ScienceWeek

SCIENCE POLICY: ON THE COST OF MEDICAL EDUCATION

The following points are made by Gail Morrison (New Engl. J. Med. 2005 352:117):

1) The cost of obtaining a medical education has been spiraling upward for the past 20 years. Despite a great deal of rhetoric in articles[1,2] and at meetings of the Association of American Medical Colleges, nothing has happened to change the alarming pattern. The average annual tuition and fees at public medical schools during the 2003-2004 academic year amounted to $16,153, and the corresponding figure for private schools was $32,588.[3] Adding $20,000 to $25,000 for living expenses, books, and equipment brings the estimated cost of four years of attendance to approximately $140,000 for public schools and $225,000 for private schools.

2) The continuing increases in tuition are the primary reason why a medical education is less affordable today than it was two decades ago. In the 1984-1985 academic year, average annual tuition and fees were $3,877 at public medical schools and $12,973 at private schools. Thus, in 19 years, the costs increased by 317 percent and 151 percent, respectively.[3] Despite low inflation during the past several years, this escalation continued. Recent shortfalls in state budgets have caused major budget crises at most public medical schools, resulting in a need for marked increases in tuition, which went up by 11.9 percent in 2003 and 17.7 percent in 2004. Private medical schools, though not as reliant on state money, raised tuition by 4.4 percent in 2003 and 5.7 percent in 2004 --increases well above the rate of inflation.

3) Accompanying this continuing escalation, there has been an enormous increase in the average indebtedness of graduating students. The average debt carried by 1984 graduates was $22,000 for public school and $26,500 for private school. By 2004, the debt had increased to $105,000 for public school and $140,000 for private school, and only about 20 percent of medical students graduated with no debt.[4]

4) Although the consumer price index is less than twice what it was 20 years ago, medical-student debt was 4.5 times as high in 2003 as it was in 1984. Tuition, however, has increased at only 2.7 times the rate of the consumer price index at private medical schools and at 3.8 times that rate at public schools. The greater increase in student debt suggests that medical students have chosen to borrow money not only for tuition but also for other costs associated with attending medical school.

5) Since more students now live in apartments rather than dormitories, are married or marry during medical school, have children, need to buy cars for travel to affiliated hospitals and outpatient facilities, and buy computers and other electronic equipment, these other costs have increased dramatically. Most students, even if they come from affluent families, prefer to take out low-cost educational loans to pay for their expenses. Moreover, unlike law or business students, who enter the workforce immediately after graduation and can begin to pay off their debt, the average graduating medical student spends an additional three to six years in postgraduate training programs while interest continues to accrue.

6) All of this means that a 2003 graduate with $100,000 in debt who begins repayment after a three-year residency will generally pay $15,000 per year for 10 years. Consolidating the debt and extending repayment over a period of 25 years will result in payments of $12,000 per year for a quarter-century. In 2003, the consolidated interest rate for Stafford loans was only 2.82 percent. When interest rates reach the maximum allowable rate of 8.25 percent, as they have been known to do, or when students need to borrow additional money from private sources, repayments can well exceed these estimates.[5]

7) What have been the effects of these increases? Although there has been a decrease in applications to medical schools from about 47,000 in 1996 to 35,700 in 2004, there are still two applicants for each of the 16,000-plus yearly medical-student positions. In 2003, the average salary of a practicing physician ranged from about $146,000 for a family practice physician to a bit more than $400,000 for an invasive cardiologist.5 So far, physicians' salaries have allowed them to pay off their debt, generally without defaulting on their loans.

References (abridged):

1. Petersdorf RG. Financing medical education. Acad Med 1991;66:61-65

2. A word from the president: you owe how much? Vol. 11. No. 10. AAMC Reporter. July 2002. Washington, D.C.: Association of American Medical Colleges.

3. Jolly P. Medical school tuition and young physician indebtedness. Washington, D.C.: Association of American Medical Colleges, March 23, 2004

4. Division of Medical Education. Medical School Graduation Questionnaire 2003: all schools report. Washington, D.C.

5. Association of American Medical Colleges, 2003. Salaries for physicians. Washington, D.C.: Bureau of Labor Statistics, 2004

New Engl. J. Med. http://www.nejm.org

--------------------------------

Related Material:

ALARM OVER DECLINE IN NUMBER OF US PHYSICIAN-SCIENTISTS

Notes by ScienceWeek:

In general, the term "physician-scientist" refers to MDs who devote all or a majority of their professional effort to research concerning health and disease. The designation thus includes basic, disease-oriented, patient-oriented, population-oriented, and prevention-oriented investigators. In the US, there is apparently a current great concern about an evident "dangerous" decline in the number of physician-scientists.

The following points are made by Leon E. Rosenberg (Science 1999 283:331):

1) The decline is not a new problem, and was already discussed 20 years ago. The problem has been repeatedly discussed during the past 2 decades, particularly with respect to patient-oriented research.

2) The author suggests the problem must be addressed now for a number of reasons: a) The entire species of physician-scientist is apparently at risk, not only those doing patient-oriented research. b) Endangering physician-scientists endangers everyone concerned with medical research. c) The actions taken to date cannot solve the problem. d) The threat can be averted only by bold, concerted action on the part of all the participants in the US medical research enterprise.

3) The number of first-time MD applicants for National Institutes of Health (NIH) research project grants decreased 31 percent from 1994 to 1997. If this progression continues linearly, there will be zero applicants by 2003. (The drop in first-time MD applicants was not compensated by an increase in applications from MD/PhD applicants.)... The data indicate that progressively fewer young MDs are interested in (or perhaps prepared for) careers as independent NIH-supported investigators.

4) Since 1992, there has been a 51 percent decrease in the total number of MD postdoctoral trainees supported by NIH through individual fellowships and training grants. If this trend continues, there will be no MDs in this pool by 2006.

5) Recent data from the Howard Hughes Medical Institute (HHMI) indicate that in the past 2 years there has been a 57 percent decrease in the number of MDs applying for the prestigious HHMI postdoctoral fellowships.

6) In 1989, 14 percent of graduating medical students expressed a strong interest in research as a career; in 1996, the number was 10 percent. Concerning disincentives, the author points out that the debt burden for US medical school graduates now averages US$80,000, a debt difficult to manage via the modest stipends paid to postdoctoral trainees. The author concludes: "We must act now to create a national environment conducive to creating a new generation of physician-scientists who have been trained rigorously and are confident in their ability to compete and succeed. Above all, these young investigators must be imbued with the belief that their efforts are essential."

Science http://www.sciencemag.org

--------------------------------

Related Material:

ON THE EVOLUTION OF MEDICAL SCHOOL ADMISSIONS TESTING

The following points are made by William C. McGaghie (J. Am. Med. Assoc. 2002 288:1085):

1) The origins of the Medical College Admission Test (MCAT) coincide with the rise of scientific psychology and quantitative approaches to mental measurement in the late 19th and early 20th centuries. The intellectual foundation of this work has been traced to ideas of Charles Darwin (1809-1882) expressed in his _On the Origin of Species_ (first published in 1859) concerning individual variation and natural selection. Sir Francis Galton (1822-1911), Darwin's cousin, subsequently attempted to measure human variation in intellectual abilities using a variety of laboratory tests. At the turn of the 20th century Alfred Binet (1857-1911) and Theodore Simon created early versions of intelligence tests, while E. L. Thorndike (1874-1949) developed several achievement tests and interpreted their results, in part, using the Karl Pearson (1857-1936) newly developed correlation coefficient. Early individual tests of intelligence and school achievement were quickly adapted for group administration in service of efficiency and extensive use.(1)

2) The technology of intelligence and achievement testing of school children was soon applied to other educational and social measurement problems. Psychological historian Murray Levine(1) observes:

"It is likely that the experience in group testing in the schools before World War I prepared the way for the group intelligence tests developed by Army psychologists for the selection of officers and for screening out those who were intellectually inadequate for service. It was after the war that group tests came into wide use and became firmly entrenched in the schools and in our culture."

3) Scholastic aptitude has been defined as a fixed trait. Furthermore, there are individual differences in knowledge or skills that contribute to educational achievement. Scholastic aptitude emphasizes verbal, quantitative, and frequently subject-centered (eg, biology, chemistry, physics) knowledge and skill that enhance educational success. Operationally, academic aptitude is similar, if not identical, to common conceptions of intelligence. From their inception, intelligence, achievement, and aptitude tests have been judged by the extent to which, when given to a group, they yield a normal distribution of scores that clearly separates high, medium, and low performers. Reliable score differences allow for objective (scientific) norm-referenced selection, classification, or placement decisions about individuals.

4) In summary: The attrition rate of 5% to 50% from US medical schools in the 1920s propelled the development of a test that would measure aptitude for medical studies. Since its development in 1928, the Medical College Admission Test (MCAT) has undergone 5 revisions. The first version was divided into 6 to 8 subtests that focused on memory, knowledge of scientific terminology, reading and comprehension, and logic. The second, which appeared in 1946, was reduced to 4 categories: verbal and quantitative skills, science knowledge, and added a category called understanding modern society. The major difference in the third version, launched in 1962, expanded the test's understanding modern society section to a broader test of general information. In 1977, the MCAT underwent its fourth change: its science section, reading and quantitative skill assessment sections were expanded; its general liberal arts knowledge section was eliminated; its scoring report structure and scoring range were altered; and its cultural and social bias was minimized. The current version, beginning in 1991, has undergone another significant change. Although it does not contain independent measures of either liberal arts or numeracy as separate categories, quantitative skills are needed to solve some of the problems in biological and physical sciences. However, its principal innovation is the writing sample section. Through its 74-year history, the various renditions of the MCAT demonstrate that the definition of aptitude for medical education reflects the professional and social mores and values of the time.

References (abridged):

1. Monroe WS. An Introduction to the Theory of Educational Measurements. Boston, Mass: Houghton Mifflin; 1923. Cited by: Levine M. The academic achievement test. Am Psychol. 1976;31:228-238.

2. Barzansky B, Etzel SI. Educational programs in US medical schools. JAMA. 2001;286:1049-1055.

3. Flexner A. Medical Education in the United States and Canada. New York, NY: Carnegie Foundation for the Advancement of Teaching; 1910. Bulletin No. 4.

4. Ludmerer KM. Learning to Heal. Baltimore, Md: Johns Hopkins University Press; 1985.

5. Scholastic Aptitude Test for Medical Schools, Form 15, First Edition. 1941; Hollinger Box No. 1; IM Box No. 501982734. Located at: Association of American Medical Colleges.

J. Am. Med. Assoc. http://www.jama.com

ScienceWeek http://scienceweek.com