The Medical College Admission Test score of a school is strongly related to physician distribution. The medical schools that have the highest MCAT scores graduate the students that have the lowest levels of distribution both geographically and socioeconomically. Higher MCAT scores are most associated with urban locations and subspecialty choice. Higher MCAT scores also mean fewer choosing family medicine, one of the key distributional indicators of physicians since family physicians are the only physician specialty that distributes geographically like the US population and also has the best socioeconomic distribution to poor, less educated (Mold, Graham Center), lower income primary care, whole county shortage areas, high poverty zip codes, and small and isolated rural areas. Family medicine has the best choice of generalist office-based careers and urban low income or high poverty practice locations, especially when considering those away from zip codes with a hundred or more physicians Birth Origins Articles
The figures above represent the career choices of 1987 - 1999 graduates of allopathic medical schools at their 2005 practice locations. Each MCAT category represents about 20 schools. Duluth and Mercer have outstanding admissions of students who are selected for distributional qualities. UC Davis, US Irvine, UCLA, and the University of Washington demonstrate similar type admissions/curricula for office primary care in poverty locations. Other schools appear to have distribution primarily based on the student types admitted and the practice environment of the surrounding state or states nearest the medical school.
The correlations for typical allopathic medical school MCATs ranged from 0.4 to 0.7 for each of the distributional careers. MCAT Correlations The correlations remained strong despite controls for population income and rurality, longitude, and state education levels. The correlations remained stead even when removing outliers to obtain a more normal distribution.
Similar graphics can be obtained by examining student income levels, student birth origins, or other socioeconomic factors.
Ranking students based on MCAT scores is likely to result in more and more student admitted who are less likely to distribute. Such a use denies students who have never had a level playing field throughout their lives from having even a chance. The MCAT can predict the group of students likely to have problems academically, but it cannot predict the individual students who will have problems or fail. The same is true of the distributional careers noted above. Birth origin characteristics and MCAT scores can be used to predict the types of students most likely to distribute, but cannot predict individual student distribution.
The MCAT is incapable of long term performance measurement, even attempts to extend relevance beyond the first 2 years of medical school are a stretch. The white students, with greater percentages of older and rural born students, are the most likely to demonstrate better USMLE 1 scores and basic science performance. The Asian students, with the greatest percentage of younger and urban born students, are the most likely to demonstrate lower USMLE 1 scores compared to MCAT based predictions. It is likely that this is not white or Asian, but socioeconomic in nature. Veloski, JJ Prediction of Student's Performances
Using the MCAT as a threshold to determine those who are acceptable candidates for admission, and then throwing out the MCAT from further consideration would be a better method of focusing on the characteristics of physicians for
There is no evidence that selections based on the MCAT improves any of the above and some evidence that higher MCAT scores can be a problem area in some. The students that have overcome significant income and education obstacles to gain admission may well have advantages in all of the above areas.
Further examinations of individual student income and MCAT score are indicated, but are unlikely given the potential consequences of the findings.
Robert C. Bowman, M.D.
[email protected]
Career Predictors and the MCAT
Physician Workforce Studies
Physician Distribution in the United States