white students, withdrew three times more frequently, were somewhat more likely to take a leave of absence, and were three times more likely to still be enrolled in medical school at the start of the sixth year of school. This cohort data reports events, but it does not report reasons for the status changes. We do not know why the students’ status was changed. Was a student dismissed for failing one course, two courses, or more? Did the student with- draw to avoid a termination? Why were some students placed in extended pro- grams? Why did a student take a leave of absence? There is a multitude of rea- sons why a student might graduate late, ranging from earning an advanced degree, to an illness, to multiple academic problems. Using AAMC individual student records, Huff and Fang were able to answer some of these questions (1999). They used data for only 13,118 students for whom there was complete data, versus the 16,289 students whom they reported matriculated in 1992. The AAMC table reports a total of 16,053 students started medical school in 1992. The AAMC reported that overall, 82% of the students for whom they had complete data in this cohort eventually graduated. Their data is not broken down by racial groups. The following gives data about the first re- ported events which resulted in a change of academic status: • 537 (4%) of the students in the cohort experienced academic difficulty, which caused graduation to be terminated or delayed. • Other known reasons why students did not graduate on time, or never graduated, include 555 students engaged in research (4%), 117 (<1%) for health reasons, three (<1%) for non-academic dismissal, nine (<1%) students died, and four (<1%) students experienced financial difficulties. • 768 (5.8%) students graduated late or not at all for “other” documented reasons. • In addition, 407 (3%) students graduated in more than four years without documented reasons. In general, we see that URM students were 97% more likely than were their counterparts in the referent group to experience academic difficulty, controlling for the effects of all other variables. Academic problems are the primary reason why students experience a change in academic status. However, the 768 students who were terminated or graduated late for “other reasons” and the 407 students who graduated in more 

336 THE RIGHT THING TO DO, THE SMART THING TO DO than four years without documented reasons, highlight a major problem in using this data. For about 9% of the cases, the medical schools did not provide suffi- cient information to determine the reasons for the status change. Based on my 30 plus years of experience in medical education I am admit- tedly jaded about the veracity of medical student transcripts and the quality of the information which schools communicate to the AAMC. Especially when it concerns minority students. I seriously question that only 3 of over 13,000 stu- dents were dismissed for non-academic problems. Or that only four students had financial problems which delayed graduation or caused them not to graduate. I believe that the numbers in these categories are much higher and are buried in the “other” category or in the “voluntary withdrawal” category as a face-saving strategy on behalf of the particular student and/or the medical school. I am also uncomfortable saying that the students who graduated on time (82%) did not have any academic problems. Undoubtedly, some of these stu- dents failed individual course exams, some may have received unfavorable clinical evaluations in their chosen field although they passed the clerkship, or some received low grades in courses in which they expected to excel. There are schools with academic policies that would allow a student to fail a course, reme- diate it quickly, and not have this failure reported on the transcript, in a dean’s letter, or to the AAMC database. Another problem with this data is that a single event, a course failure, or failure on a NBME Step exam would trigger different responses from different schools based on their academic policies. Along the same lines, the threshold for multiple problems might produce different actions based on academic policy or the actions of a school’s “promotions committee” or “academic standing committee.” Based on the Huff and Fang data, we now know that of the 1,449 leaves of absence given to an unknown number of individual students (a student could have more than one LOA over a six-year period), 555 were for research. Thus, there were probably about 900 leaves of absence granted for academic and per- sonal reasons. If we eliminate the 126 reported instances of death or illness we might reduce the number of leaves to slightly less than 800. Again, a leave of absence could well be granted to a student who has an academic problem com- pounded by a personal or health problem. The student could be granted a leave of absence as a way to avoid failing a course that would result in a dismissal or to discourage the student from withdrawing from medical school. In my experience, many medical schools under-report academic and personal problems encountered by their students. Confidentiality, paternalism, poor record keeping, concerns about lawsuits, unwillingness to disclose problems with the school’s academic policy, and enforcement of policy are underlying reasons. Finally, there is the issue of when these problems arise. Huff and Fang re- ported, “The evidence also highlights the critical time periods for encountering problems. Students with lower mean MCAT scores and lower undergraduate science GPAs tended to experience academic problems throughout the first three 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 337 years of medical school, as did URMs, whereas older students tend to have prob- lems only in the first two years. Overall, students already identified by their scores as being at risk experienced the majority of problems during the first year.” (1999, p. 459.) The bottom line is that about one of three minority students encounters significant changes in academic status during medical school, most often beginning in the first year, which result in leaving medical school or gradu- ating late with a troublesome transcript. Why is this the case? WHY DO URM STUDENTS HAVE MORE ACADEMIC AND RETENTION PROBLEMS THAN THEIR NON- MINORITY CLASSMATES? The following are five problem areas that need to be explored: Admissions, Curriculum, Faculty, Support Services, and Remedial Strategies. At the conclu- sion of this paper I will make specific recommendations for improvement of medical school URM student retention. Admissions For several decades, medical school deans have strongly articulated the de- sirability of a diverse medical student body that reflects the composition of the patient population to be served. It is the medical school admissions committee, the gatekeeper to the medical profession, that plays the major role in increasing the number of minority students enrolled in medical schools. Unfortunately, deanships of U.S. allopathic medical schools have been experiencing significant turnover in recent years. Of the 125 deans who were in office on August 1, 1999, in either a full or interim capacity, 22 (18%) were no longer occupying those positions on July 31, 2000 (Barzansky, Jonas, & Etzel, 2000). High turn- over and short tenure of medical education leadership do not bode well for strong and consistent support for the increasing admission and retention of URM students in our medical schools. In general, underrepresented minority students apply to medical school with lower average grades and lower average MCAT scores than members of the admissions committee feel are needed to predict success in their medical school. Sometimes these cut-off numbers are based on national or school studies. But often these cut-off numbers are really used to reduce a large applicant pool down to a manageable interview size, given limited faculty time to interview applicants. We know that average grades and MCAT scores are reported rather than the range of grades and MCAT scores because there are many students, minority and non-minority, with scores below the cut-off values who are inter- viewed, accepted, matriculate, and graduate on schedule. Admissions committees seem to believe that they are the final gatekeepers to the medical profession, that everyone they admit ultimately graduates. They 

338 THE RIGHT THING TO DO, THE SMART THING TO DO see their decisions as “high stakes” decisions protecting society from medical school graduates who are ill-suited to the practice of medicine and the medical profession. Therefore it is their duty to admit only students with acceptable grades and MCAT scores who also impressed the interviewers. Ironically, it is the students themselves who, by their acceptance or rejection of a medical school’s offer, actually determine the composition of the first-year class at each medical school in the fall. A school with one hundred places in the class reviews several thousand applications, interviews hundreds of pre-screened applicants, selects a few hundred acceptable candidates, and offers acceptances on a rolling basis. The actual overall composition of the class on the first day of classes is determined by the students who show up, not by the committee. For the 1992 retention study cohort of students previously presented, the following table contains information about their average MCAT scores and grade point averages: TABLE 3 Academic Profile of 1992 Matriculants by Status as of November 1997 URM Graduated Enrolled1 Withdrew2 Dismissed3 BCPM GPA 2.97 2.89 2.83 2.78 AO GPA 3.32 3.27 3.38 3.19 Total GPA 3.12 3.04 3.07 2.96 Biological Sciences 7.8 6.8 6.3 6.4 Physical Sciences 7.3 6.9 6.1 6.3 Verbal Reading 7.8 7.1 7.6 6.8 Writing Sample O4 O O N (median) Non-Minorities Graduated Enrolled1 Withdrew2 Dismissed3 BCPM GPA 3.44 AO GPA 3.56 3.50 3.52 3.32 Total GPA 3.49 3.57 3.65 3.47 3.53 3.58 3.37 Biological Sciences 9.5 10.2 8.9 8.3 Physical Sciences 9.3 10.1 8.8 8.8 Verbal Reading 9.6 10.0 9.6 8.8 Writing Sample O O O O (median) 1“ Enrolled” includes those who are on official leave of absence. 2“ Withdrew” category denotes voluntary withdrawal for academic, financial, health, and other reasons. 3 “Dismissed” category covers those dismissed from medical school for academic or non- academic reasons. 4 “O” and “N” are MCAT scores (range is from J to T). NOTE: Racial/Ethnic categories do not include foreign students. SOURCE: AAMC (1998). Minority Students in Medical Education: Facts and Figures XI, 1998. Washington, DC: AAMC. 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 339 Academic information about these students shows that the non-minority students across the board had higher average college grades and higher average MCAT scores than did the URM students. URM students who graduated on time had higher average grades and higher average MCAT scores than did URM students who were dismissed, withdrew, or took more than four consecutive years to graduate. In the case of non-minority students, this same pattern holds, with two exceptions. The white students who withdrew or delayed graduation had higher grades and MCAT scores than did the students who were dismissed, withdrew, or took more than four consecutive years to graduate. No information was provided to explain this anomaly. Perhaps they were students who left medical school to pursue master’s or Ph.D. degrees. It should be remembered that these are average scores for each group. Clearly, for both groups of students, there will be students with high grades and high MCAT scores who do not graduate, and students with low grades and low MCAT scores who do graduate on schedule with no academic difficulties. Many medical schools, seeking out the most academically qualified URM applicants, typically offered acceptances to the same subset of the entire URM applicant pool. Many schools were not willing to give strong consideration to URM students with lower academic credentials. It is easy for the chair of the admission’s committee to say, “We recruited, interviewed, and accepted enough URM students to do our part to meet the goal of 3000 by 2000, but less than a handful actually matriculated to our school. We did our part. Don’t blame us. We tried.” For the 3000 by 2000 challenge to be achieved, however, given the current rate of acceptance, there need to be at least 6,000 minority applicants in the pipeline. This has not happened yet. There is a growing body of literature regarding non-academic or non- cognitive factors which may be useful in predicting academic success in college and professional schools. Since the typical URM applicant has college grades and MCAT scores below the cut-off levels, admissions committees have become in- terested in non-cognitive factors which can be derived from the application or the interview process that would be helpful in selecting URM and non-traditional students with the greatest chances of academic success in medical school. After reviewing the literature, particularly with respect to students in the health professions, I am not convinced that anyone has found a combination of academic and non-cognitive factors which highly predicts success in all phases of medical education. In fact, it is even questionable if the MCAT and GPA, sepa- rately or in combination, are reliable predictors. Tucker suggests several factors which are believed to be helpful in promoting successful college transition: vi- sion, the image which students hold for their future, and a sense of community in which a student feels a sense of belonging to a new educational environment. (Tucker, 1999). Sedlacek and colleagues suggest a number of factors including positive self-concept or confidence, realistic self-appraisal, understanding and dealing with racism, preferring long-range goals, successful leadership experi- 

340 THE RIGHT THING TO DO, THE SMART THING TO DO ence, demonstrated community service, and knowledge acquired in the field (Sedlacek & Prieto, 1990). Cariaga-Lo and colleagues, in their study of medical student attrition, reported that non-cognitive characteristics that influenced stu- dents’ chances of academic difficulty were “being more norm-favoring, less self- realized, lower tendency for achievement via independence, age (generally older), gender (women) and race (nonwhite).” (Cariaga-Lo, Enarson, Crandall, Zaccaro, & Richards, 1997). In 1999, Greg Strayhorn put together a very useful literature review of non- cognitive variables and research (Strayhorn, 1999). There is much food for thought in this line of research. These variables have the potential to be useful in screening URM students whom otherwise would not be considered, and in looking for clues to help the student in academic difficulty to identify underlying problems and to seek solutions. Non-cognitive factors probably play a role in how well students interview, how well they adapt socially to the new environ- ment of medical school, how well they cope with racism and other stressors, and how well they are able to function in the clinical setting. I have worked with medical students from a variety of different back- grounds, who perform below their own and/or the school’s expectations. I be- lieve that there are additional factors, discoverable as part of the admissions pro- cess, that are predictive of success in medical school. These include the students’ knowledge of their own most efficient and effective learning strate- gies, willingness to try new ways of learning, appropriate use of learning strate- gies, openness to seek out help, appropriate use of time management skills, feelings of being “imposter” medical students, family support, adequate finan- cial support, lack of role models and mentors, and inability to cope with stress and failures. At one time or another many medical students, even successful students, wonder if they have made a serious career choice mistake. Anyone who has served on a medical school admissions committee knows that selecting applicants to medical school is a demanding and difficult task. Is a 3.1 science GPA from a selective college a better predictor of success in medical school than a 3.4 from a less selective school? Will a student with a very high verbal reasoning score and only average biological sciences and physical science scores perform better than an applicant with an average verbal reasoning score and above-average physical science and biological science MCAT scores? Will an older student with impressive life experiences be a stronger clinical student than a 21-year-old right out of college with impressive grades and no extracur- ricular accomplishments? The database used to admit or deny a novice student entry into the profes- sion of medicine includes: 1) a four-page application/transcript, 2) an essay which may or may not have been written by the applicant, 3) the results of a one-day multiple choice exam, 4) a few carefully crafted, positive faculty letters of recommendation, and 5) one or two hour-long interviews. Admissions com- mittee meeting discussions about each applicant are often relatively short. 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 341 Committees often have little or no information about how well students they have accepted in the past have performed in medical school, or how well stu- dents they rejected did in the medical school which they subsequently attended. As anyone who has served as a member of an admissions committee knows, selecting successful medical students is more an art form, less a science. The selection of URM students most likely to succeed is compounded by the problem that medical school admissions committees typically have few URM faculty members. In 1989, for example, URM faculty represented only 2.9% of the clinical faculty at U.S. medical schools (Petersdorf, Turner, Nick- ens, & Ready, 1990). Kondo and Judd reported in 2000 that on average, there were 4.1 URM members (16%) per admissions committee. Physicians with URM status comprised 8% of committee membership. Half (51%) of medical schools had one or no URM representatives (Kondo & Judd, 2000). This means there are usually few URM faculty with whom interviewing URM students can identify. In committee deliberations, there are few URM fac- ulty, even fewer with senior rank, who can provide insight about life experiences of URM students, have knowledge about the colleges they attended, and can provide insight into non-cognitive factors which might be helpful in selecting successful URM candidates. Prediction studies provide us with clues about which applicants might en- counter academic problems. However, since no medical school openly admits to conducting controlled admissions experiments, it is unlikely that we will ever know with exact precision what personal and academic characteristics most ac- curately predict academic performance. The absence of a published controlled study is ironic given that most members of a typical medical school admissions committee consider themselves to be scientists, or physicians who practice medicine scientifically. The Curriculum Medical research and technology have significantly increased the fund of knowledge medical students must master in four short years of medical school. The content of the medical school curriculum is under constant change as new discoveries are made. But the way this content is delivered to medical students has changed relatively little since the Flexner Report in 1910 (Flexner, 1910). GPEP, ACME-TRI, and, more recently, the AAMC Milbank Study of Curricular Change all point to tremendous internal resistance to changing the format of curriculum delivery in both the basic sciences and clinical years. (AAMC, 1984; AAMC, 1999; AAMC & Milbank Memorial Fund, 2000). There is resistance to moving away from focusing on content delivery to focusing on student learning. A cutting-edge lecture presentation, which results in no learning on the part of the audience, is not education. In particular, there has been a reluctance to re- place the lecture format as the primary method of presenting core material to 

342 THE RIGHT THING TO DO, THE SMART THING TO DO students. Small groups and the use of computers, for example, have made rela- tively little inroad into curriculum delivery. Labs have virtually disappeared in medical education with the exception of the gross anatomy dissection labs, which may soon be replaced with computer simulations. In an environment where the emphasis is on hours of lecture delivered, rather than knowledge ac- quired, it is the weakest and least-prepared student who will suffer the most. Nowhere in the curriculum is this more evident than in the first semester of the first year. The typical college “pre-medical curriculum” at best prepares medical students for the content of the first several weeks of the medical school curriculum. It does not prepare students well for the culture of medical educa- tion or for the profession of medicine. It does not prepare students adequately for the pace of the curriculum, for the volume of material to be mastered, and for the detail-oriented questions on multiple-choice exams. Many students typically are not facile with the learning tools needed to survive in the medical school environment. It has been my observation that medical students are usually so smart that they often have not learned how to learn effectively and efficiently. When strategies acquired in high school and college start to fail them, they are clueless about what to do. They tend to put in more hours studying using their customary strategies rather than consider trying different, and potentially more effective, strategies. Seeking help is a strategy of last resort. URM students, by and large, are at greater risk, given their diverse back- grounds, of not thriving in the unfamiliar, fast-paced environment of medical school, especially during the first semester. The retention cohort data presented earlier suggests that many academic problems that result in attrition start in the first semester and are compounded in the second semester. By the second year, students who are not performing well have either left medical school or have already been placed on a remedial plan, allowing them more time to learn the material. The first semester of the first year at most medical schools consists of a gross anatomy course and a biochemistry and/or molecular biology course, plus perhaps one or two smaller courses. There is a “boot camp” mentality operative during the first several months of medical school. Faculty create an environment where the workload increases dramatically and the early exams are designed to produce relatively low scores. The curriculum during the first semester is delivered primarily in a lecture format. Last year’s lectures are read again by the faculty and students follow along reading the photocopied notes from the previous year. The pace is fast, there is no time for questions, and it is a good day when the AV equipment works properly. Anatomy lab is scheduled for three hours but there are five hours of expected dissection work. The tests use the multiple-choice format with a heavy emphasis on the recall of detailed information. The pace is so fast that comprehension and long-term retention are nearly out of the question for many students. Memorization is the most common strategy and cutting one class to 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 343 cram for a test in another course is the norm. The word “survival” is common when students talk about the first semester. Learning is secondary. There are two possible explanations for this behavior on the part of course directors. One is that many faculty think it is their job to weed out those students who should not be in medical school; that is, to correct admissions committee errors. The other is to create a boot camp environment to toughen up the students, focus their attention on academics, and motivate (scare) them into giving 110% to learning the material in their course. Within a few weeks, even the students with very strong science backgrounds are starting to flounder and non-science majors are wondering if they have made a horrible error in career choice. The medical school basic science faculty are researchers and content experts. Teaching medical students is something they do in order to be able to hold a fac- ulty appointment in the medical school with its rich research resources. Many do not feel that they will be promoted or given raises based on their teaching per- formance. They would rather teach graduate students who also work in their labs, and who will also be their proteges. Faculty are expected to deliver an expanding body of knowledge, in an organized way, in a pre-set number of hour-long lec- tures. Typically medical students have little input into faculty promotion and ten- ure decisions. They either are not asked or their comments are given little or no consideration. After all, there is only one anatomist on staff who can teach the medical student gross anatomy course. If she is not promoted and given tenure, who will teach the course? Course directors and primary lecturers in courses are typically not members of an underrepresented minority group. Medical school faculty tend to look at lecturing, reading, and testing as the only ways students can learn the material and the only ways that mastery of their knowledge can be tested. Faculty equate “teaching” with “student learning.” They have been resistant to any form of small-group teaching, especially prob- lem-based learning (PBL). Even in the anatomy lab, the groups of four to five students assigned to each cadaver are regarded as dissection or work groups, not as learning groups. What would happen, for example, if the group was tested as a team and all received the same grade? This would encourage all members of the group to make sure that they were, as a group, well prepared for the test. Perhaps the “gunner” in the group would be less likely to come into the lab after hours to do additional dissection so he will get a better grade on the next exam, while depriving his partners of a learning experience. For many beginning medical students, the test scores they receive in the first semester are lower than they expected. Many report they are the lowest test scores they have ever received. They wonder how they have suddenly become so “stupid.” They typically do not share these scores with classmates, non- medical student friends, spouses, partners, or family. Since the grade distribution is announced in class, it is known that 20% of the class failed the exam. Students who failed, or almost failed, the exam tend to isolate themselves so that class- mates will not ask them how they did on the exam. And unfortunately, many 

344 THE RIGHT THING TO DO, THE SMART THING TO DO students in the class believe incorrectly that all the minority students are in the group that failed. Medical school faculty, by and large, assume that every student learns the same way that they learn. UMR students, by and large, are at greater risk, given their lower grades and MCAT scores, of not thriving in the educational culture of medical education. One study has suggested that perhaps URM students pre- fer a different learning style compared with their non-minority classmates. Tay- lor and Rust suggested white students are “assimilators” and prefer the lecture format, while URM students tend to be “convergers,” “divergers,” or “accom- modators” and are at a disadvantage when more interactive styles of teaching methods are not used (Taylor & Rust, 1999). Another possibility is that a URM student may have an undiagnosed learning disability that may not be recognized because of the presumption that URM stu- dents always perform at a lower level than non-minority students. A URM student who suspects a problem may also choose not to submit to testing because she lacks the funds to pay for it. She may also be reluctant to be labeled as “disabled.” Most medical schools conduct an orientation program prior to the official start of the first year. For the most part, this program is an orientation to the medical school facility. With the exception of a talk by the dean and a White Coat ceremony, it is not an orientation to the profession of medicine. Other than handing out textbook lists and the class schedule, there is little attention given to helping students manage the curriculum and navigate through the first semester. The students are more likely to be given information about the best student- friendly bars in the area than information on how to prepare for medical school lectures, how to take notes, how to read medical textbooks, how to study for and take multiple-choice exams, and other learning strategies which might be useful in the fast-paced, content-rich medical school curriculum. They receive no ad- vice about being effective adult learners. The sophomores who are invited back early to “orient” new students or to tutor first-year students are typically those students who did extremely well academically the previous year. They probably have little understanding of the problems some classmates experienced. Many do not even know why they themselves were successful. It has also been my experience that minority students often do not participate actively in the orientation programs. They often need the orientation week to take care of housing and financial aid. I have, in the past, attended workshops con- ducted by Leon Johnson, the former president of National Medical Fellowships (NMF), for minority students preparing for medical school. He makes a strong case regarding minority students who have become too dependent on easy-to- obtain, high-interest credit cards for daily living expenses. Students who do not pay off credit cards on time develop poor credit histories which make it very dif- ficult for them to qualify for financial aid, to obtain credit, to open bank accounts, and to pass credit checks for apartment leases and utilities hookups. As a result they often miss portions of the orientation program so that they can take care of 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 345 “life” problems which non-minority students in the class were able to take care of weeks ahead of time. Some students with high consumer debt try to secretly work part-time while in medical school just to pay off their debts. This usually leads to poor academic performance, missed classes, and lack of sleep. Medical schools that do not devote time and resources to preparing students for the medical school curriculum are missing an opportunity to improve the quality of life for students during the first year. This information can help to level the academic playing field for URM and non-traditional students. The money “lost” on scholarships and financial aid given to students who do not graduate and default on loans would more than pay for the cost of providing more support services for students in academic difficulty. A study recently published by the AAMC and the Milbank Memorial Fund takes a close look at curricular change at 10 medical schools. “The good news is that many medical educators are implementing curricular changes that are re- sponsive to the latest advances in biomedical science, to the social and policy sciences relevant to medical practice, to the burden of disease, to the organiza- tion and financing of health care, and to the changing demography of the Ameri- can population.” (AAMC & Milbank Memorial Fund, 2000, p. v). Northwestern was the only medical school featured in the report which specifically addressed the issue of diversity of its student body. The Northwestern report stated, “Fi- nally, we will be working to further increase the cultural diversity of our medical classes and to enhance the ‘cultural competence’ of each individual student . . . “Nevertheless, we still fall far short of our goals for enrollment of under- represented minority students . . . With respect to each student’s skills in work- ing with patients from other cultures . . . We are creating a new program to en- sure that every student has the opportunity to learn basic medical Spanish and increasing opportunities for clerkship experiences in ethnic neighborhoods.” (AAMC & Milbank Memorial Fund, 2000, p. 146). Other than this one exception, I did not find acknowledgement of the in- creasing diversity of the medical student body or that URM students had higher levels of attrition than non-minority students. There was an occasional comment that faculty needed to know more about adult learning, and that perhaps it would be beneficial to increase small-group learning formats, such as PBL. Few medical schools have elements of the curriculum which involve cul- tural competency. Typically this term is used to describe curricular efforts to inform future practitioners of different cultural values, norms, and beliefs as they apply to birth, life, growing up, puberty, pregnancy, illness, old age, and death. Ideally, the curriculum would include language training sufficient for a student to do a basic physical exam in two languages, but the term could also be applied to helping students understand the backgrounds of other members of their class. URM students could share their own knowledge and experience with other members of the class as discussion group leaders and participants, helping 

346 THE RIGHT THING TO DO, THE SMART THING TO DO to teach physical diagnosis language courses and being available to the hospital as translators and patient advocates. Curriculum accountability is an area that has attracted considerable attention in recent years. At many medical schools there is no curricular authority and no central education budget. Departments own most of the courses except for a few interdisciplinary “dean’s” courses. The curriculum committee primarily sched- ules classes, adjudicates disputes about course hours and use of lecture halls and labs, and determines exam schedules. Lip service is often given to evaluating the curriculum, the quality of individual courses, faculty educational efforts, and overall student performance. If a large number of first-year students fail bio- chemistry, for example, would this be the province of academic affairs or student affairs? Would the promotions or academic standing committee focus on the course or on the failing students? Typically, student failure of a course is consid- ered to be the fault of the student, not the instructor or the course itself. Who would have jurisdiction if all 12 students who failed were URM students? Faculty Basic science and clinical faculty with medical school appointments have come under increased pressure to “earn their keep” through funded research projects and clinical practice revenue. As noted earlier, teaching medical stu- dents is not valued or rewarded when tenure is granted or salary increases are determined. Medical schools and associated research and clinical facilities are expensive to operate. Tuition pays only a fraction of the total cost of medical education. Medical students are transitory “visitors” to the medical school cam- pus. Faculty stay forever. There is a severe shortage of URM faculty teaching core courses. “Al- though African Americans, Native Americans, Mexican Americans, and Main- land Puerto Ricans make up almost 25% of the U.S. population, they account for less that 8% of all practicing physicians. Only 3% of medical school faculty members belong to one of these minority groups.” “First, minority faculty, by virtue of their small numbers in a given medical school, is disadvantaged by comparative isolation within the academic community. Second, minority faculty often feel disproportionately obliged to serve on time-consuming committees, to mentor students with complicated nonacademic problems, and to participate in community service activities that are not typically career advancing. There is even a term that has been coined for these contributions, the black tax. Third, attainment of senior faculty rank by minority faculty is tantamount to crashing a long-running party at which a relatively circumscribed group of invitees has had privileged access to the trappings of power” (Cohen, 1998). A study published in 1998 reported faculty in each ethnic group working similar hours, but black faculty spent more time in clinical activities and less time in research. More black and Hispanic faculty felt pressure to serve on 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 347 committees due to their race or ethnicity, although they actually did not spend more time in hours per month on committee-related activities than the other fac- ulty groups. They also reported that controlling for age, URM faculty were sub- stantially less likely to have attained senior rank than white faculty. URM fac- ulty were also found to have a greater debt burden than other faculty, which may explain more time spent in clinic and less time engaged in research (Palepu, Carr, Friedman, Amos, Ash, & Moskowitz, 1998). The medical school faculty is predominately composed of white males, par- ticularly persons in positions of educational leadership and administration. URM students are marginalized in this environment. They do not encounter faculty with whom they can identify. They have difficulty finding faculty who over- came the same problems that they are now encountering. How could a faculty member who never needed to study hard, never failed a course, and did not have to borrow large amounts of money understand what a URM student, who is at risk of flunking out of medical school, is going through on a day-to-day basis? Because URM faculty tend to hold lower rank and to experience difficulty gaining promotion and tenure, they do not have much time to mentor minority students, provide them with survival strategies, be advocates for them with deans and promotions committees, provide career advice, or help with applying to resi- dency programs (Fang, Moy, Colburn, & Hurley, 2000). The absence of URM faculty decreases the chances that a URM student can find a faculty mentor, someone who has “walked the same path” and can serve as a guide. The increas- ing diversity of medical school classes will, in the next 10 to 20 years, begin to diversify the predominately white male faculty of medical schools, especially the clinical faculty. In 1989 URM faculty represented only 2.9% of the clinical fac- ulty at U.S. medical schools (Petersdorf, Turner, Nickens, & Ready, 1990). In the AAMC’s 2000 Medical School Graduation Questionnaire, seniors anonymously responded to a wide range of questions about their experiences at their school. Each school received their own aggregate results as well at national results. When asked to what degree the racial and ethnic diversity of the school’s student body positively fostered professional growth and development, only 34.2% of the students said it was a moderate or strong influence. When students were asked if they had been personally mistreated during medical school, 20% of the almost 14,000 respondents said yes. When this 20% were asked if they were denied opportunities for training or rewards because of their race ethnicity, 12% said one or more times. When asked if they had been subjected to racially or ethnically offensive remarks/names directed at them personally, 15.8% said one or more times. Asked if they felt they had received lower evaluations or grades based solely on their race or ethnicity rather than performance, 12.3% said one or more times. When asked the source of this mistreatment, faculty in the clinical setting and interns/residents were cited as the most common sources of mistreatment. While the numbers are small, these instances of racial discrimi- nation or harassment loom large in the eyes of the medical student victims, es- 

348 THE RIGHT THING TO DO, THE SMART THING TO DO pecially when the perpetrators are physicians who are also their teachers, men- tors, and professional role models. These behaviors are perpetuating harassment from one generation to another (AAMC, 2000). Support Services I recently conducted a small study of the minority affairs officers listed in a publication called 2000–01 Diversity of American Medical Education published by the AAMC (AAMC, 2001). This publication listed the names, degrees and titles of all persons identified as each school’s minority affairs officer. Of the 120 predominately white medical schools (excluding Meharry, Howard, More- house, and the three schools in Puerto Rico) there were 69 (57%) schools who listed a minority affairs officer. Of these, 38 persons (55%) were physicians and 37 (54%) held the title of associate dean or higher. Half the schools had a mi- nority affairs officer at some administrative level, but only one-quarter of the schools had a minority affairs officer at the senior level of administration. Race was not identified in the publication, but based on my own knowledge, a large percentage of these minority affairs officers are members of an underrepresented minority group. According to the current issue of AAMC Medical Schools Admissions Re- quirements, every medical school lists a contact person for URM applicants. A total of 101 (78%) persons listed held the title of Dr. (although it was not speci- fied whether it was a M.D. or other earned doctorate). A total of 68 persons (53%) held some variation of the title minority affairs officer. A total of 52 (40%) held the title of associate dean or higher. Overall, only 26 (20%) persons listed had a doctorate degree, the title of associate dean, and the title of minority affairs officer (AAMC, 2001). Overall, while a majority of predominately white medical schools list a mi- nority affairs officer, only about one in four has one with a Ph.D. or M.D. who holds senior rank in the administration. This is not to say that a minority person with an M.S. in degree counseling and the title of director could not be an effec- tive advisor, counselor, and advocate for minority students. However, a person without an advanced degree and a senior-level administrative title is less likely to be a voting member of the admissions, academic standing or promotions com- mittee, or the curriculum committee, and is less likely to have access to the dean. A relatively recent addition to the administrative staff of some medical schools is a learning assistance specialist. I found no resource that provides data on learning assistance specialists at medical schools or that specifies that they are readily available to medical students on the university’s main campus. I sus- pect that most were initially hired because of problems that minority students were encountering. Once in place, they provide service to a wide variety of stu- dents who wish to perform academically at a higher level. A few even have linkages to faculty development programs for medical school faculty. The avail- 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 349 ability of these specialists increases support resources available to medical stu- dents. More are needed. Even if support resources are available to URM students, they still may have to seek help for personal or academic problems. It has been my experience that URM students, like most medical students, are reluctant to ask for help. The students, when asked, are aware of the services available, but are less likely than non-minority students to make appointments or to respond to offers of assis- tance. Denial, fear of exposure, concerns about confidentiality, lack of trust, pride, the possible high out-of-pocket cost of diagnostic testing are all possible reasons. The fact that most of the persons providing assistance are not URM faculty or staff is an additional contributing factor to explaining poor utilization. A high percentage of URM students come from economically disadvan- taged backgrounds and cannot afford the high cost of medical tuition and living expenses without assistance. Most need loans and scholarships to attend medical school. It has already been mentioned that some of these students or their par- ents have credit rating problems, which preclude loans. Scholarships are often used by the admissions committee to attract URM students with high academic credentials. URM students, for example, receive about 6% of the cost of medical education from their families, compared with 20% for non-minority students. URM students are considerably more dependent on loans and scholarships for their medical education (AAMC, 1998). They are also more likely to come to medical school with a higher undergraduate debt than non-minority students. Many URM students often contribute some of their financial aid to help their parents or siblings. However, virtually all forms of financial aid assume that the student is in good academic standing. If a student has to repeat a year, she or he may lose a scholarship or may not be eligible for a loan. During the repeat year the student may have difficulty finding financial aid and will have to seek employment while also attending medical school or studying for a remedial exam. Students who need more than four years to graduate may find expensive loans or work part-time while doing clinical rotations. None of these options are desirable and they may significantly interfere with the student’s ability to recover academi- cally. The financial aid officer needs to be an involved member of the support team capable of dealing with the difficult financial issues which many of the URM students face, especially when they encounter academic problems. Failure to obtain adequate financial support during periods when they are experiencing academic problems, are in a remedial mode, or are on leave of absence, can contribute to eventual dismissal or withdrawal from school. Remedial Strategies There is the temptation to identify students at risk of academic and/or per- sonal problems in medical school, and to attempt to intercede, either by requir- 

350 THE RIGHT THING TO DO, THE SMART THING TO DO ing them to attend a special pre-orientation program or to see a coun- selor/advisor soon after classes start. There are several problems with these strategies. First, these strategies run the risk of stigmatizing the student doing poorly and of increasing his or her already high level of anxiety. Often, the stu- dent refuses to participate in special programs and/or resists preventive advise- ment appointments. He or she firmly believes that all that is needed is to “study more” and things will be fine. Some medical schools have experimented with “lightened load” programs in which students spend three years completing the first two years of medical school. While this strategy may help some students who have non-academic problems which may limit their ability to fully participate in the curriculum at full pace (for example, a mother with a young child), in most cases these pro- grams can be stigmatizing, making the participants feel like second-class citi- zens. In some schools, students who get into early academic difficulty are forced into lightened load programs without giving them the chance to succeed on their own. Overall, I have not been a proponent of obligatory or forced lightened load programs. While they may lighten the load, they do not help students develop requisite survival skills and, too often, they stigmatize the participating students. One option is to encourage faculty to give a test one or two weeks after the start of classes which does not count heavily toward the final grade, but is con- structed just like a real exam with the same level of difficulty. Then conduct a post mortem. It would be helpful if the faculty can provide information about questions that students missed. Was this information they should have known from college or was it new material? Did the students make careless errors sug- gesting they did not read the question carefully? Did they change right answers to wrong answers? Did the students finish the exam? Providing this information to the students and to the learning assistance specialist opens up an opportunity for each student and counselor to discuss study skills and testing strategies. This is important because the counselor may not be a content expert. Waiting until the sixth week before the first test results are available is often too late for initial feedback because it allows relatively little time for corrective action to occur before the mid-term exam. The early first exam score gives the student a “ticket” to seek help and the counselor/advisor a specific reason to call the student in for a counseling session. If a medical student fails a first-year course, there is typically a six- to eight-week break between the first and second years which the student can use to study and take a remedial exam or take a summer course at one of a dozen medical schools which offer remedial courses. After the second year there is usually a month before the junior year starts and at most medical schools, stu- dents are encouraged or required to take the Step 1 exam before they start clini- cal rotations. The last two years of medical school run almost continuously with periodic week-long breaks and end one or two months before graduation. In this extremely compressed curriculum, the only mechanism for allowing a student to 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 351 remediate a failed course is to derail the student from normal progress and give the student extra time to deal with the course failure. This basic strategy of giving more time is fundamentally flawed. First of all, if a student could not learn in a supervised, structured environment the first time through, why is it assumed that she can learn better in an unsupervised, un- structured environment? If the student is using inappropriate learning tech- niques, additional time is a waste of time. Counselors need to work with the student to determine why she is failing and help her acquire strategies that will assure success in the future. Support, not time, is what is needed more. Unfortunately, a common strategy is to give the student a leave of absence, either to study the material for a make-up test or to take a time out and wait for the course to be offered again. Another strategy is to tell the student that he or she has to take a make-up test at the end of the summer before the next year starts. Unless the student has access to financial resources, this means that the student often has to return home. This removes the student from sources of aca- demic support from faculty, tutors, classmates, learning assistance specialists, and administrators. Faculty, who are already complaining about too much work, are reluctant to advise or tutor remedial students and often do nothing more than provide the student with a generic reading list. Over time the “banished” student will feel estranged from the school, will be less likely to seek help, and be less inclined to return to medical school to try again. Instead, every effort needs to be made to keep the student on campus, in contact with resources and, if possible, in the classroom. In the absence of health or personal problems that require the student to leave campus, I would strongly support remedial strategies that would keep the student in an academic environment. Often there are untapped resources in the form of medical students in M.D./Ph.D. programs who would be interested in tutoring a student, which in turn helps them to review material prior to resuming their own medical training. Junior faculty, who someday will be teaching them- selves, could help a struggling student as part of their own professional devel- opment. Other URM students who have been successful could collectively or- ganize tutorial or help sessions for struggling URM students. A list of URM faculty who would be willing to serve as tutors, or coaches, or just someone to talk with should be readily available to struggling students. Another approach is to ask classmates who are doing well in the course to provide assistance, either on a voluntary basis or with compensation provided by the dean’s office. Rather than use the “best” students in the class, it might make more sense to identify students with prior teaching experience or students with master’s degrees. A student who will have to repeat first semester will still profit by sitting in on, or auditing, second-semester courses. No one in the class needs to know that the student did not pass all first-semester courses. This allows the student to see what is coming, may help him or her to put first semester material into perspec- 

352 THE RIGHT THING TO DO, THE SMART THING TO DO tive, and allows him or her some degree of dignity to stay with his or her class- mates and friends. This approach will necessitate creative ways to finance the student’s living expenses for the year. Intervention should be coordinated between the student affairs, minority af- fairs and academic affairs staff, the learning skills specialist, and the course di- rector of the failed course. This will also alert the academic affairs dean in cases where a large number of URM students are failing the same courses. This in- formation should also be provided to the admissions dean and the admissions committee. It has been my experience, at several medical schools, that in some first- semester courses the content “ramp” is too steep. Too little time is taken by faculty to review material which they assume all students should have learned prior to starting medical school. A summary of the academic records of the class should be made available to first-semester faculty. For example, how many students had biochemistry courses in college, how many have advanced science degrees, how many students were non-science majors. The pace of first-semester courses is of- ten too fast, causing most students to very quickly fall behind. A few questions at the start of a lecture might help the lecturer to determine if the pace is appropriate and if key concepts from the previous lecture have been understood. There is nothing sacred about the first semester concluding before the De- cember holiday period. Giving students a block of time to study in December might improve overall learning and exam performance. Scheduling first- semester exams after the holiday season would also help decompress the first semester. There is also nothing sacred about giving medical students a month or two break at the end of the first year. While this is a prime time for remedial efforts, perhaps a few extra weeks should be added to the end of the semester to review basic core concepts before the final exam. All of these strategies would allow all students more time to learn the material and result in fewer failures during the critical first year. Once a student experiences an academic failure that results in a projected delayed graduation date, there appears to be a cumulative effect that signifi- cantly increases the chances the student will never graduate. The student no longer has the support of friends and classmates. She has increased financial problems. She believes that her failures are common knowledge. She may be- lieve that future faculty will know she has failed a course and will pre-judge her. And, of course, her transcript may keep her from getting desirable electives at other hospitals. Residency program directors are less likely to grant her an inter- view with a flawed transcript and graduation in more than four years. Early identification of academic failure, swift and intense efforts to provide assistance by faculty and administration, making every effort to keep the student on schedule, and providing continuing and adequate financial aid are essential elements of a successful remedial strategy, especially for first- and second-year academic problems. 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 353 SUMMARY Project 3000 by 2000 and prior initiatives increased the size and quality of the URM medical school pipeline and probably the number of URM candidates for admission to the other health professions as well. The application credentials of the URM candidates for admission to medical school increased and so did the interest in identifying non-cognitive factors which would predict success in medical training. However, one of three accepted URM medical students still fails to gradu- ate on time. Most academic problems begin in the first semester of the first year and are not resolved, causing delays in graduation or a failure to graduate. Ad- missions committees need to continue to look beyond grades and MCAT scores for indicators of academic success among URM applicants, and the faculty need to take a close look at elements of the medical school curriculum, especially the first semester, which may be causing avoidable academic failures. Faculty need to explore other instructional methodologies than lecture, to create a more hospitable and effective learning environment for URM students. This should also include a curriculum which promotes cultural competence. URM faculty are few in number and struggle to be successful in their own ca- reers and also be available to help future URM faculty succeed in medical school. It should be recognized that when a URM student gets into academic difficulty, it is not a good practice to just give the student more time without supervision and structure. The student needs to stay on campus with adequate financial support and with access to all available support services. Minority af- fairs officers, in conjunction with other deans, need to be aggressive and vigilant advocates for URM students. The financial challenges facing students who need additional time to complete their medical training must be met by the medical school. The school needs to actively preserve the financial, as well as profes- sional, investment which they have made in the students accepted by the admis- sions committee. RECOMMENDATIONS With limited sources, limited time, and an uncertain political climate, which strategies would have the greatest bang for the buck and the greatest probability of success? Below are 34 specific recommendations to improve retention of URM medical students that would also be applicable to students in the other health professions. Admissions 1. The word “diversity” should be part of the mission statement of every medical school accredited by LCME. 

354 THE RIGHT THING TO DO, THE SMART THING TO DO 2. Schools should continue to recruit, interview, and accept URM students to meet a new AAMC goal of 20% URM enrollment by the year 2010. 3. Medical schools should get more directly involved in their own URM pipe- line which would involve increasing the numbers of URM students on cam- pus prior to the start of the admissions process. 4. Deploy senior URM faculty, residents, and students to serve on the admis- sions committee as recruiters, interviewers, and voting committee members. 5. Track the progress of admitted students in the curriculum and use both cog- nitive and non-cognitive factors to determine the success profile for a school consistent with (1) above. 6. If scholarships and loans are available, commit resources for no less than five years without a requirement of academic progress. 7. Encourage URM applicants to attend classes, labs, make return visits, and to come to the entire extended orientation program. Curriculum 1. Orientation should be at least two weeks long and should include an orien- tation to the curriculum, learning styles, testing strategies, and small-group work as well as an introduction to the medical school and the community. 2. During the extended orientation program some classes should be held cov- ering prerequisite material. Material presented should be tested in the same way that first-semester courses are tested. Feedback which identifies areas of strength and weakness, should be provided to students. 3. Students who are identified as potential risks during orientation should be involved in on-going coordinated assistance immediately. 4. Decompress, slow the pace, and extend the length of the first year, espe- cially the first semester. Lower the entry ramp a few degrees. 5. Increase URM faculty representation in every year of the curriculum. URM clinical faculty could, for example, provide clinical correlates, present pa- tients, and discuss cases as part of first-year courses. 6. The curriculum committees should mandate that lectures be significantly reduced and replaced with small-group learning experiences and other al- ternate methodologies. There are a variety of ways in which an electronic curriculum would foster diversity in educational modalities. 7. Learning assistance specialists should work with faculty on courses, pres- entations, and tests. 8. The structure of the course should reflect the learning styles of the students in the course. 9. Cultural competency components need to be added to all phases of the cur- riculum starting in the first semester. This can be done using small-group- based courses, which focus on social, psychological, economic, and profes- sionalism issues in medical practice. 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 355 10. Medical students at the end of the second year should be able to do an ac- ceptable patient examination in two languages. This should be an LCME standard and tested by USMLE. 11. Decrease dependence on MCQ exams, especially in the first year. Use com- puters to test students, employing a variety of testing formats. 12. Offer early systematic academic support to students during the first semester. Faculty 1. Explore ways to help URM faculty earn tenure and promotion at the same rate as non-minority faculty. 2. Find ways to involve URM faculty in the curriculum design and delivery, especially in the first year. 3. Strategically deploy URM faculty to student and education-related com- mittees. 4. Address the issues of clinical faculty and resident/intern discrimination and harassment directed at URM students in the school. Support Services 1. The office of minority affairs should be staffed with high-ranking, visible, and available staff and should have resources to provide support services to URM students. 2. Deans of student, academic, and minority affairs should work together to eliminate attrition in the first year. 3. Learning assistance support should be available within the medical school and work in conjunction with the offices of minority, student, and academic affairs. 4. Implement strategies to make students and faculty aware of differences in learning styles, and alter the curriculum and support services to maximize learning for all students. 5. Find alternative remedial strategies which are not based on “time out.” Stu- dents in academic difficulty should be on campus, working with faculty, fully supported, and able to continue their education, even if they are in a remedial mode. 6. Increase available financial aid funds for URM students and guarantee sup- port for a minimum of five years. 7. Find creative ways to encourage URM students to seek help when they en- counter academic or personal problems. Find ways to reduce further stigma- tizing students who are already coping with the prospect of academic failure. 

356 THE RIGHT THING TO DO, THE SMART THING TO DO Miscellaneous 1. Medical schools should keep detailed records of reasons why URM students experience academic difficulty, evaluate remedial strategies employed, and document outcomes. They should publish results of this research. 2. Attention should be given to both cognitive and non-cognitive variables and academic problems which are linked to specific courses. 3. Feedback should be provided to the admissions, academic standing and cur- riculum committees as well as to the office of academic affairs, student af- fairs, minority affairs and the dean’s office. 4. The AAMC should start another, more detailed cohort study. 5. LCME should take a close look at accreditation standards relative to im- proving retention of URM students and on-time graduation rates. REFERENCES American Association of Colleges of Nursing and the National Organization of Nurse Practitioner Faculties. (2000). 1999–2000 Enrollment in baccalaureate and graduate programs in nursing. [Online]. Available: www.aacn.nche.edu and www.nonpf.com [accessed December 12, 2000]. American Association of Colleges of Pharmacy. (2001). Pharmacy education facts and figures. [Online] Available: www.aacp.org/students/pharmacyeducation.html [ac- cessed January 14, 2001]. Association of American Medical Colleges (AAMC). (1970). Report of the Task Force on Expanding Educational Opportunities for Blacks and Other Minorities, 1970. Washington, DC: AAMC. AAMC. (1984). Physicians for the twenty-first century, The GPEP Report. Washington, DC: AAMC. AAMC. (1997). Minority students in medical education: Facts and figures XI, 1997. Washington, DC: AAMC. AAMC. (1998). Minority students in medical education: Facts and figures XI, 1998. Washington, DC: AAMC. AAMC. (1999). Educating medical students: Assessing change in medical education— The road to implementation. ACME-TRI Report. Washington, DC: AAMC. AAMC. (2000). LCME graduation questionnaire. Washington, DC: AAMC. AAMC. (2001). 2000–01 Diversity of American medical education. Washington, DC: AAMC. AAMC. (2001). Medical school admission requirements United States and Canada 2001– 2002. Washington, DC: AAMC. AAMC & Milbank Memorial Fund. (2000). The education of medical students: Ten sto- ries of study of curricular change. New York: Milbank Memorial Fund. American Dental Association. (1999). Dental practice. [Online]. Available: www.ada.org/prof/ed/careers/factsheets/dentistry.html [accessed December 18, 2000]. Barzansky, B., Jonas, H.S., & Etzel, SI. (2000). Educational programs in U.S. medical schools, 1999–2000. Journal of the American Medical Association 284(9):1114– 1120. Cariaga-Lo, L.D., Enarson, C.E., Crandall, S.J., Zaccaro, D.J., & Richards B.F. (1997). Cognitive and noncognitive predictors of academic difficulty and attrition. Aca- demic Medicine 72(10 suppl.):S71. 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 357 Cohen, J.J. (1998). Time to shatter the glass ceiling for minority faculty. Journal of the American Medical Association 280(9):821. Fang, D., Moy, E., Colburn, L., & Hurley, J. (2000). Racial and ethnic disparities in fac- ulty promotion in academic medicine. Journal of the American Medical Association 284(9):1085. Flexner, A. (1910). Medical Education in the United States and Canada. New York: Carnegie Foundation. Bulletin 4. Huff, K.L., & Fang, D. (1999). When are students most at risk of encountering academic difficulty? A study of the 1992 matriculants to U.S. medical schools. Academic Medicine 74(4):454–460. Kondo, D.G., & Judd, V.E. (2000). Demographic characteristics of U.S. medical school admission committees. Journal of the American Medical Association 284(9)1111– 1113. Laudicina, R.J. (1999). Minority student persistence in clinical laboratory education pro- grams. Journal of Allied Health 28(2):80–85. Ludmerer, K. (1999). Time to heal. New York: Oxford; p. 251. Nickens, H.W., & Ready, T. (1999). A strategy to team the “savage inequalities.” Aca- demic Medicine 74(4):310–311. Palepu, A., Carr, P.L., Friedman, R.H., Amos, H., Ash, A.S., & Moskowitz, M.A. (1998). Minority faculty and academic rank in medicine. Journal of the American Medical Association 280(9):767. Petersdorf, R.G., Turner, K.S., Nickens, H.W., & Ready, T. (1990). Minorities in Mmdi- cine: Past, present and future. Academic Medicine 65(11):663–670. Sedlacek, W.E., & Prieto, D.O. (1990). Predicting minority students’ success in medical school. Academic Medicine 65(3):161–166. Strayhorn, G., (Ed.) (1999). Literature review on non-cognitive variables. Chapel Hill: University of North Carolina, Fall. Taylor, V., & Rust, G.S. (1999). The needs of students from diverse cultures. Academic Medicine 74(4):302–304. Tucker, J.E. (1999). Tinto’s model and successful college transitions. Journal of College Student Retention: Research, Theory & Practice 1(2):163–175. 

358 THE RIGHT THING TO DO, THE SMART THING TO DO DISCUSSION CASE STUDY Angela is a 22-year-old Hispanic student in her first year of medical school. She was a psychology major at State College. She earned a 3.0 GPA overall, 3.1 in BCPM, and 7s on the MCAT. She was accepted by two medical schools, and chose this medical school because a classmate from State was also accepted. She was late arriving to freshman orientation because she had car trouble driving to the school 300 miles from her home. She missed the White Coat Ceremony and the presentations by the administration. When she went to the financial aid office she discovered that her parents had not submitted the neces- sary tax forms in time and she would not be receiving her financial aid package. Tuition payment was deferred until the end of the month. In the first semester, she was quickly overwhelmed by both the gross anat- omy and biochemistry courses. She had taken only the basic pre-medical cur- riculum at her school. She received 50% on the first gross anatomy test and 45% on the biochemistry test. The class average in both exams was in the low 70s. For the rest of the semester Angela focused on gross anatomy because she felt she had a better chance of passing it, especially because she did very well (85%) on the practical exam. She passed gross anatomy but failed biochemistry by five points. Angela was allowed by academic policy to take second-semester courses knowing that she would have to take and pass a remedial exam in biochemistry during the summer after her first year. She did better in the second semester, passing all courses with grades in the low 70s. Lacking financial resources to live near campus during summer, she went home to study for the biochemistry make-up exam. She worked part-time in the local library. A few weeks later her mother suffered a mild heart attack. Her parents are divorced and Angela spent a lot of time with her mother at the doc- tor’s office, translating what the doctor said. At home she helped her mother take care of her two younger sisters and the house. Angela only had few weeks during the summer to intensively study bio- chemistry. She returned to medical school the week before classes, took the re- medial exam, and failed it by one point. Angela reviewed the exam and chal- lenged two answers which the instructor had marked wrong. He refused to consider her petition to reconsider her answers, saying, “You are a marginal student. It would do you well to repeat the year. Maybe you will study harder.” She sought help from Dr. Green, the Assistant Dean for Minority Affairs. Dr. Green is a relatively young, black physician in the Department of Family Medi- cine who works part-time as the Medical School’s Minority Affairs Officer. She reviewed the entire exam with Angela. She was quite surprised to see several very poorly written questions on the exam, including the two that Angela had challenged. Dr. Green called the biochemistry instructor, who reluctantly agreed to meet with her that afternoon. 

HOW DO WE RETAIN MINORITY HEALTH PROFESSIONS STUDENTS? 359 Dr. Green pointed out the poorly worded questions. After 20 minutes of somewhat heated discussion, the instructor finally agreed that one of the ques- tions Angela had challenged should be thrown out. He pointed out, however, that after throwing out the question and recalculating the grade, Angela still re- ceived a failing grade of 69.7%. His printed rule is that the final exam grade must be at least a 70% to pass his course. Dr. Green then went to the associate dean for academic affairs and explained the situation. A week later, and several days after second-year classes had already started, Dr. Green told Angela that her grade had been rounded up to a 70 by the chair of the biochemistry depart- ment and that she has been promoted to the second year. Angela began attending second-year classes at the start of the second week. In the third semester she failed the pharmacology course. According to aca- demic policy, failure of a second course results in a “invitation” to meet with the academic standing committee. She explained to the committee that she did not have a strong science background, that she is not strong in memorization, and that she was preoccupied by her mother’s continuing health problems. She was put on probation, told to find a tutor, and to keep in close touch with her instructors. She was also told to attend more of the help sessions offered in the evenings by graduate students. She was warned that, should she fail another course, she would be dismissed. She had to use loan money she budgeted for her food to pay for a tutor. She also went to the learning assistance specialist and discovered that she was a slow reader and employed a poor strategy for taking multiple-choice exams. By mid-semester she was “just” passing all courses. A month later she received a note from the associate dean for student affairs, asking her to come in for an advisement appointment. She made the appointment, but did not keep it. Terrified that she was going to be dismissed, she started stay- ing up very late at night, studying pharmacology and other third-semester courses. She managed to continue passing exams, but began to experience severe head- aches. She thought she might need glasses but did not have the time or the money to get the glasses. She did manage to pass all third-semester courses. In her fourth semester she did extremely well in the physical diagnosis course. She received very positive reviews about her ability to conduct a com- petent patient interview in both English and Spanish, and helped to organize a physical diagnosis Spanish course for classmates. However, she had a great deal of difficulty passing the organ systems course exams. On most exams she passed by only a few points. At the start of the class, the instructor told the students, “Students who can’t pass my course never pass Step 1.” Afraid that she might fail pharmacology and later Step 1, she went back to Dr. Green and asked for help. If you were Dr. Green what would you advise? • What are Angela’s options? • What are the advantages and disadvantages of each option? • What could the medical school have done differently in Angela’s situation?