Physicians serve as the central decision makers in the American health care system, with their clinical judgments influencing how up to 90 percent of every health care dollar is allocated. As health spending continued to climb, the ranks of the uninsured expanded, and quality of care remained uneven, the need for reliable, current data on physician demographics and practice patterns grew more pressing. The Center for Studying Health System Change (HSC) 2008 Health Tracking Physician Survey, a nationally representative mail survey of U.S. physicians providing at least 20 hours per week of direct patient care, addressed that need. This Data Bulletin, authored by Ellyn Boukus, Alwyn Cassil, and Ann S. O'Malley and published in September 2009, presents key findings across multiple dimensions of physician practice, including demographics, practice organization, compensation structures, career satisfaction, patient acceptance patterns, managed care participation, and charity care provision.
The physician sample was drawn from the American Medical Association master file and included active, nonfederal, office-based and hospital-based physicians. Residents, fellows, radiologists, anesthesiologists, and pathologists were excluded. The survey collected responses from more than 4,700 physicians, achieving a response rate of 62 percent. Due to changes in survey administration from prior HSC physician surveys (including a shift from telephone to mail, altered question wording, and differences in sample structure), direct comparisons with earlier Community Tracking Study Physician Surveys are not appropriate. The 2008 survey instead establishes a new baseline for tracking how physicians organize and practice medicine.
In 2008, nearly one-third of physicians worked in solo or two-physician practices, 15 percent were in groups of three to five physicians, and 19 percent practiced in groups of six to 50. Thirteen percent were based in hospital settings, with 44 percent of hospital-based physicians working in office practices or clinics and the rest divided between emergency departments and hospital staff roles. Three percent practiced in community health centers (CHCs) and 4 percent in group or staff-model health maintenance organizations (HMOs).
Almost three-quarters of U.S. physicians were men in 2008. However, among physicians under age 40, more than 41 percent were women, signaling a likely shift in the gender composition of the physician workforce in coming years. Three out of four physicians identified as white, non-Hispanic, while 3.8 percent were Black, non-Hispanic, 5.3 percent were Hispanic, and 17.2 percent were Asian or of other racial backgrounds. Among physicians under 40, approximately two-thirds were white and one-third belonged to minority groups, including 4 percent Black, 5.4 percent Hispanic, and 24 percent Asian or other.
More than 80 percent of physicians surveyed worked full time. Over half (53 percent) were between 40 and 55 years old, and close to four in ten had been in practice for more than 20 years. Nine in ten were board certified, and 22 percent had received their medical training outside the United States or Canada. Roughly 40 percent were primary care physicians, 35.1 percent were medical specialists including psychiatrists, and 25.7 percent were surgeons including obstetrician/gynecologists. A slight majority (56 percent) were full or part owners of their practices, while 44 percent worked as employees or independent contractors.
The geographic distribution of physicians relative to the general population was uneven. Physicians were overrepresented in the Northeast and in large metropolitan statistical areas, likely reflecting the concentration of specialized services and academic medical centers in those regions. Conversely, physicians were underrepresented in the South. In 2008, 32.5 percent of physicians practiced in the Northeast compared with 18.1 percent of the general population, while 36.7 percent of physicians were located in the South compared with 22.4 percent of the population. By urban-rural status, 60.4 percent of physicians practiced in large metropolitan areas (population exceeding one million), compared with 54.7 percent of the general population. Only 10.3 percent of physicians were in non-metropolitan areas, while 16.4 percent of the general population lived in such regions.
In 2008, 44 percent of physicians reported receiving some form of performance-adjusted salary, such as a base salary with adjustments tied to their own productivity. About one-quarter indicated payment by fixed salary, and 20 percent received a share of practice revenue. Roughly one-third of physicians earned less than $150,000 annually, about one-third earned between $150,000 and $250,000, and 30 percent earned $250,000 or more. Slightly more than 45 percent of physicians had bonuses available to them.
Productivity measures and overall practice financial performance were the most common factors influencing physician compensation. Approximately 61 percent of physicians reported that these factors were moderately or very important in determining their pay. By contrast, fewer than one in ten physicians said that practice profiling, which compares a physician's resource utilization patterns to those of peers, played a meaningful role in their compensation. Similarly, about one in eight reported that quality-of-care measures (13 percent) and one in seven cited patient satisfaction surveys (15 percent) as moderately or very important to their pay.
The relative weight of compensation factors varied by specialty, income level, and practice setting. Productivity and practice financial performance tended to gain importance as physician income increased, while quality measures, patient satisfaction, and profiling diminished in importance at higher earnings levels. Among specialties, compensation for surgical specialists was more closely linked to productivity and less tied to the other measures relative to general internists. Physicians in group or staff-model HMOs were more likely to cite patient satisfaction surveys, quality measures, and practice profiling as important incentives compared with those in solo or two-physician practices.
Forty-three percent of physicians described themselves as somewhat satisfied with their overall medical career, and 39 percent reported being very satisfied. Satisfaction levels varied by physician and practice characteristics. Physicians in practice for more than 20 years exhibited more polarized responses: they were more likely to be either very satisfied or very dissatisfied relative to newer doctors. Pediatricians and both medical and surgical specialists reported greater career satisfaction than general internists, family practitioners, psychiatrists, and obstetrician/gynecologists. Physicians with higher incomes also expressed greater career satisfaction.
On average, close to half of physicians' practice revenue came from public sources, approximately 31 percent from Medicare and 17 percent from Medicaid. Practices' willingness to accept new patients differed substantially by insurance type. Almost nine in ten physicians (87 percent) reported that their practices accepted all or most new privately insured patients, and nearly three-quarters (74 percent) accepted all or most new Medicare patients. However, only slightly more than half (53 percent) accepted all or most new Medicaid patients, while 28 percent accepted no new Medicaid patients at all.
Several patterns emerged in the relationship between physician and practice characteristics and new patient acceptance. Black physicians were more likely to accept new Medicaid patients. Medical and surgical specialists, whose services tend to be episodic rather than ongoing, were more willing to accept new patients regardless of insurance type. Psychiatrists were notably less likely to accept new patients across all insurance categories. Group practices, particularly those with six or more physicians, and institutional practices were more likely to accept new patients across the board. Physicians in rural areas also reported greater willingness to accept new patients.
Despite anecdotal reports that many physicians had abandoned insurance networks, the vast majority of physicians (87.6 percent) held managed care contracts in 2008. Nearly 70 percent reported having five or more managed care contracts. Physicians without managed care contracts tended to have practiced for more than 20 years, worked fewer than 40 hours per week, lacked board certification, practiced in solo or two-physician settings, resided in the western United States, and operated in noncompetitive environments.
Options for physicians who chose not to contract with managed care plans ranged from seeing only patients covered by insurance products without provider networks (such as fee-for-service Medicare) to establishing cash-only practices to serving managed care patients as out-of-network providers who could balance bill patients for charges beyond insurer allowances. Among specialties, psychiatrists were far less likely to hold managed care contracts, with about one-third having none, likely reflecting both low reimbursement rates and the burden of utilization management from behavioral health companies. Pediatricians, in contrast, were more likely to participate in managed care, probably because Medicaid managed care plans represent a major source of coverage for children.
In 2008, slightly fewer than six in ten U.S. physicians (59.1 percent) reported providing charity care, defined as free or reduced-cost care for patients in financial need. On average, physicians who delivered charity care spent 9.5 hours on it in the month preceding the survey, amounting to slightly more than 4 percent of their time in all medically related activities.
Provision of charity care varied considerably across physician and practice characteristics. Surgical specialists (73.5 percent) were the most likely to provide charity care, probably because many are required to be on call at hospitals where they encounter uninsured patients needing emergency services. Pediatricians (45.6 percent) were the least likely, perhaps reflecting the broader availability of public coverage for children. Levels of charity care were highest among physicians in solo or two-physician practices (71.5 percent). Physicians in larger groups, HMOs, and institutional practices, where charity care policies are typically set at the organizational level, were substantially less likely to provide charity care individually.
Physicians at the highest income levels reported the greatest provision of charity care: 67 percent of those earning more than $250,000 provided such care in 2008, compared with 54.7 percent of those earning less than $150,000. Physicians who had been in practice for more than 10 years were also more likely to report providing charity care than those with fewer years of experience.
For more research on healthcare costs, physician workforce data, and practice patterns, visit the section.
This snapshot of U.S. physicians was originally published by the Center for Studying Health System Change (HSC) in September 2009. Physician workforce trends continue to shape healthcare costs and access. For current resources: