County Hospitals and Regional Medical Care in Texas

An Analysis of Out-of-County Costs

Texas Medicine Logo

Journal Article -- June 2004

By Ruben Amarasingham, MD; Sue Pickens, MEd; Ron J. Anderson, MD

The current system of regional medical service delivery in Texas places large demands on the state's urban public hospitals. To assess the nature and scope of such demands, we examined financial data from five of the state's largest public hospital districts. During fiscal year 2002, these hospitals reported 103,381 encounters with out-of-county patients, resulting in $66 million in unreimbursed costs. Given the current economic outlook, Texas requires a more effective regional model that centralizes tertiary care, disperses primary and secondary care, and preserves key public health goods.


In Texas, responsibility for the health care of indigent persons falls to the counties. The 1985 Indigent Health Care and Treatment Act specified 3 ways this responsibility might be met: counties may establish a hospital district, support a public hospital, or create a county-based indigent health care program (CIHCP). Although effective in establishing minimum standards of care for indigent persons, the act does not acknowledge geographic, economic, and demographic differences that complicate the uniform delivery of health care across counties. In the absence of a more considered regional system, urban county hospitals have assumed substantial extrajurisdictional health care responsibilities.

Several factors can explain this phenomenon. First, many rural counties cannot support critical tertiary services such as trauma (level 1 or 2 capability), burn care, neonatal care, pediatric intensive care, comprehensive cancer services, and other complex medical or surgical interventions. 1 Patients in these counties, particularly those without health insurance, rely upon larger urban hospitals with these capabilities. Second, 138 of the 254 counties in Texas operate under the auspices of a CIHCP plan. In these systems, indigent patients must navigate a labyrinth of county clinics, sometimes limited by inadequate specialty and referral capabilities. 2,3 When hospital services are needed, the only recourse might be a local private hospital, whose charity might already be strained, or the closest public hospital. Third, counties without a public hospital may, and often do, use the strictest possible criteria to define indigent status: 21% of the federal poverty level (currently $163 per month for a single individual). 4 In 2002, 10.3 million Texas residents earned less than 250% of the federal poverty level; 39%, or 4 million, of these persons were uninsured. 5 Lacking insurance, and without county assistance, patients who earn more than 21% of the federal poverty level are forced to delay care until disease is extreme. At that point, many of these uninsured patients, including those from surrounding counties, will turn to an urban public hospital for help. For these reasons, the burden on metropolitan public hospitals is increasingly regional, and not provincial, in nature.

To understand the scope, cost, and consequences of regional health care provided by urban public facilities, we examined the aggregate financial data of the largest public hospital districts in Texas.

Study Hospitals, Data, and Methods

Five major Texas public hospital districts were selected for this analysis: Parkland Hospital in Dallas County (Dallas), R.E. Thomason Hospital in El Paso County (El Paso), Harris County Hospital District (Houston), University Hospital in Bexar County (San Antonio), and John Peter Smith Hospital in Tarrant County (Fort Worth). All three hospitals in the Harris County Hospital District -- Ben Taub General, LBJ General, and Quentin Mease Community -- were included in this analysis.

Each of these hospitals is governed by a hospital district mechanism in which a local property tax levy funds the bulk of its operations. In addition, each hospital is a critical medical training site, 3 being the primary teaching hospital for local medical schools. All of the hospitals offer a substantial breadth of primary, secondary, and tertiary services.

Historically, uncompensated care (the sum of bad debt and charity care) has been used as a measure of a hospital's commitment toward safety net services. In 2001, 354 general hospitals in Texas provided $5.2 billion dollars in uncompensated care.* Of this amount, the 5 study hospitals provided $1.2 billion, or 23%. By almost any previous definition, these hospitals are among the most critical safety net centers in Texas. A typical safety net hospital demonstrates a ratio of uncompensated care to operating expense in the 90th percentile or higher. 6 Ratios for all 5 study hospitals fall above the 95th percentile. In 2001, Zuckerman et al developed a method of calculating the market share of uncompensated care expense (UCE), which suggests that safety net hospitals should score 2 or higher. 7 On this measure, each of the study hospitals scored 4.5 or higher ( Table 1 [PDF]).

Each hospital returned a detailed financial survey outlining the cost of services provided to out-of-county patients during fiscal year 2002. The instrument was divided into four sections: inpatient (self-pay or charity); inpatient (third-party payer); outpatient (self-pay or charity); and outpatient (third-party payer). Third-party payers include Medicaid, Medicare, and commercial insurers. This category also includes various other payer systems. (For a full list, as well as a copy of the financial instrument used, contact the authors.)

We obtained the total number of patients, charges, costs, and unreimbursed costs for each of the four sections. Because we were concerned with specialty services that may or may not be offered in surrounding counties, we asked each hospital to exclude outpatient visits to primary care clinics. In this analysis, outpatient visits reflect encounters in specialty-based clinics, such as cardiology or hematology/oncology, or outpatient emergency room visits. Each hospital was able to make this separation without limitation. Data from each of the hospitals were then aggregated to produce a picture of the total costs of providing out-of-county care ( Table 2 [PDF]). The cost recovery rate is the percentage of costs recovered in each category.


The 5 hospitals treated nearly 14,000 out-of-county patients admitted during 2002 ( Table 2 ). This substantial inpatient volume is roughly equivalent to 4.2 times the full capacity of all 5 hospitals. As more than a quarter of these patients had no funding source (self-pay or charity), unreimbursed costs totaled $28.5 million. As expected, the cost recovery rate for these uninsured patients was an abysmal 8%. Importantly, the recovery for those with some third-party payer coverage, which includes commercial insurance, was only 82%. Historically, safety net hospitals have used surpluses from paying patients to cross-subsidize the losses incurred from uninsured patients. This process, known as "cost-shifting," is not possible here. Even in the individual case, no hospital had sufficient revenue from third-party payers to cover losses from self-pay or charity patients. Considering self-pay and charity patients and patients with third-party coverage, the 5 hospitals lost nearly $49 million in unreimbursed inpatient costs.

Specialty-based (and emergency room) outpatient care is unavailable in many rural counties because of geographic isolation and maldistribution of specialists. Among the 5 hospital systems, 89,492 out-of-county outpatient visits occurred, with an extremely low total cost recovery rate of 43% ( Table 2 ). Again, cost-shifting was not possible given the large losses experienced from paying patients as indicated by a 60% cost recovery rate.

The 5 study hospitals provided 103,381 clinic or inpatient encounters to persons who live outside of the hospitals' mandated jurisdictions. The cumulative cost of these admissions and clinic visits was nearly $174 million. Of this, $66 million was not reimbursed, producing a cost recovery rate of 62% ( Table 2 ). In no single category did cumulative cost reimbursements equal 100%. In context, the $66 million in unreimbursed costs represents $186 million in unreimbursed charges (bad debt and charity care charges). Considering that the total uncompensated care for all 5 hospitals equaled $1.2 billion dollars, out-of-county care represented nearly 16% of that amount. For these hospitals, out-of-county care clearly represented a substantial loss. Importantly, in the case of both inpatients and outpatients, revenue from paying patients (commercial, Medicaid, or Medicare) was not sufficient to cross-subsidize the enormous losses generated by uninsured patients.


Simply stated, regionalization aims to concentrate limited or expensive health care services locally within an area while dispersing primary and secondary care more broadly. Theoretically, regional health services might develop in one of two ways. In some cases, development is anticipatory; a regional plan develops from the considered logic of public health planning, coordination, and market considerations. Sophisticated referral patterns and provider relationships are created at micro- and macro-levels to support larger regional health goals. In other cases, regionalization is forced into existence: only those providers with the necessary economies of scale maintain a market presence. Smaller hospitals, perhaps in more rural territories, cannot and do not compete in the provision of such services. In Texas, market forces and population dynamics, rather than public health planning, is the determining calculus. Urban counties, the only counties with the willingness and capability to support a large infrastructure of services, inevitably become the focus of health care delivery for an entire region.

However, in the absence of a concrete regional plan governing health care delivery and reimbursement, costs associated with out-of-county admissions threaten the solvency of urban public hospitals for several reasons. First, at a time of diminishing support for public institutions at the federal, state, and local levels, millions of dollars in losses are simply unsustainable. Second, the increasing volume of uninsured patients (both in- and out-of-county) requires hospital districts to expand rapidly and replace facilities with major capital improvements. High levels of uncompensated care prevent this in two ways. First, such losses diminish the internal funds necessary for plant improvement. And, second, the financial vulnerabilities imposed by uncompensated care raise the cost of externally generated capital, preventing beleaguered hospitals from borrowing what they really need for capital replacement.8 As fiscal concerns grow and other sources of support decrease, safety net providers will be forced to cut out-of-county services, avoid capital improvements, or fail altogether.

Finally, rather than dispersing care as widely as possible, the present environment urbanizes primary, secondary, and tertiary care for low-income residents, thereby limiting resources, increasing wait and travel times, and congesting tertiary facilities with patients  requiring less sophisticated care.

Effective solutions must focus on three distinct but critical goals: preserving comprehensive tertiary services, increasing health insurance coverage, and improving health services for rural patients. Texas requires a multipronged strategy to successfully achieve such aims. How might such a plan be constructed?

First, regional plans must consider indigent health care from a larger vantage than the county unit. In Texas, the 11 public health regions may represent the best organizational structure. The regions are well established and already form the basis for the Texas Department of Health's approach to public health. Addressing indigent and tertiary services in a regional manner will eliminate the heavy dependence on the circumstances, and willingness, of a few counties.

We recommend that regional authorities or boards, working closely with local hospitals, design plans that meet the individual needs of a region. This strategy prevents a "one size fits all" scheme to health services. For example, a model in West Texas may emphasize development of a rural medical care infrastructure at the primary and secondary care levels. In Northeast Texas, emphasis may instead be on developing more efficient referral patterns to facilitate tertiary care. In the end, regional plans must identify and develop critical infrastructure needs for the region, enhance the capability of participating rural facilities, establish efficient referral and pre-hospital care (emergency medical services) patterns for the region, improve indigent health care reimbursement, develop strategies to disperse primary and secondary care in rural areas, and maintain timely access to centralized tertiary services in urban areas.

Furthermore, we suggest that each regional authority develop a model around 3 classes of participating hospitals with different levels of responsibility. Numeric criteria should be established by the state to identify these facilities on the basis of their historic commitments to tertiary and indigent services. This classification might be divided roughly into urban public hospitals providing significant tertiary services, nonprofit community hospitals with some tertiary services, and rural hospitals. Explicit funding mechanisms could then be developed to reimburse hospitals for primary, secondary, and tertiary indigent care.

Finally, to be effective, a regional plan must draw upon novel funding strategies at the state and local levels. Efforts can be made in two directions. First, equitable mechanisms to support care should be implemented across counties. Ad valorem taxes could be imposed on a region, rather than simply a county, to support regional missions. Statewide matches, similar to those currently in effect for counties, could be created to alleviate the burdens faced by urban hospitals for out-of-county indigent care. Funds established for a region could then be distributed to hospitals by the regional authority or a statewide department such as the Texas Department of Health. Second, every effort to decrease the level of uninsured should be pursued. Policymakers can move in new and creative directions such as developing rural health care cooperatives, employing waivers under the Health Insurance Flexibility and Accountability (HIFA) initiative to achieve "Medicaid buy-in," and increasing the floor of eligibility criteria for indigent care.


Texas needs an organized, flexible, and creatively funded model for regional medical services. The current alternative leans heavily on already overburdened urban public hospitals and does little to coordinate regional processes across counties. We support the development of a more effective system that decreases the rates of uninsured, improves primary and secondary medical services for rural patients, and preserves comprehensive tertiary services across Texas.  


The authors thank the 5 Texas hospitals that participated in this analysis. We would especially like to thank Billy Moore, PhD, director of market research, Baylor Healthcare; Jennifer Cutrer, PhD, director of governmental affairs, Parkland Hospital; Eduardo Sanchez, MD, MPH, Texas commissioner of health; Mark Teresi, former chief financial officer, Parkland Hospital; Jorie Klein, director of trauma services, Parkland Hospital; Charlotte Forswall, director of decision support, Parkland Hospital; and the Texas Institute for Health Policy Research.

*Uncompensated care is reported as the sum of bad debt and charity care charges from the 2001 Texas Department of Health/American Hospital Association annual survey of hospitals. Of the 367 acute general medical and surgical hospitals in Texas , 12 were excluded from analysis because of incomplete data.

Back to article


  1. Rickets TC, ed. Rural Health in the United States.   New York, NY: Oxford University Press; 1999:41, 141.
  2. Gaskin DJ. Safety Net Hospitals: Essential Providers of Public Health and Specialty Services . February 1999. Available at: Accessed March 29, 2004.
  3. Singer I, Kuzner J, Fagnani L. America's Safety Net Hospitals and Health Systems, 2000. Washington, DC: National Association of Public Hospitals and Health Systems; July 2002. Available at: Accessed March 29, 2004.
  4. Health and Human Services 2004 Poverty Guidelines. Available at: Accessed March 29, 2004.
  5. Employee Benefit Research Institute estimates of the Current Population Survey, March 2003 supplement. This represents 2002 data.
  6. Fishman LE. What types of hospitals form the safety net? Health Aff. 1997;16(40):215-222.
  7. Zuzkerman S, Bazzoli G, Davidoff A, LoSasso A. How did safety-net hospitals cope in the 1990s? Health Aff . 2001;20(4):159-168.
  8. Sloan FA, Vlavona J, Mullner R. Identifying the issues: a statistical profile. In: Sloan FA, Blumstein JF, Perrin JM, eds. Uncompensated Hospital Care: Rights and Responsibilities. Baltimore, Md: The Johns Hopkins University Press; 1986:25.

Dr Amarasingham, Robert Wood Johnson Clinical Scholar, Johns Hopkins University School of Medicine, Baltimore, Md; Ms Pickens, director, Strategic Planning and Population Development, Parkland Health & Hospital System, Dallas; and Dr Anderson, chief executive officer and president, Parkland Health & Hospital System, and professor of internal medicine, The University of Texas Southwestern Medical School at Dallas. Send reprint requests to Ruben Amarasingham, MD, Johns Hopkins University School of Medicine, 600 N. Wolfe St , Carnegie 291, Baltimore, MD 21287-6220; email: .  


Texas Medicine Main Page
Texas Medicine Back Issues

Comment on this (Must be logged in to comment)

Add Comment

Text Only 2000 character limit

Looking for more?