Losing Ground: Physician Income, 1995-2003

Tracking Report No. 15

June 2006

Ha T. Tu, Paul B. Ginsburg

From 1995 to 2003, the average net income physicians earned from practicing medicine fell roughly 7 percent in inflation-adjusted terms, based on a nationally representative study conducted by the Center for Studying Health System Change (HSC). This erosion of physicians' real earnings contrasts sharply with income patterns among other professional workers, who experienced approximately a 7 percent gain after accounting for inflation over the same timeframe. Among physician categories, those in primary care were hit hardest, seeing a 10.2 percent drop in real income between 1995 and 2003, while surgeons experienced an 8.2 percent inflation-adjusted decline. Medical specialists, however, saw their real income remain essentially stable. Physicians indicated they worked marginally fewer hours overall yet devoted more time to direct patient care. Stagnant or falling reimbursement rates from both government and private insurers appear to be a central driver behind shrinking real incomes for physicians. The sustained downward trajectory in real earnings since the mid-1990s is likely a significant reason behind physicians' growing reluctance to engage in pro bono activities, including charity care and volunteer service on hospital committees.

Decline in Physicians' Real Income Continues

Between 1995 and 2003, the average net income that physicians derived from medical practice dropped approximately 7 percent after adjusting for inflation, according to HSC's nationally representative 2004-05 Community Tracking Study Physician Survey (see Figure 1 and Data Source). Primary care doctors and surgeons struggled the most to keep up with the rising cost of living, while medical specialists maintained their footing better than any other group.

When adjusted for inflation, the incomes of medical specialists held essentially steady from the mid-1990s onward. By contrast, primary care physicians -- already the lowest-paid category among all doctors -- experienced significant erosion of their purchasing power, losing 10.2 percent in real income since the mid-1990s. Surgical specialists also faced a notable reduction of over 8 percent in inflation-adjusted earnings between 1995 and 2003.

The negative trajectory for physician real income is a stark departure from the trends observed among workers in professional, specialty, and technical occupations.(1) From 1995 to 2003, wages and salaries for these workers rose roughly 7 percent in inflation-adjusted terms. The gap in income trends between physicians and other professional workers, which was particularly pronounced during the 1995-1999 period, narrowed somewhat from 1999 to 2003, yet physicians continued to trail their professional counterparts (see Table 1).

Notwithstanding the downward trend in real earnings, medicine as a whole continues to rank among the highest-compensated professions in the United States: At least half of all patient care physicians earned in excess of $170,000 in 2003, and the average physician net income stood at roughly $203,000. Furthermore, even though surgical specialists have lost ground to inflation since the mid-1990s, they remain the top earners among all physicians: Their average income of $272,000 in 2003 exceeded that of medical specialists by 29 percent and surpassed primary care physician incomes by 86 percent.

Table 1: Physicians' Net Income from Practice of Medicine, 1995, 1999 and 2003, and Percent Change, 1995-2003

All Patient Care Physicians: Average Reported Net Income -- 1995: $180,930; 1999: $186,768; 2003: $202,982. Average Net Income, Inflation Adjusted (1995 Dollars) -- 1995: $180,930; 1999: $170,850; 2003: $168,122. Percent Change in Inflation-Adjusted Income -- 1995-1999: -5.6*; 1999-2003: -1.6; 1995-2003: -7.1*.

Primary Care Physicians: Average Reported Net Income -- 1995: $135,036; 1999: $138,018; 2003: $146,405. Average Net Income, Inflation Adjusted (1995 Dollars) -- 1995: $135,036; 1999: $126,255; 2003: $121,262. Percent Change in Inflation-Adjusted Income -- 1995-1999: -6.5*; 1999-2003: -4.0*; 1995-2003: -10.2*.

Specialists: Average Reported Net Income -- 1995: $210,225; 1999: $218,819; 2003: $235,820. Average Net Income, Inflation Adjusted (1995 Dollars) -- 1995: $210,225; 1999: $200,169; 2003: $195,320. Percent Change in Inflation-Adjusted Income -- 1995-1999: -4.8*; 1999-2003: -2.4; 1995-2003: -7.1*.

Medical Specialists: Average Reported Net Income -- 1995: $178,840; 1999: $193,161; 2003: $211,299. Average Net Income, Inflation Adjusted (1995 Dollars) -- 1995: $178,840; 1999: $176,698; 2003: $175,011. Percent Change in Inflation-Adjusted Income -- 1995-1999: -1.2; 1999-2003: -1.0; 1995-2003: -2.1.

Surgical Specialists: Average Reported Net Income -- 1995: $245,162; 1999: $255,011; 2003: $271,652. Average Net Income, Inflation Adjusted (1995 Dollars) -- 1995: $245,162; 1999: $233,276; 2003: $224,998. Percent Change in Inflation-Adjusted Income -- 1995-1999: -4.9; 1999-2003: -3.6; 1995-2003: -8.2*.

Private Sector Professional, Technical, Specialty Occupations: Average Reported Net Income -- 1995: N/A; 1999: N/A; 2003: N/A. Average Net Income, Inflation Adjusted (1995 Dollars) -- 1995: N/A; 1999: N/A; 2003: N/A. Percent Change in Inflation-Adjusted Income -- 1995-1999: 4.3; 1999-2003: 2.5; 1995-2003: 6.9.

Notes: The Bureau of Labor Statistics (BLS) Employment Cost Index of wages and salaries for private sector "professional, technical and specialty" workers was used to calculate estimates for these workers. Significance tests are not available for these estimates. All inflation-adjusted estimates were calculated using the BLS online inflation calculator. The composition of the physician population changed between 1995 and 2003 -- a fact that makes some estimates of percentage changes in real income appear inconsistent (for example, estimates of income changes for all patient care physicians not falling between estimates for primary care physicians and specialists). These data patterns occur because the proportion of medical specialists steadily increased from 1995 to 2003 (32% to 38%) while the proportions of primary care physicians and surgical specialists both declined by about 3 percentage points.

* Rate of change is statistically significant at p <.05.

Source: Community Tracking Study Physician Survey

Physicians Work Fewer Hours but Spend More Time on Patient Care

Although physician earnings fell in real terms from 1995 to 2003, the average weekly hours physicians devoted to all medically related activities also edged downward during this span (see Table 2). Medically related activities encompass time spent on administrative duties, professional activities, and direct patient care, but exclude time on call when physicians are not actively treating patients. Across all patient care physicians, the weekly hours spent on medically related activities decreased from 55.5 in 1995 to 53.2 in 2003. This trend held for both primary care physicians and medical specialists; only surgical specialists -- who on average logged the most hours -- did not experience a statistically significant decline in average hours worked.

Even as physicians' total work hours declined, the time they spent in direct patient care actually increased (see Supplemental Table 1). Patient care activities include face-to-face encounters with patients, patient record keeping and office work, travel time associated with seeing patients, and communication with other physicians, hospitals, pharmacies, and others on a patient's behalf. The gains were generally moderate in scale, with the notable exception of surgical specialists, whose average weekly patient care hours rose considerably -- from 47.2 hours in 1995 to 50.1 hours in 2003. Because patient care hours grew while total medically related work hours fell, physicians now devote a significantly larger share of their work time to caring for patients than they did in the mid-1990s -- 86 percent compared with 81 percent (see Supplementary Table 2).

Given that time spent on patient care rose between 1995 and 2003, it is unlikely that the overall reduction in work hours played a major role in falling physician incomes, since revenue is more directly tied to patient care activities. Hours devoted to direct patient care increased for all physician categories between 1995 and 1999, then stabilized or dipped slightly thereafter. This pattern likely reflects the fact that the volume of office visits grew at a healthy pace during the late 1990s, but that growth rate decelerated starting in 2001.(2)

The overall reduction in work hours and in time spent on activities other than direct patient care may partly reflect shifts in physician practice settings over time. Physicians have been transitioning from solo and small group practices into large group practices and institutional environments, such as hospitals (see Supplementary Table 3). As physicians join larger practices that can invest more heavily in administrative staff and information technology to manage billing and other administrative functions, less physician time is likely needed for these tasks. Another possible contributor to reduced non-patient-care work hours is a reported retreat from unpaid activities by physicians, particularly in hospital settings.(3) Willingness to serve on hospital committees, for instance, appears to have declined as income pressures have intensified. Volunteerism may also have decreased for activities in the wider medical community, such as serving on specialty society committees.

Table 2: Average Hours per Week Spent on Medically Related Activities, 1995-2003

All Patient Care Physicians: Average Hours per Week -- 1995: 55.5; 1999: 54.4; 2003: 53.2. Percent Change in Hours -- 1995-1999: -1.9*; 1999-2003: -2.2*; 1995-2003: -4.1*.

Primary Care Physicians: Average Hours per Week -- 1995: 53.8; 1999: 52.7; 2003: 51.4. Percent Change in Hours -- 1995-1999: -2.0*; 1999-2003: -2.5*; 1995-2003: -4.5*.

Specialists: Average Hours per Week -- 1995: 56.6; 1999: 55.6; 2003: 54.3. Percent Change in Hours -- 1995-1999: -1.8*; 1999-2003: -2.3*; 1995-2003: -4.0*.

Medical Specialists: Average Hours per Week -- 1995: 54.6; 1999: 54.0; 2003: 51.5. Percent Change in Hours -- 1995-1999: -1.0; 1999-2003: -4.8*; 1995-2003: -5.8*.

Surgical Specialists: Average Hours per Week -- 1995: 58.8; 1999: 57.7; 2003: 58.5. Percent Change in Hours -- 1995-1999: -1.8*; 1999-2003: 1.3; 1995-2003: -0.6.

* Rate of change is statistically significant at p <.05.

Source: Community Tracking Study Physician Survey

Fees Fail to Keep Pace with Inflation

Stagnant or declining reimbursement rates from both government programs and private insurers appear to be a primary driver behind the erosion or stagnation of real incomes for physicians. Medicare payment rate increases for physician services totaled 13 percent from 1995 to 2003,(4) falling well behind inflation, which amounted to 21 percent over that eight-year span.

While Medicare fees have eroded in real terms since the mid-1990s, the trajectory for private insurer reimbursements to physicians has been even more unfavorable: In 1995, commercial fees averaged 1.43 times Medicare fees; by 2003, that ratio had dropped to 1.23.(5) Medicaid fees have consistently been well below Medicare levels, so even though Medicaid payment rates increased relative to Medicare and outpaced inflation from 1998 to 2003, higher Medicaid reimbursements alone would not have been sufficient to generate meaningful income gains for most physicians.(6) One probable exception would be primary care physicians with sizable Medicaid patient panels, particularly those practicing in states -- such as New York and South Carolina -- that began with low Medicaid fee levels and raised them most aggressively.

Volume of Physician Services Increases

Although physician payment rates have been constrained, the volume of physician services expanded significantly, driven primarily by growth in the number of tests and procedures. Among Medicare beneficiaries, minor procedures grew 6 percent per year on average between 1999 and 2003.(7) This rate of growth outstripped that of office visits (4% per year) and major procedures (3% per year). The robust growth in tests and procedures helps explain in part why medical specialists have seen their incomes keep pace better than primary care physicians, who depend more heavily on cognitive services -- such as evaluation and management of patients -- to generate revenue. Among medical specialists, those with procedure-driven practices, such as gastroenterologists or cardiologists, are better positioned to invest in specialty facilities and generate additional income than, for example, a psychiatrist, who like a primary care physician relies on cognitive-based services.(8)

Although Medicare's physician fee schedule is designed to reflect relative physician work and practice expense and is periodically updated, the failure to identify many services for which relative values should decrease due to rising productivity has led to overpayments for certain services, particularly procedures associated with rapidly advancing technology.(9) A widening gap between costs and fees is most likely to emerge for services where technological improvements make physicians, staff, and equipment more productive over time, allowing them to perform more procedures per day and operate at higher capacity. The net effect is that the relative compensation for evaluation and management services and other cognitive-based services -- where major productivity gains are unlikely -- is declining sharply over time. Because many commercial insurers incorporate Medicare's relative value scale into their own payment schedules, the adverse impact on specialties like primary care, for which office visits represent a key revenue source, extends well beyond Medicare patients.

Policy Implications

The sustained downward trend in real incomes since the mid-1990s is likely a key factor behind the declining willingness of physicians to engage in pro bono work, whether that means providing charity care to patients who cannot afford to pay(10) or volunteering for other medically related activities, such as serving on hospital committees.

Downward pressure on earnings is also likely connected to the migration of physicians away from primary care -- already lower-paying and experiencing steeper income declines -- toward certain medical specialties that offer higher compensation and have better kept pace with inflation. The makeup of the physician population shifted between 1995 and 2003, with the share of medical specialists rising steadily from 32 percent to 38 percent, while the proportions of primary care physicians and surgical specialists each fell by roughly 3 percentage points. In deciding which area of medicine to pursue, many physicians today already demonstrate preferences for medical specialties that afford greater control over work hours. When these preferences are reinforced by the widening income gap between these specialties and primary care, the likely outcome is an imbalance in the physician workforce and potentially a future shortage of primary care physicians and other specialties that deliver primarily cognitive services.

For policy makers aiming to realign the price signals directed at physicians to ensure the nation's medical needs are adequately met, the primary policy lever remains Medicare and Medicaid payment rates. For the foreseeable future, the congressionally enacted Medicare physician payment formula will generate significant annual reductions in payment rates, unless Congress steps in to reverse the cuts, as it has done every year since 2003. Few believe that Congress should directly determine physician incomes, and the debate over Medicare physician payments is largely framed in terms of beneficiary access to physician services(11) and the long-term fiscal consequences of Congress continuing to override Medicare physician payment reductions.

For certain physicians, particularly some medical specialists, declining fees do not necessarily translate to declining incomes, because their productivity for procedures is growing rapidly alongside the volume of procedures. But for primary care physicians and other specialists who depend heavily on office visits rather than procedures, falling fees are likely to translate directly into falling incomes.

This underscores the need for more effective updating of Medicare relative values to better reflect the current relative costs of different services, and policy makers have already begun taking steps in that direction. Recognizing the pressures confronting physicians in cognitive specialties, the Relative Value Scale Update Committee, or RUC, recently voted to increase the value of cognitive services relative to that of procedures. Additionally, the Medicare Payment Advisory Commission in March 2006 recommended that the U.S. Department of Health and Human Services create an expert panel -- including members with expertise in health economics and clinical practice -- to help identify overvalued services and review RUC recommendations.

Because many commercial insurers follow Medicare's relative value system, timely and accurate relative value updates would trigger a broader reallocation of physician resources that would yield benefits extending well beyond Medicare patients.

Notes

1. Changes in wage/salary income for private sector "professional, specialty and technical" workers are based on the Bureau of Labor Statistics (BLS) Employment Cost Index. Data were adjusted for inflation using the BLS online inflation calculator.

2. National Center for Health Statistics, National Ambulatory Medical Care Survey, Summaries from 1996, 2001 and 2003. Advance Data from Vital and Health Statistics, Nos. 295, 337, 365, Hyattsville, Md.

3. During HSC's 2005 site visits to 12 nationally representative communities, hospital administrators reported decreased physician willingness to serve on hospital committees. They also reported decreased physician willingness to provide emergency on-call coverage, but this is not a factor in trends on hours worked, since the questionnaire specifically directs physicians not to include time spent on call when not treating patients.

4. Medicare Payment Advisory Commission, A Data Book: Healthcare Spending and the Medicare Program (June 2005).

5. Medicare Payment Advisory Commission, Report to the Congress: Medicare Payment Policy (March 2005).

6. Zuckerman, Steve, et al., "Changes in Medicaid Physician Fees 1998-2003: Implications for Physician Participation," Health Affairs, Web exclusive (June 23, 2004).

7. Medicare Payment Advisory Commission, A Data Book: Healthcare Spending and the Medicare Program (June 2005).

8. Pham, Hoangmai, H., et al., "Financial Pressures Spur Physician Entrepreneurialism," Health Affairs, Vol. 23, No. 2, (March/April 2004).

9. Ginsburg, Paul B. and Joy M. Grossman, "When The Price Isn't Right: How Inadvertent Payment Incentives Drive Medical Care," Health Affairs, Web exclusive (August 9, 2005).

10. Cunningham, Peter J. and Jessica H. May, A Growing Hole in the Safety Net: Physician Charity Care Declines Again, Tracking Report No. 13, Center for Studying Health System Change, Washington, D.C. (March 2006).

11. Cunningham, Peter, Andrea Staiti and Paul B. Ginsburg, Physician Acceptance of New Medicare Patients Stabilizes in 2004-05, Tracking Report No. 12, Center for Studying Health System Change, Washington, D.C. (January 2006).

Data Source

This Tracking Report presents findings primarily from the HSC Community Tracking Study Physician Survey, a nationally representative telephone survey of physicians involved in direct patient care in the continental United States conducted in 1996-97, 1998-99 (results not shown for ease of presentation), 2000-01, and 2004-05. The sample of physicians was drawn from the American Medical Association and the American Osteopathic Association master files and included active, nonfederal, office- and hospital-based physicians who spent at least 20 hours a week in direct patient care. Residents and fellows were excluded. The 1996-97, 1998-99, and 2000-01 surveys each contain information on about 12,000 physicians, while the 2004-05 survey includes responses from more than 6,600 physicians. The response rates ranged from 52 percent to 65 percent.

Physicians were asked to report their incomes for the last full year prior to the initial fielding of each survey (e.g., respondents to the 1996-97 survey were asked to report their 1995 incomes). Other measures used in this report use a different, more recent time frame (e.g., physicians are asked to report the number of hours they worked in the week prior to their survey participation). For ease of presentation, all measures are reported in this tracking report as being for the years 1995, 1999 and 2003. More detailed information on survey content and methodology can be found at www.hschange.org.

Supplementary Tables

Supplementary Table 1: Average Hours per Week Spent on Direct Patient Care, 1995-2003

Supplementary Table 2: Hours Spent on Direct Patient Care as a Percentage of Medically Related Activity Hours, 1995-2003

Supplementary Table 3: Distribution of Patient Care Physicians by Practice Type, 1995-2003

TRACKING REPORTS are published by the Center for Studying Health System Change.

600 Maryland Avenue, SW, Suite 550, Washington, DC 20024-2512. Tel: (202) 484-5261. Fax: (202) 484-9258. www.hschange.org

President: Paul B. Ginsburg; Vice President: Jon Gabel

Sources and Further Reading

Much of the reimbursement data discussed in this report stems from Medicare's payment methodology for physician services. The CMS Medicare Physician Fee Schedule provides the official rate-setting framework that determines how physicians are paid under Medicare, including the relative value units and conversion factors referenced throughout this tracking report.

The inflation adjustments central to this analysis rely on compensation benchmarks maintained by the federal government. The Bureau of Labor Statistics Employment Cost Index tracks quarterly changes in labor costs across industries and occupational groups, and served as the basis for comparing physician income trends against those of other professional, specialty, and technical workers from 1995 to 2003.

Physician practice patterns, including hours worked and practice settings, have been tracked extensively by organized medicine. The American Medical Association's Physician Practice Benchmarks offer survey-based data on practice characteristics, ownership structures, and work hours that complement the Community Tracking Study findings on how physicians have shifted from solo and small-group settings into larger organizational arrangements.

Several of the original studies cited in this report were published in peer-reviewed health policy journals. Health Affairs published key research on Medicaid physician fee changes, physician entrepreneurialism driven by financial pressures, and the distortions created by Medicare's relative value system -- all of which directly informed the conclusions in this tracking report about why primary care incomes fell faster than specialist incomes.

The policy recommendations in this report draw heavily on advisory work conducted for Congress. The Medicare Payment Advisory Commission (MedPAC) publishes annual reports to Congress on Medicare payment policy, including analyses of physician fee updates, service volume growth, and the spending implications of overriding scheduled payment reductions -- the same issues at the heart of the income trends documented here.