Medicaid Patients Increasingly Concentrated Among Physicians

Tracking Report No. 16

August 2006

Peter J. Cunningham, Jessica H. May

Even though Medicaid reimbursement rates and program enrollment have grown, the share of U.S. doctors willing to treat Medicaid beneficiaries has edged downward over the past ten years, according to a nationwide analysis by the Center for Studying Health System Change (HSC). In 2004-05, 14.6 percent of physicians indicated they received zero revenue from Medicaid, up from 12.9 percent in 1996-97. Modest increases also emerged in the proportion of doctors declining new Medicaid patients. Perhaps more notably, treatment of Medicaid enrollees is becoming progressively concentrated among a smaller subset of physicians who tend to work in large group practices, hospitals, academic medical centers, and community health centers. Comparatively low reimbursement levels and substantial administrative expenses are likely driving decreased Medicaid involvement among doctors in solo and small group practice settings.

Physician Medicaid Participation Down Slightly

Medicaid reimbursement rates, which fall well below what physicians receive from Medicare or private insurers, have long discouraged doctor participation in the program. Roughly one in five physicians (21 percent) reported turning away all new Medicaid patients in 2004-05 -- a rate six times greater than for Medicare beneficiaries and five times greater than for those with private coverage, based on HSC's nationally representative Community Tracking Study Physician Survey (see Data Source and Table 1). Furthermore, only about half of doctors indicated they were taking all new Medicaid patients in 2004-05, in contrast to over 70 percent who were open to all new Medicare and privately insured patients. Research has demonstrated that limited physician participation in Medicaid adversely affects enrollees' ability to obtain medical care. [1]

Between 1998 and 2003, Medicaid reimbursement rates rose modestly relative to Medicare rates, even as some states cut or froze payment levels in the early 2000s amid severe fiscal constraints and escalating Medicaid expenditures. [2] Concurrently, Medicaid enrollment expanded significantly during this span -- growing 8 percent overall between 2000 and 2003 -- likely driven by broadened eligibility criteria, steep annual increases in private health insurance premiums, and sluggish economic conditions following the 2001 recession that cost many workers their jobs and employer-provided coverage. [3]

Notwithstanding these developments, physician engagement with Medicaid declined marginally over the past decade. From 1996-97 to 2000-01, the share of doctors earning no revenue from Medicaid rose from 12.9 percent to 14.6 percent, while the percentage refusing all new Medicaid patients climbed from 19.4 percent to 20.9 percent. Between 2000-01 and 2004-05, there was no statistically significant change in the share of physicians deriving any Medicaid revenue or in the proportion with practices closed to new Medicaid enrollees.

Table 1: Physician Participation by Insurance Type

Medicaid

No Medicaid Revenue: 1996-97: 12.9% | 2000-01: 14.6%# | 2004-05: 14.6%#

Accepting No New Patients: 1996-97: 19.4 | 2000-01: 20.9# | 2004-05: 21.0

Accepting All New Patients: 1996-97: 51.1 | 2000-01: 51.9 | 2004-05: 52.1

Privately Insured

Accepting No New Patients: 1996-97: 3.6 | 2000-01: 4.9# | 2004-05: 4.3

Accepting All New Patients: 1996-97: 70.8 | 2000-01: 68.2# | 2004-05: 71.8*

Medicare

Accepting No New Patients: 1996-97: 3.1 | 2000-01: 3.8# | 2004-05: 3.4

Accepting All New Patients: 1996-97: 74.6 | 2000-01: 71.1# | 2004-05: 72.9

# Change from 1996-97 is statistically significant at p <.05. * Change from 2000-01 is statistically significant at p <.05.

Source: Community Tracking Study Physician Survey

Medicaid Patients More Concentrated

For the majority of physicians who treat Medicaid beneficiaries, Medicaid constitutes a relatively minor portion of their practice. Among all doctors providing any care to Medicaid patients, approximately 60 percent derive less than 20 percent of their total practice revenue from the program (findings not shown). Yet these physicians account for only about 28 percent of physician care delivered to Medicaid patients in dollar terms (see Table 2). By contrast, roughly one-quarter of physicians earn 30 percent or more of their practice income from Medicaid, and these doctors are responsible for more than half of all physician care rendered to Medicaid enrollees.

Additionally, treatment of Medicaid patients is becoming more heavily concentrated among the minority of doctors who deliver a substantial volume of Medicaid care. The proportion of total Medicaid physician revenue attributed to doctors earning 30 percent or more of their practice income from Medicaid grew from 43.1 percent in 1996-97 to 51 percent in 2004-05. Meanwhile, the share of Medicaid physician revenue accounted for by doctors with less than 20 percent of practice revenue from Medicaid fell from about 38 percent to 28.4 percent.

At least part of this shift can be attributed to the growing reluctance of physicians with lower Medicaid involvement to accept new Medicaid patients. For doctors deriving between 1 percent and 9 percent of practice revenue from Medicaid, the proportion declining new Medicaid patients rose from 20.7 percent in 1996-97 to 27.1 percent in 2004-05 (see Table 3). In contrast, fewer than 3 percent of physicians who earn 30 percent or more of their revenue from Medicaid were turning away new patients, and this figure has remained stable over the past decade.

Table 2: Distribution of Medicaid Physician Practice Revenue

Percent of Revenue from Medicaid

0-9%: 1996-97: 10.6% | 2000-01: 9.0% | 2004-05: 7.8%

10-19%: 1996-97: 27.2 | 2000-01: 24.3 | 2004-05: 20.6

20-29%: 1996-97: 19.1 | 2000-01: 20.7 | 2004-05: 20.6

30% or Higher: 1996-97: 43.1 | 2000-01: 46.1 | 2004-05: 51.0

Note: Physicians who derived no revenue from Medicaid are excluded.

Source: Community Tracking Study Physician Survey

Table 3: Physicians Accepting No New Medicaid Patients

All Physicians: 1996-97: 19.4% | 2000-01: 20.9%# | 2004-05: 21.0%

Percent of Revenue from Medicaid

1-9%: 1996-97: 20.7 | 2000-01: 23.9# | 2004-05: 27.1#

10-19%: 1996-97: 5.9 | 2000-01: 7.2 | 2004-05: 7.9#

20-29%: 1996-97: 4.4 | 2000-01: 3.4 | 2004-05: 2.6#

30% or Higher: 1996-97: 2.7 | 2000-01: 3.2 | 2004-05: 2.8

# Change from 1996-97 is statistically significant at p <.05.

Source: Community Tracking Study Physician Survey

Physician Practice Characteristics

The growing concentration of Medicaid patient care is also marked by a migration away from small, office-based practices toward larger group settings and institution-based environments, including hospitals, academic medical centers, and community health centers. A substantially higher percentage of physicians in solo or small group practices earn little or no Medicaid revenue compared with those in institutional settings and other practice configurations, while more than half of institutional providers generate 20 percent or more of their practice income from Medicaid (see Supplementary Table 1). Nonetheless, solo and small group practice physicians still deliver a significant amount of care to Medicaid patients, representing over 40 percent of all physician Medicaid revenue in 2004-05 -- compared with 30.5 percent for institutional providers.

However, Medicaid patient care has increasingly shifted away from smaller practices over the past ten years. The share of total Medicaid physician revenue attributable to solo and small group practices fell from 52.4 percent in 1996-97 to 41.7 percent in 2004-05 (findings not shown). Conversely, medium and large group practices, along with institutional providers, captured a greater share of Medicaid revenue during the same period.

Part of this transition reflects physicians' broader movement out of smaller practices and into larger groups and other settings, as documented in earlier HSC research. [4] However, physicians who remain in solo practice or smaller groups are also increasingly refusing to accept new Medicaid patients. For instance, 35.3 percent of doctors in solo and two-physician practices were not accepting new Medicaid patients in 2004-05, up from 29 percent in 1996-97 (see Table 4). By contrast, fewer physicians in larger group practices and institutional environments were turning away new Medicaid patients, and this figure actually decreased slightly among institutional providers.

Table 4: Physicians Accepting No New Medicaid Patients, by Practice Type, Specialty and Location

All Physicians: 1996-97: 19.4% | 2000-01: 20.9%# | 2004-05: 21.0%

Practice Type

Solo/2 Physicians: 1996-97: 29.0 | 2000-01: 35.5# | 2004-05: 35.3#

Small Group: 1996-97: 16.2 | 2000-01: 18.0 | 2004-05: 24.0#*

Medium Group: 1996-97: 10.0 | 2000-01: 13.3 | 2004-05: 12.0

Large Group: 1996-97: 15.0 | 2000-01: 15.6 | 2004-05: 13.3

Group/Staff HMO: 1996-97: 15.1 | 2000-01: 21.7# | 2004-05: 13.5*

Institutional Provider (1): 1996-97: 8.3 | 2000-01: 9.7 | 2004-05: 6.6#*

Other: 1996-97: 19.0 | 2000-01: 17.1 | 2004-05: 18.9

Specialty

General Internal Medicine: 1996-97: 27.0 | 2000-01: 27.0 | 2004-05: 30.5

Family Practice: 1996-97: 25.5 | 2000-01: 28.4# | 2004-05: 27.3

Pediatrics: 1996-97: 15.8 | 2000-01: 14.9 | 2004-05: 15.0

Medical Specialists: 1996-97: 18.9 | 2000-01: 17.0 | 2004-05: 18.0

Surgical Specialists: 1996-97: 13.9 | 2000-01: 19.5# | 2004-05: 18.8#

Location

Large Metropolitan Statistical Area (MSA) (100,000+): 1996-97: 21.3 | 2000-01: 23.1# | 2004-05: 23.6#

Small MSA: 1996-97: 13.7 | 2000-01: 15.5 | 2004-05: 13.3

Non MSA: 1996-97: 9.5 | 2000-01: 9.9 | 2004-05: 10.6

# Change from 1996-97 is statistically significant at p <.05. * Change from 2000-01 is statistically significant at p <.05.

(1) Includes hospitals, academic medical centers, community health centers and other free clinics.

Source: Community Tracking Study Physician Survey

Physician Specialty

Willingness to accept new Medicaid patients differs across the major physician specialty categories. General internists and family practitioners are the most likely to report having closed their practices to new Medicaid enrollees, whereas pediatricians and specialists are the least likely to have done so. Because children are far more likely than adults to be enrolled in Medicaid and the State Children's Health Insurance Program (SCHIP), pediatricians have less flexibility to decline Medicaid patients. Similarly, many specialists carry on-call duties at hospital emergency departments, which reduces their ability to select which patients they treat in those settings.

From 1996-97 through 2004-05, the only statistically significant shift in Medicaid acceptance rates by specialty occurred among surgical specialists. In 2004-05, 18.8 percent of surgical specialists indicated they were refusing all new Medicaid patients, compared with 13.9 percent in 1996-97. This change may be linked to the broader trend of physicians -- particularly surgeons -- no longer practicing exclusively in hospital environments. [5] As more doctors reduce their hospital emergency department coverage and work in freestanding ambulatory surgery centers, surgical specialists may now find it easier to avoid treating Medicaid patients than they could in the past.

Physician Location

Doctors practicing in large metropolitan areas (populations exceeding 200,000) were less inclined to accept new Medicaid patients compared with their counterparts in smaller metropolitan areas and rural communities. Moreover, the share of physicians in large metropolitan areas refusing new Medicaid patients rose modestly, from 21.3 percent in 1996-97 to 23.6 percent in 2004-05. The far greater density of both residents and healthcare providers in major urban centers gives physicians there more latitude in choosing which patients to serve compared with rural doctors. Additionally, the perceived responsibility to accept Medicaid patients may carry more weight in rural areas, where there are fewer alternative physicians available to Medicaid enrollees, particularly for specialty services.

Low Pay, Administrative Hassles Deter Medicaid Participation

Comparatively low Medicaid reimbursement rates and heavy administrative demands are primary reasons physicians cite for refusing Medicaid patients. Among doctors who were not accepting new Medicaid patients in 2004-05, approximately five out of six (84%) identified insufficient reimbursement as a moderately or very important factor in their decision (see Supplementary Table 2). Billing requirements and paperwork were mentioned by 70 percent of physicians as contributing factors, while roughly two-thirds pointed to delayed reimbursement. A smaller proportion of doctors cited concerns about already having a full patient panel or the elevated clinical demands of Medicaid patients.

These concerns also likely account for why physicians in smaller practices are increasingly closing their doors to new Medicaid patients. The administrative burden associated with treating Medicaid beneficiaries may have intensified in recent years, as more states require prior authorization for prescription medications and other tests and procedures. [6] For doctors in solo or small group practices, these administrative expenses may be prohibitively high on a per-patient basis given the limited number of Medicaid patients they see. Indeed, physicians in solo or small group settings are considerably more likely to cite billing requirements and paperwork as reasons for turning away new Medicaid patients compared with doctors in larger group practices and institutional environments, where centralized billing operations and economies of scale can mitigate the administrative burden of treating Medicaid enrollees.

Policy Implications

Despite increases between 1998 and 2003, average Medicaid physician reimbursement in 2003 stood at just 69 percent of Medicare reimbursement levels and was even lower when compared with private insurance payment rates. [7] Given the paperwork and administrative burdens that physicians report, it is unsurprising that a growing number of doctors in small private practices are choosing to stop serving Medicaid patients. The consequence is that Medicaid patient care is becoming progressively concentrated among physicians working in larger groups, as well as those in hospitals, academic medical centers, and community health centers.

This trend of rising concentration will likely persist in the near term. Physicians face substantial financial pressures and declining real incomes as a result of stagnant reimbursement rates from both Medicare and private payers. [8] These pressures are leading some doctors to scale back the time they devote to volunteer work and other less profitable aspects of their practice, which may include caring for Medicaid enrollees.

Growing concentration is also likely to be accelerated by the expansion of Medicaid managed care enrollment and the emergence of Medicaid-only health plans. Managed care enrollment climbed from roughly 40 percent of Medicaid enrollees in 1996 to approximately 60 percent by 2004, and further growth is anticipated. [9] While Medicaid managed care plans previously included numerous commercial plans serving a mixture of Medicaid and privately insured individuals, most Medicaid managed care plans now serve Medicaid beneficiaries either predominantly or exclusively. [10] Physician networks that contract with these plans are likely to include practices that deliver a disproportionate share of care to Medicaid patients (e.g., clinics, hospital-based physicians), while excluding those that treat relatively few Medicaid patients (e.g., solo and small group practices).

Fundamental restructuring of the Medicaid program could effectively lower Medicaid physician payment rates and diminish physician participation in the program even further. The Deficit Reduction Act (DRA) of 2005 was projected to reduce federal Medicaid spending in part by raising enrollee cost sharing for premiums and health services. While copayments for services had been restricted to $3 or less under prior law, the DRA permitted states to charge certain Medicaid enrollees coinsurance of up to 20 percent for some services. [11] If beneficiaries are unable to pay -- as many expect given the low income levels of most Medicaid enrollees -- physicians will either have to accept reduced Medicaid payments or increase their administrative costs to collect from patients.

Since low reimbursement and high administrative expenses in Medicaid already represent serious concerns among physicians, some doctors are likely to respond to increased enrollee cost sharing by shutting their practices to Medicaid patients. Enrollees will continue to gravitate toward providers who depend on Medicaid revenue or who are committed by their organizational mission to serving Medicaid beneficiaries. Whether the growing concentration in itself harms enrollee access to medical care remains unclear, since many large Medicaid providers are situated in areas where enrollees tend to reside, such as inner cities and medically underserved communities. However, if these major Medicaid providers face mounting financial pressures and rising patient demand, quality of care and access to certain services could suffer.

Notes

1. Cunningham, Peter J., and Len M. Nichols, "The Effects of Medicaid Reimbursement on the Access to Care of Medicaid Enrollees: A Community Perspective," Medical Care Research and Review, Vol. 62, No. 6 (December 2005).

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

3. Holahan, John, and Arunabh Ghosh, "Understanding the Recent Growth in Medicaid Spending, 2000-03," Health Affairs, Web Exclusive (Jan. 26, 2005).

4. Tu, Ha T., and Paul B. Ginsburg, Tracking Report No. 15, Losing Ground: Physician Income: 1995-2003, Center for Studying Health System Change, Washington, D.C. (June 2006).

5. Berenson, Robert A., Paul B. Ginsburg and Jessica H. May, "Hospital-Physician Relations: Cooperation, Competition, or Separation?" (under review).

6. Crowley, Jeffrey S., et al., State Medicaid Outpatient Prescription Drug Policies: Findings From a National Survey, 2005 Update, Kaiser Commission on Medicaid and the Uninsured, Washington, D.C. (October 2005).

7. Zuckerman, et al., op. cit.

8. Tu and Ginsburg, op. cit.

9. Centers for Medicare and Medicaid Services, 2004 Medicaid Managed Care Enrollment Report.

10. Felt-Lisk, Sue, Rebecca Dodge and Megan McHugh, Trends in Health Plans Serving Medicaid, 2000 Update, Kaiser Commission on Medicaid and the Uninsured, Washington, D.C. (2001).

11. Kaiser Commission on Medicaid and the Uninsured, Deficit Reduction Act of 2005: Implications for Medicaid, Washington, D.C. (February 2006).

Data Source

This Tracking Report draws on findings from the HSC Community Tracking Study Physician Survey, a nationally representative telephone survey of physicians engaged in direct patient care across the continental United States. The survey was conducted in 1996-97, 1998-99 (results not shown for ease of presentation), 2000-01, and 2004-05. The physician sample was drawn from the American Medical Association and the American Osteopathic Association master files and included active, nonfederal, office- and hospital-based physicians who devoted at least 20 hours per week to direct patient care. Residents and fellows were excluded from the sample. The 1996-97, 1998-99, and 2000-01 surveys each captured data on approximately 12,000 physicians, while the 2004-05 survey collected responses from more than 6,600 physicians. Response rates ranged from 52 percent to 65 percent. More detailed information on survey methodology can be found at www.hschange.org.

Supplementary Tables

Supplementary Table 1: Provision of Medicaid by Practice Type, 2004-05

Supplementary Table 2: Moderately or Very Important Reasons for Not Accepting New Medicaid Patients, 2004-05

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

Medicaid.gov — Official CMS resource on Medicaid program enrollment, eligibility, benefits, and provider participation data by state.

KFF: Medicaid Physician Fee Levels — Analysis of Medicaid-to-Medicare fee ratios across states and the relationship between payment rates and physician willingness to accept Medicaid patients.

Health Affairs Journal — Peer-reviewed research on Medicaid access, physician workforce distribution, and the effects of reimbursement policy on provider participation.

AHRQ Medical Expenditure Panel Survey (MEPS) — National data on health care utilization and expenditures, including physician visit patterns by insurance type referenced in this study.

Bureau of Labor Statistics: Physicians and Surgeons — Federal employment and wage data for physicians across specialties, providing context for physician workforce trends.