A Census of Actively Licensed Physicians in the United States, 2012

Over the past 150 years, state medical licensing and regulatory boards in the United States have steadily evolved from entities that simply issued medical licenses — based on minimal qualifications that at one time did not include a high school diploma prior to admission into medical school — to multi-faceted and multi-staffed authorities responsible for protecting the public by granting licenses to only the most qualified individuals and ensuring that disciplinary and competency standards are upheld.2 Today, each of the 70 state and territorial medical boards in the United States are governed by statutes and regulations in a Medical Practice Act that establishes a board's legal rights and responsibilities in service to the public.

Because an active license is required to legally practice medicine, and physicians sometimes have more than one license, accurate information about a physician's credentials and licensure status has always been crucial to state medical boards to enable them to monitor a physician's practice, protect the public and promote quality health care. Accurate and current aggregate information about physicians' licensure status and credentials is also of critical value to state and federal policymakers interested in health care workforce assessments, predictions and planning.

This article provides a summary, analysis and discussion of the most recent physician licensure data, collected in 2012 from each of the state medical boards in the United States and the District of Columbia by the Federation of State Medical Boards (FSMB). This is the second such census of actively licensed physicians, following one that was published two years ago,3 and contains expanded state-specific information and additional data sets. In aggregate, the information offers a useful and current snapshot for health care workforce determinations of the number, gender, age, American Board of Medical Specialties (ABMS) certification and location by state of all physicians with an active license to practice medicine.

The FSMB maintains a comprehensive, central repository of data from every state medical board in the United States that contains comprehensive biographical, educational and disciplinary information about all licensed allopathic (MD) and osteopathic (DO) physicians. The repository is unique in that it is the only national database that contains the most current information about which jurisdictions have granted physicians a license, or renewal of a license, to practice medicine. The complete database, known since its development in 2004 as the Federation Physician Data Center (FPDC), is continuously updated and currently contains more than 1.7 million physician records, including information about physicians who are currently licensed, no longer licensed or deceased. To obtain an accurate count and precise information about physicians with an active, current license to practice medicine, we conducted a census using the most recent data obtained by the FPDC during the 2012 calendar year.

Licensure data is continuously provided throughout the year to the FPDC by the 51 state medical boards (which regulate both allopathic and osteopathic physicians) and 14 state osteopathic boards (which only regulate osteopathic physicians) in the United States and the District of Columbia. These state boards are specifically authorized by each of their Medical Practice Acts to license and discipline physicians. Four additional territorial medical boards (Guam, U.S. Virgin Islands, Commonwealth of Northern Mariana Islands and Puerto Rico) are also member boards of the FSMB but their physician data was excluded from the current analysis (just as it was in 2010) because current data from these jurisdictions was not available. Because of their differing capacities and resources, state boards submit information to the FPDC at varying intervals throughout the year. The majority (85%) of state boards provide medical licensure information to the FPDC on a monthly or quarterly basis, with some boards able to provide such data weekly or even daily (e.g., Maine Board of Licensure in Medicine, New York State Board for Medicine and the Oklahoma State Board of Medical Licensure and Supervision).

A physician record in the FPDC is typically initiated when a U.S. medical school student or an international medical graduate (IMG) first registers to take the United States Medical Licensing Examination (USMLE), a program created in 1992 that is co-sponsored by the Federation of State Medical Boards and the National Board of Medical Examiners and is required of U.S. and IMG allopathic physicians for licensure eligibility by state medical boards. For U.S. osteopathic medical students who do not register for the USMLE^* and for physicians who were first licensed prior to the introduction of the USMLE and the Comprehensive Osteopathic Medical Licensure Examination (COMLEX-USA) in the early 1990s, licensure files from state boards serve as the initial FPDC record and as the primary source for a physician's record of successful completion of a licensure examination, which may include older examinations such as the National Board of Medical Examiners (NBME), the National Board of Osteopathic Medical Examiners (NBOME) or the Federation Licensure Examination (FLEX).

When the FPDC receives additional physician data, each record is matched to a master physician identity table using a set of algorithms developed by the FSMB. The five data elements used for matching the information to ensure that it is accurate include name, date of birth, Social Security number, medical school name and medical school graduation year. If there is a physician record match with three of the five data elements, the information is automatically entered. When there are fewer than three data elements available for matching a new physician record, the record is sent to the FSMB's Data Integration department for manual review. More than 90% of physician records received in files from state boards are matched automatically. This systematic process also allows the FSMB to track the same physician across multiple jurisdictions if more than one state license is sought at any time during his or her professional career.

Though physicians in the United States are not licensed based on their specialty or practice focus, and specialty board certification is not a requirement for medical licensure, the FPDC receives and supplements licensure data provided by state boards with specialty and subspecialty certification information obtained from the American Board of Medical Specialties (ABMS).^** The ABMS represents 24 independent specialty boards^† that certify physicians in more than 145 specialties and subspecialties of medicine and surgery.4 Deceased physicians are also identified and flagged in the FPDC by cross-referencing physician records with the Death Master File of the Social Security Administration (SSA), a federal database that contains more than 86 million records of reported deaths.

An analysis of data collected in 2012 about physicians in the United States and the District of Columbia reveals that there were 878,194 physicians with an active license to practice medicine, representing a net increase since 2010 of 28,109 (3%) physicians. State medical boards issued 134,456 new licenses to physicians since the FSMB's 2010 physician census, a figure which includes physicians obtaining their first license, one or more additional licenses (enabling practice in multiple jurisdictions) or a new license when moving from one jurisdiction to another. Between 2010 and 2012, 48,219 physicians received their first medical license from a state medical board.

Actively licensed allopathic physicians represent the vast majority (93%) of the licensed physician population in the United States, while actively licensed osteopathic physicians account for 7%, figures essentially unchanged from 2010 (Table 1).

The osteopathic medical profession continues to experience exponential growth in its numbers, however. From 2010 to 2012, the number of physicians with a DO degree and an active license increased by 8%, compared to a 3% increase in the number of licensed physicians with an MD degree.

The actively licensed physicians identified in our 2012 census graduated from a total of 1,881 medical schools located in 166 countries around the world. Nearly 76% of physicians graduated from a U.S. or Canadian medical school (allopathic or osteopathic), 22% graduated from a medical school outside the United States and its territories or Canada, and for 2% the medical school of graduation could not be determined. Table 2 shows the ten allopathic and osteopathic medical schools that had the largest number of graduates with an active license to practice medicine in the United States. Table 3 provides a listing of the ten medical schools outside the United States that had the largest number of graduates with an active license to practice medicine in the United States.

Of the 196,573 IMGs who graduated outside the United States or Canada, more graduated from medical schools in India (45,558 or 23%) than any other country in the world. IMGs from schools in the Philippines were a distant second, at (14,785 or 8%), while Pakistan (10,880 or 6%), and Mexico (10,012 or 5%), ranked third and fourth, respectively. Our census revealed a substantive increase since 2010 in the number of actively licensed IMGs who graduated from a medical school in the Caribbean.^†† In 2012, 13% (25,726) of IMGs with an active license to practice medicine in the U.S. graduated from a medical school in the Caribbean (Figure 1). While the total number of IMGs with an active license in the United States increased by only 4% since 2010, the number of physicians who graduated from a school in the Caribbean increased during that time period by 14%.

The age composition of the actively licensed physician population reflects the gradual, but significant and certain, shift seen in the general population. The average age for physicians with an active license to practice in 2012 was 51 years, up from 50 years in 2010. Furthermore, the actively licensed physician population grew at a faster rate in the older age groups than in the younger. In 2012, 26% of actively licensed physicians were 60 years of age or older, compared with 24% in 2010 (Figure 2); this represents a two-year increase of 11% compared with less than a 1% increase for the number of physicians 49 years of age or younger.

As with age, a shift in the gender makeup continued from 2010 to 2012. Though two-thirds of actively licensed physicians were male in 2012, the number of female physicians with an active license increased by 8% compared with only a 2% increase for male physicians. In 2010, females represented 29% of licensed physicians in the United States, and by 2012 they comprised 30% of the population. In 2012, the average age for female physicians was 46 years compared with 52 years for males. Furthermore, 34% of female physicians were 39 years of age or younger, compared to only 18% of male physicians (Figure 3).

Overall, 77% of physicians with an active license to practice medicine were certified by at least one ABMS specialty board in 2012, up from 75% in 2010. While 80% of MDs and 39% of DOs were found to hold ABMS certification, the count is considerably lower for DOs because many osteopathic physicians achieve specialty board certification, in addition to or in place of ABMS certification, through the 18 specialty boards of the American Osteopathic Association's (AOA) Bureau of Osteopathic Specialists: a total of 23,819 DOs (38% of actively licensed osteopathic physicians) have AOA BOS certification, according to data reported by the AOA in 2012. There was a difference in ABMS certification rates based on country of medical graduation. U.S. and Canadian medical school graduates (both MDs and DOs) were more likely than IMGs to have ABMS specialty certification (78% vs. 73% in 2012). However, IMGs have narrowed the gap slightly over the past two years (a five-percentage-point difference now, compared with a seven-point difference in 2010).2 A strong and predictable relationship continues to exist between specialty certification status and age. The percentage of physicians who are ABMS certified appears to increase dramatically for physicians aged 30 to 39, peaks for physicians aged 40 to 49 and then begins to diminish during the 60–69 age range and beyond (Figure 4).

The data about the number of active medical licenses maintained by physicians have essentially remained static from 2010 to 2012. Seventy-eight percent of physicians held only one active license to practice medicine from a state medical board, 16% had active licenses in two jurisdictions and 6% had active licenses in three or more jurisdictions. Twenty-four percent of male physicians, compared with 19% of female physicians, had a license to practice medicine in more than one jurisdiction. Physicians with specialty certification from an ABMS board were more likely to have two or more active licenses (23%) than physicians without ABMS certification (17%).

Analyses by state, and within the nine geographic divisions of the United States as defined by the U.S. Census Bureau (Figure 5), were used to illustrate the location of actively licensed physicians in the United States during 2012. Similar to 2010, the South Atlantic, Pacific, Middle Atlantic and East North Central divisions accounted for a little more than two-thirds of the actively licensed physicians in 2012 (Figure 6). Some areas experienced growth in their physician populations, while others saw declines. The number of physicians in the South Atlantic, Pacific and West South Central divisions increased by 9% or more from 2010 to 2012, for instance, while the New England and West North Central divisions experienced decreases of more than 5% (Table 4).

The 878,194 physicians with an active license to practice medicine represent a physician-to-population ratio of 280 actively licensed physicians for every 100,000 people in the United States and Washington, D.C. The 2012 state-by-state totals in Table 5 provide additional information on the geographic breakdown of physicians with an active license to practice medicine in the United States.

Our census results show that the actively licensed physician population in the United States since 2010 is steadily growing in total number, slightly older, with more women and with more diversity in terms of medical school of graduation, with a substantive increase in licensed physicians who graduated from a medical school in the Caribbean. Though the vast majority of the 878,194 physicians with an active license continued to hold an MD degree in 2012, the number of physicians holding an osteopathic medical degree grew at a faster rate in our census, consistent with data reported elsewhere that one in five medical students in the U.S. is now enrolled in an osteopathic medical school.6 The overall percentage of actively licensed physicians graduating from international medical schools remained steady from 2010 to 2012. The undergraduate medical education programs in the Caribbean continue to expand,7 however, and the number of IMGs with an active license who graduated from a medical school in the Caribbean, many of whom are U.S. citizens,8 grew at a faster pace (14%) from 2010 to 2012 than the number of actively licensed physicians who graduated from all other international medical schools (4%).

Beyond the changes by degree and medical school type, gradual, but significant, shifts in the age and gender composition of the actively licensed physician population also continued and will likely have a substantial impact on the U.S. health care system in the years to come. Our census indicates that 26% of actively licensed physicians in the U.S. are 60 years of age or older, signaling an actuarial need for increasing the supply of physicians as older physicians retire from practice or become deceased. Our findings also indicate that nearly one-third of actively licensed physicians are women, and the average age for female physicians (46 years) is considerably younger compared with their male counterparts (52 years). With women making up nearly half of all enrolled U.S. medical and osteopathic medical students during the 2011–2012 academic year,9,10 this trend will likely continue among those who become actively licensed in the years ahead.

The aging of the actively licensed physician population and the shift in gender composition could have a considerable effect on health workforce determinations considering the different work patterns ascribed, even in contemporary studies, to both male and female physicians. Research conducted with data from the U.S. Census Bureau's Current Population Survey (CPS) indicates that male physicians are generally less active than their female counterparts in the latter part of their career (55 to 64 years) while female physicians are less active in the early stages of their career (25 to 34 years).11 A recent nationwide survey of 13,575 physicians indicated that 22% of physicians above the age of 40 work 40 hours a week or fewer, compared with 15% of physicians who are age 40 or younger. A breakdown by gender in that survey showed 27% of female physicians working 40 hours a week or less, compared to 18% percent of male physicians.12 In a similar manner, an analysis of data from the New York State Survey of Residents Completing Training concluded that “the growing number of female physicians will probably create a new set of provider preferences that includes more predictable schedules and less time pressures on other aspects of life.”13 

Like actively licensed physicians, the general population in the United States is projected to undergo extensive and steady growth in the older age cohorts. The number of Americans aged 65 and older is projected to more than double to 88 million by 2050.14 Baby boomers — those individuals born between 1946 and 1964 — started turning 65 years old in 2011 and will be largely responsible for the increase as an estimated 10,000 boomers reach age 65 each day for the next 17 years.15 All baby boomers will be 65 years of age or older by 2030 and are expected to represent nearly 20% of the total U.S. population.16 Moreover, data on health care expenditures indicates that Americans 65 years and older use more health care per capita than any other age group.17 A report published by the Institute of Medicine (IOM) confirms this finding, showing that those aged 65 years and over represented only 12% of the U.S. population in 2008 but account for 26% of all physician office visits, 34% of all prescriptions, 35% of all hospital stays, and 38% of the emergency medical services responses.18 Those percentages are almost certain to increase as baby boomers age in the next two decades.

The full impact of the Patient Protection and Affordable Care Act (PPACA), enacted in 2010 and upheld by the U.S. Supreme Court in 2012,19 is yet to be realized, but it is safe to assume that the law represents a historic, comprehensive approach to health care reform that is expected to touch nearly every aspect of the U.S. health care system. The law predominantly addresses what is held to be a societal need to extend health care coverage to more individuals, improve health system quality and efficiency, promote health and prevent disease, limit fraud and abuse, provide more availability for long-term care and increase the health care workforce, particularly in primary care.20 The ability of young adults to enroll in a parent's health plan up to the age of 26 is already in place, having gone into effect in September of 2010. On the horizon are significant expansions in eligibility for Medicaid and the implementation of state-based health insurance exchanges beginning in 2014.21 While the outcome of negotiations between the federal government and the states over procedural aspects of the new law (and coverage for various health care services) will play out in the months ahead, there is little argument that health care reform will greatly expand coverage in the coming years and that as many as an additional 32 million Americans may become insured by 2019.22 

An aging population and millions of Americans becoming eligible for health insurance highlight the importance of health workforce data determinations and predictions. Even before the Affordable Care Act was passed, demand for physicians in the U.S. was on the rise with nearly three out of four hospitals planning to increase physician employment.23 This phenomena continues to bolster the already decades-long swing to more physicians working in hospitals, integrated delivery systems or large group practices rather than as solo practitioners.24 

According to the Health Resources and Services Administration (HRSA) of the U.S. Department of Health and Human Services, there are 5,864 Primary Care Health Professional Shortage Areas (HPSAs) with 57 million people living in them,25 and the most recent projections suggest national physician shortages as high as 130,000 by 2025.26 Similarly, a growing number of specialty-specific and state-specific studies have concluded that the physician workforce is facing current or future shortages.27 For example, there were 7,356 specialty-certified geriatricians in the U.S. in 2012, one for every 2,551 Americans 75 years of age or older, but few physicians appear to be choosing to pursue specialty training in geriatrics; by 2030 the ratio is expected to drop to 1 in 3,798.28 California is facing a striking physician shortage, as nearly 30% of physicians in that state near retirement age, the second highest percentage in the nation.29 Some lawmakers in the state are contemplating filling the gaps by redefining who may provide care and by expanding the scope of practice of non-physician health care providers.30 

The Association of American Medical Colleges contends that existing physician shortages will worsen without significant interventions.31 It proposes a supply-side solution that expands funded graduate medical education positions, increases utilization of non-physician health care providers and increases physician productivity through team-based care that encourages physicians and nonphysician clinicians to work with care coordinators and improve efficiencies through the use of electronic health records, online communications and practice-management innovations.

A minority view holds that the use of teams, information technology, better sharing of clinical data and the use of non-physician providers have the potential to offset the demand for physician services in the future.32 Where there is little argument is that the U.S. health care system is financially stressed and that the cost of health care will continue to rise, even with long-term benefits said to derive from the Affordable Care Act. Health care spending now constitutes 18% of the nation's gross domestic product (GDP) and is projected to increase to 21% by 2023.33 

Our findings in this article build upon baseline physician census data we reported in 2010, which continues to highlight the need for a better understanding and accurate assessment of the current supply of physicians. The number of actively licensed physicians and where they are licensed are important contributions to health workforce studies but these data elements provide a limited picture. At the FSMB's Annual Meeting last year in Fort Worth, Texas, its House of Delegates adopted as policy a recommendation that all state medical boards adopt a framework for a physician minimum data set (MDS) of questions for physicians to answer at the time of licensure renewal.34 The MDS for physicians was developed by an FSMB Working Group, chaired by Richard Whitehouse, Esq., of Ohio and more recently by Mark Eggen, MD, of Minnesota, that worked collaboratively with HRSA's National Center for Health Workforce Analysis, which is led by Edward Salsberg, MPA.

A state medical board's license renewal process is a unique opportunity for collecting additional, up-to-date workforce information from physicians. Implementing an MDS using a uniform, basic set of questions that all state boards can ask of actively licensed physicians will provide more robust census information that can be used to improve access and delivery, and reduce waste. The MDS will provide a window into physicians' clinical participation (including entry, retention, exit and re-entry to practice); a better understanding of the geographic distribution of health care delivery and physician migratory patterns; and a consistent body of information that can be tracked over time. As the MDS is implemented by state boards in the months and years ahead, a census of licensed physicians that includes information about how many physicians are actively engaged in the practice of medicine — whether full-time or part-time, in their area of primary residency or fellowship training or not, and where — could offer greater insight to state and federal policymakers, managed care organizations, physicians and the public as coordinated efforts are made across the country to deliver quality health care that is affordable, efficient and accessible.

the authors wish to thank Cyndi Streun, Frann Holmes, Bradley Dunn, Lucie Maomanivong, Martha Buchholz, Cassandra Irving, Amanda Johnson and Sheila Still for their valuable assistance in the preparation of this manuscript.