Robert C. Bowman, M.D.
Latest at Physician Workforce Studies
Family practice is the specialty most associated with access to health care, improved health care costs, and increased health care quality (Phillips Starfield, Baicker Chandra). Students choosing family medicine often come from rural (Boulger, Rabinowitz). or inner city origins and have made their career decision earlier, a decision shaped by personal values and early role models (Xu) The selection process in admissions may therefore be more critical for family medicine choice.
The decline in the family medicine match coincides with changes in the admissions of students most likely to choose family medicine. These include declines in students born in the same state as their medical school and students born in rural areas.
According to Association of American Medical Colleges annual reports, medical students of rural background have decreased from 27 % to 16 % between from 1983 to 1999. Birth origin studies involving 600,000 US physicians confirm that medical students with the most urban origins chose family medicine only 13 % of the time; those in urban areas outside of core metro areas, 16 – 19%; students born in the largest rural areas, 20.3%; students from medium-sized rural areas, 24.6%; and those from isolated rural areas, 26.1%. The impact of birth origin is consistent throughout the past 30 years, across states and medical school types, and regardless of rural-urban coding scheme.
The average decline in rural born medical students in all US medical schools was 47% from 1976 to 2000. All types of medical schools posted declines, including public and private, allopathic and osteopathic, those with a rural mission, those in the most rural states, and even the 47 medical schools (Barzansky) who claimed special admissions preference for rural background students
Students born in the same state as their medical school also chose FP more often. Over the past 30 years 11.7 % of out-of-state medical students chose family medicine as compared to 16.4% of those born in-state. In-state birth origin admissions peaked in 1981 with 8303 or 49 % medical students, declining to 6300 or 35% in recent years. Out-of-state medical students have continued to climb to over 11000 or 65% of recent graduating classes. Students attending medical school can meet in-state residency requirements in only a few months in some states. More conservative in-state residency requirements could improve student choice of family medicine as a career and would stimulate medical schools to become more involved in improving state and local education quality.
Regression studies help identify the contributions of key variables regarding medical school characteristics and family medicine choice. Regressions explained over 80 % of the variance. Medical schools with an average Medical College Admission Test (MCAT) score 1 point lower (example of 9.5 compared to 10.5) graduate 2 - 3 percentage points more family physicians. This means 3 - 6 more students choosing family medicine per year per school. The MCAT as currently utilized may also screen out other primary care physicians and obstetricians. MCAT scores have continued to rise and family medicine choice has declined. Schools with a higher percentage of rural students, students over age 30, and students born in the same state graduate more family physicians. Other variables linked to FP choice are longitude (west is best), schools with family medicine departments, those with a rural mission or rural medical education person, and those with increased state education opportunity as measured by the percentage of adults continuing into college after high school graduation
Age and Choice of Family Medicine
Students born in rural areas gained entry to medical school half as often as those of urban birth. States such as SD, NE, NM, ID, WV, MT, KS, MN, AK, AR, and HI have higher rates of rural birth admission. Such states tend to have better education opportunity, higher per capita education expenditures, and specific efforts to link rural education and medical education. Regressions involving rural birth admissions to medical school note that state education opportunity and special admissions tracks explain 44% of the variance. Similar regressions involving other specialties did not link them significantly to state education.
Expansions of medical schools and medical school class sizes could improve the probability of admission of rural born students, but not without specific measures such as special tracks for rural students. Past expansions 25 years ago paradoxically decreased the rural probability of admission. Prior to expansion was the best probability for rural admission at 70% compared to urban students, decreasing to less than half that of urban students during the expansion years, then a slow improvement to just above 50% in the past decade, coincident with the improvements in admission of underrepresented minority students.
Without a foundation of admissions, there seems to be little point in special medical school or residency programs involving family medicine. Indicators linked to family medicine and primary care such as birth origin, in-state admission, state educational opportunity, and use of the MCAT suggest that a comprehensive solution to the nation's physician workforce problems must involve the integration of education and health professions education.
Regardless of policy, research, or probability, medical school admissions involve one decision at a time for each candidate. Current pressures seem to have shifted choices away from students that are more likely to make a decision for family medicine. Setting the bar too high means exclusion of the doctors most needed across the nation. Setting the bar too low means that more students will fail, with great trauma to them, the school, and others. Schools desiring to fulfill missions for rural health and diversity must make an effort beyond current evaluations.
Although national studies are helpful, medical schools and states and students vary far too much. For the nation to have an affordable, accessible, quality physician workforce, admissions committees must involve detailed studies of these and other variables, individualized to the particular medical school and relevant to local, state, and regional needs. The nation must also step forward to hold medical schools accountable for workforce.
Phillips RL, Starfield B, Why Does a U.S. Primary Care Physician Workforce Crisis Matter? Am Fam Physician 2003;68:1494, http://www.aafp.org/afp/20031015/editorials.html
Baicker K, Chandra A, Medicare Spending, The Physician Workforce, And Beneficiaries Quality of Care, Health Affairs W4 184-197 April 2004 http://content.healthaffairs.org/cgi/reprint/hlthaff.w4.184v1.pdf, accessed 8/10/2004
Xu G, Veloski JJ, Barzansky B, Hojat M, Diamond J, Silenzio VMB Comparisons Among Three Types of Generalist Physicians: Personal Characteristics, Medical School Experiences, Financial Aid, and Other Factors Influencing Career Choice Advances in Health Sciences Education 1996, vol. 1, no. 3, pp. 197-207
Boulger, JG. (1991). Family medicine education and rural health: A response to present and future needs. The Journal of Rural Health, 7(2), 105-115.
Rabinowitz HK, Diamond JJ, Markham FW, Paynter NP, Critical factors for designing programs to increase the supply and retention of rural primary care physicians, JAMA 2001; 286: 1041-1048
Barzansky B, Jonas HS, Etzel SI Educational Programs in US Medical Schools 1996 - 1997, JAMA 1997;278:744-749
Cost, Quality, Access, and Physician Workforce Expansion
Probability of admission tables