Footnotes

[2] The education, training, credentialing, and licensing of MDs and DOs is similar. The main difference between the two degrees is the DO emphasis on the musculoskeletal system and how an injury or illness in one area can affect another.

[3] The AMA defines “active” as working more than 20 hours per week in professional activities. The estimates provided in this paper include only physicians under age 75.

[4] The PWS was conducted by The Sheps Center at the University of North Carolina on behalf of BHPr’s National Center for Health Workforce Analysis. Estimates of physician retirement rates were obtained via personal correspondence with Bob Konrad, principal investigator for the PWS.

[5] The CPW combines physicians, lawyers, accountants, architects, and other licensed professionals into an occupation entitled licensed professionals, and we estimate retirement rates for this group as a proxy for physician retirement patterns.

[6] The Congressional Budget Office (CBO) projects a 3% annual growth rate in real Gross Domestic Product (GDP) between 2003 to 2013, which is about 2% average annual growth in real GDP per capita. Real economic growth, controlling for changing demographics, occurs through an increase in productivity. CBO projections, therefore, assume that worker productivity will increase by approximately 2% annually, on average, throughout the economy. Physician productivity will likely increase less rapidly than overall productivity in the United States due to the labor intensiveness of physician services.

[7] More detailed supply projections are reported in The Physician Workforce: Projections and Research into Current Issues Affecting Supply and Demand (BHPr, 2006).

[8] To obtain hours worked per week by patient care specialty for the FTE conversion, we regressed the log of total hours worked per week (by post residency patient care MDs by specialty) on age variables, sex , and country of medical education (USMG,IMG) using 1998 data from the American Medical Association's Patient Care Physician Survey. The data contains estimates for 13 specialty categories: general/family practice, general internal medicine, medical subspecialties, general surgery, surgery subspecialties, general pediatrics, obstetrics/gynecology, radiology, psychiatry, anesthesiology, pathology, emergency medicine, and "other" specialties. Data for these specialties were mapped into the 37 specialties projected in the PSM. FTEs are defined to be equal to head counts in base year 2000, and thus for each specialty and physician type (USMG or IMG) the number of FTEs equals the head count in the base year. For each projection year, the number of physicians projected for each combination of physician type, specialty, sex and age is multiplied by the expected hours worked for the appropriate combination, and the sum of the products by specialty and physician type is divided by the baseline FTE definition in terms of hours worked per week for each specialty to produce projections of FTE physicians by year, physician type, and specialty.

[9] Alternative approaches described in the literature to estimate physician requirements include a needs-based approach and use of benchmarking (i.e., a specific form of the utilization-based approach). The needs-based approach defines physician requirements based on a clinical assessment of prevalence rates for medical problems and the amount of time physicians need per patient encounter. This approach has been criticized because it ignores the economic realities that influence use rates. The benchmarking approach was used extensively in the 1990s by applying HMO physician-to-enrollee estimates to the United States population under a scenario with projected growth in managed care enrollment.

[10] The eight categories are ages 0–4, 5–17, 18–24, 25–44, 45–64, 65–74, 75–84, and 85 and older.

[11] As with the physician supply estimate, this count uses AMA and AOA Masterfile data on physicians’ activity status for physicians younger than age 75.

[12] For the three insured categories, the PRM further distinguishes between private health insurers and government-sponsored insurance plans for a total of seven insurance categories.

[13] Differences in healthcare systems make comparing use of physician services difficult. Also, measuring GDP and other measures of economic wellbeing across countries is an inexact science.

[14] Over the past 20 years, the percentage of total Federal and nonfederal physicians engaged primarily in non-patient care activities has steadily declined from around 9% to its current level of about 6%.

[15] Specialties hypothesized to be in this low-sensitivity category include general and family practice, general internal medicine, pediatrics, obstetrics/gynecology, and emergency medicine.

[16] Specialties hypothesized to be in this medium-sensitivity category include cardiology, internal medicine subspecialties, general surgery, otolaryngology, urology, anesthesiology, radiology, pathology, and “other” specialties.

[17] Specialties hypothesized to be in this high-sensitivity category include orthopedic surgery, ophthalmology, “other” surgery, and psychiatry.