Skip Navigation U.S. Department of Health and Human Services
Agency for Healthcare Research Quality
Archive print banner
AHRQ Annual Report on Research and Management, FY 2001

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

Goal 3—Costs, Use, and Access to Health Care

Addressing Challenges to Care

Adequate access to health care services continues to be a challenge for many Americans, particularly those who are poor, uninsured, minorities, rural residents, disabled individuals, and members of other priority populations. Also, continuing changes in the organization and financing of care have raised new questions about access to a range of health services, including emergency and specialty care. At the same time, examples of inappropriate care—including overuse and misuse of services—continue to be identified. Through ongoing development of nationally representative and more specialized databases, the production of public use data products, and research and analyses conducted by AHRQ researchers and AHRQ-funded researchers outside the Agency, we continue to address critical policy issues pertaining to the cost of health care, use of health care services, and access to care.

Examples of findings from recent AHRQ-funded research on health care costs, use of services, and access to care include:

  • In many cases, women who have mild to moderate pelvic inflammatory disease (PID) can be successfully treated as outpatients, which would result in substantial cost savings. PID affects more than 1 million U.S. women each year, with annual estimated direct and indirect costs of more than $4 billion. A recent study of more than 800 women with clinical signs and symptoms of mild-to-moderate PID found no apparent differences in rates of pregnancy, infertility, chronic pelvic pain, ectopic pregnancy, or tubal occlusion among women who were hospitalized and those treated as outpatients.
  • Hospital mergers may produce lower savings (expenses and revenues per admission) than previously estimated. Researchers examined changes in costs and prices for nearly 1,800 short-term hospitals from 1989 to 1997. They separated nonmerging hospitals into two groups: those that were rivals of the merging hospitals and those that were not competitors. When they compared merging hospitals in high HMO-penetration markets with their nonmerging rivals, the researchers found that the former group's average cost savings were only a modest 2.3 percentage points. Also, the average price growth of merging hospitals in high HMO-penetration markets was almost identical to that of their competitors. On the other hand, the researchers found that mergers in low HMO-penetration markets appear to produce greater cost and price savings for the merging hospitals.
  • Increasing AIDS drug assistance program benefits may reduce States' medical expenditures for HIV care. According to this study, if all States that have AIDS drug assistance programs with restricted access to protease inhibitors were to relax those restrictions, their costs would decline because hospital inpatient expenditures would decrease. Also, increasing access to protease inhibitors would reduce a State's total drug expenditures for public beneficiaries with HIV by lowering their need for prescription medicines to fight opportunistic infections. The researchers warn, however, that expanding these State programs would likely increase costs per public beneficiary.
  • Coronary angiography is underused for both Medicare managed care and fee-for-service heart attack patients. This study showed that Medicare patients enrolled in managed care plans are significantly less likely than those with traditional Medicare fee-for-service coverage to receive needed coronary angiography following a heart attack, a potentially life-saving procedure. These findings also bear out the underuse of the procedure in general; 66 percent of class I patients in Medicare managed care did not receive the procedure, compared with 54 percent of class I Medicare enrollees with fee-for-service coverage. Class I patients are those judged most in need of the procedure according to American College of Cardiology and American Heart Association guidelines.

BRIC Research on Health Care Costs

The Building Research Infrastructure and Capacity (BRIC) program launched by AHRQ in FY 2001 provided six awards involving institutions in nine States, including Idaho, Kentucky, Louisiana, Mississippi, Montana, Nevada, New Jersey, Utah, and Wyoming. Many of the projects support the development of partnerships between State agencies and universities to develop data systems useful in evaluating the effects of various programs on the cost and financing of health services. For example:

  • Rutgers Center for Health Services Research. This grant will enhance a partnership between the State University of New Jersey and State officials to develop and link State health data for use in addressing an array of health services research issues. Investigators will begin by examining the effectiveness of an innovative certificate-of-need strategy to improve access and reduce disparities in cardiac catheterization.
  • Intermountain BRIC Consortium. This project with the National Association of Health Data Organizations focuses on improving and linking State hospital discharge data with clinical data sets in Intermountain States to assess the economic consequences of policies that influence competition in health care markets.
  • Mississippi Building Research Infrastructure and Capacity. These researchers will examine the costs and use of health services in order to improve the delivery of primary care to rural, low-income populations in the Mississippi Delta area. The researchers will identify gaps in services and how these gaps could be addressed through telemedicine and increasing the efficiency of mobile care units.

Return to Contents

Medical Expenditure Panel Survey

AHRQ's Medical Expenditure Panel Survey (MEPS) provides highly detailed information on how Americans use and pay for health care. In addition to the core MEPS survey of households, it includes surveys of medical providers and establishments to supplement the data provided by household respondents on medical expenditures and health insurance coverage. The design of the MEPS survey permits both person-based and family-level estimates. The scope and depth of this data collection effort reflect the needs of government agencies, legislative bodies, and health professionals for comprehensive national estimates for use in the formulation and analysis of national health policies.

The MEPS collects data on the specific health services that Americans use, how frequently they use them, the cost of these services, and how they are paid for, as well as data on the cost, scope, and breadth of private health insurance held by and available to the U.S. population. This ongoing survey of about 15,000 households each year provides estimates for the country as a whole and for important priority populations. MEPS is unparalleled for the degree of detail in its data and its ability to link health service use, medical expenditures, and health insurance data to the demographic, employment, economic, health status, and other characteristics of survey respondents. Moreover, the MEPS provides a foundation for estimating the impact of changes affecting access to insurance or medical care on economic groups or special populations of interest, such as the poor, the elderly, veterans, the uninsured, and racial and ethnic minorities.

  • Databases. MEPS produces a number of analytical databases and, consistent with privacy policy, releases a number of databases to the public. These databases include demographic, health care use, access, expense, and insurance coverage information for all survey participants. Additional files detailing conditions, the specific content of health care events, and employment of household respondents also are made available to the public.
  • Printed data. In addition to providing databases for research use, AHRQ publishes MEPS data in tabular form on a range of topics. For example, each year AHRQ releases hundreds of tables on the health insurance coverage offered by employers. The data are available for the Nation as a whole, for important economic sectors, and for many States.
  • Web site. To maximize the use of this important investment, AHRQ has developed a Web site specific to the MEPS. This Web site rapidly disseminates databases and other products to the research community and quickly responds to inquiries from MEPS data users. In FY 2001, AHRQ was responding to more than 100 inquiries made through the Web site each month.
  • Training. To develop a cadre of sophisticated MEPS users outside of AHRQ, the Agency conducts a series of workshops, which range in length from a few hours to several days. They provide orientation to the policymaker and researcher about the range of questions that MEPS can answer and how the data can be properly used.
  • Data Center. AHRQ's Center for Cost and Financing Studies operates a Data Center to facilitate access to and use of MEPS data and answer questions from users.

How MEPS data are used:

  • In the public sector (e.g., Office of Management and Budget, Medicare Payment Advisory Commission, and Treasury Department): Government agencies and Congress rely on MEPS data to evaluate health reform policies, the effects of tax code changes on health expenditures and tax revenue, and proposed changes in government health programs such as Medicare.
  • In the private sector (e.g., RAND, Heritage Foundation, and the Urban Institute): Businesses, foundations, and academic institutions use these data to develop economic projections.
  • For research: These data represent a major resource for the health services research community at large. In the past year, data on premium costs from the MEPS Insurance Component have been used by the Bureau of Economic Analysis to produce estimates of the gross domestic product (GDP) for the Nation.

MEPS Products and Key Findings

Key findings, 2000:

  • In the first half of 2000, about 16 percent of the U.S. civilian noninstitutionalized population was uninsured.
  • Among those under age 65, Hispanics accounted for about 25 percent of the uninsured population in 2000, even though they represented only about 13 percent of the overall population for this age group.

Key findings, 1999:

  • Among adults under age 65, married people were more likely to have health insurance.
  • About 82 percent of Americans under age 65 had public or private insurance coverage.

Key findings, 1998:

  • More than 50 percent of elderly Americans had private insurance, but 40 only public coverage (Medicare with or without Medicaid).>
  • Less than half of all Hispanic Americans and about half of black Americans were covered by private health insurance, compared with three-quarters of whites.

Key findings, 1997:

  • More than one-third of young adults (aged 19-24) were uninsured in 1997.
  • Among all racial/ethnic groups, Hispanic males were the most likely to be uninsured; nearly 37 percent lacked coverage.
  • During the first half of 1997, almost 30 percent of children under age 4, 25 percent of those ages 4-6, and about 20 percent of children ages 7-12 had only public insurance coverage.

MEPS Household Component1

Key findings, 1996:

  • Inpatient hospital care accounts for nearly $4 of very $10 spent on health care.
  • Prescription medicines account for about 12 percent of total expenditures.
  • Almost 53 percent of children covered by Medicaid have a parent that works.
  • Only about 43 percent of the population received dental care in 1996.
  • Nearly 32 percent of Hispanics and 21 percent of blacks were insured in 1998, compared with only 12 percent of whites.

MEPS Insurance Component2

Key findings:

  • The average health insurance premium in 1996 was $1,997 for single coverage and $4,953 for a family; in 1999, the average premiums were $2,325 (single) and $6,058 (family).
  • Over the period 1996-1999, rates of contribution for employers and employees have remained the same.
  • In every State, establishments in large firms (50 or more employees) were more likely to offer health insurance than those in small firms (less than 50 employees).
  • For both single and family plans, more workers were covered by mixed-provider (PPO-type) plans than were covered by any-provider (conventional indemnity-type) plans or exclusive-provider plans (like HMOs). Enrollment rates in mixed-provider plans increased over the 1996-1999 period.
  • About 70 percent of U.S. establishments that offer health insurance offer only one plan; employers in Hawaii and California are most likely to offer their employees a choice of plans.

1. Full-year data have been released for 1996-1998; partial-year data have been released for 1999-2000.
2. Data are available for 1996-1999.

Return to Contents

Healthcare Cost and Utilization Project

The unprecedented volume and pace of change in the U.S. health care system, and the fact that changes are not occurring uniformly across the country, require a new information standard. We at AHRQ have long recognized the need for scientifically sound, standardized databases and tools for using them, as well as the need to make these resources available at the national, regional, and State levels. The Healthcare Cost and Utilization Project is one of many ways in which AHRQ is addressing this need.

HCUP is a Federal-State-industry partnership to build a standardized, multi-State health data system. This long-standing partnership has built and continues to develop and expand a family of administrative databases and powerful, user-friendly software to enhance the use of administrative data. Included in HCUP is hospital discharge information from State-specific hospital and ambulatory surgery databases, as well as a national sample of discharges from community hospitals. HCUP data are used at all levels to inform decisionmaking. HCUP continues to be a very valuable resource in light of recent findings that about 40 percent of personal health care expenditures in the United States go towards hospital care—making it the most expensive component of the health care sector.

State Inpatient Databases (SID)

The SID comprise non-Federal hospital discharge data from the participating States (see below), which represent about 67 percent of the over 22 million inpatient discharge abstracts in the United States.

New York*
South Carolina*
New Jersey*

* Participants in AHRQ's designated central distributor or single point of contact to facilitate access to their databases.

FY 2001 accomplishments include increasing the number of States participating in HCUP; now half (25) of all States are HCUP partners, an increase of roughly 15 percent over the previous fiscal year. New State partners were selected based on the diversity—in terms of geographic representation and population ethnicity—they bring to the project, along with data quality performance and their ability to facilitate timely processing of data.

Nationwide Inpatient Sample (NIS)

The NIS is the largest all-payer inpatient database in the United States. It provides information on about 7 million inpatient discharges from about 1,000 hospitals, including data from 1988-1999. According to NIS data:

  • About 135,000 hospital stays a year for treatment of depression, and alcohol- and substance-related mental disorders are not covered by either private insurance or public insurance programs such as Medicare and Medicaid.
  • Childbirth is the leading type of hospital care not covered by private insurance or public coverage. About 5 percent of all hospitalizations for childbirth—roughly 191,000 hospital stays a year—are uninsured.
  • Two chronic diseases, which if appropriately treated in primary care practices do not ordinarily result in hospitalization, also are among the top 10 types of uninsured inpatient care—asthma and diabetes. Together they account for 65,000 hospital admissions a year.

AHRQ also expanded HCUP beyond inpatient hospital settings to include hospital-based State ambulatory surgery databases (SASD). The number of States participating in the SASD increased from 9 in FY 2000 to 13 in FY 2001.

State Ambulatory Surgery Databases (SASD)

The SASD includes data on surgeries performed on the same day in which patients are admitted and released from hospital-affiliated ambulatory surgery sites.

South Carolina
New Jersey*
New York*

* Participates in AHRQ's designated central distributor or single point of contact to facilitate access to their databases.

Additionally, a pilot of emergency department databases was expanded from one to five States. The State Emergency Department Databases (SEDD) capture hospital-affiliated emergency department encounters from data organizations in participating States.

AHRQ recently announced the availability of the Kids' Inpatient Database (KID), the first comprehensive research database exclusively concerned with inpatient care of children and adolescents in the Nation's community hospitals. The KID is the only dataset on hospital use, outcomes, and charges for children age 18 and younger, including newborns, regardless of whether they are privately insured, receive public assistance, or have no health insurance. The KID contains national estimates for 6.7 million pediatric discharges and data on various hospital characteristics such as region, location (urban/rural), size, ownership, and pediatric hospital status.

During the past year AHRQ began a multifaceted effort to make HCUP data more accessible to researchers and other interested users. A centerpiece of this effort is HCUPnet, a free, interactive, menu-driven online service that allows easy access to national statistics and trends and selected State statistics about hospital stays.

HCUPnet answers questions about conditions treated and procedures performed in hospitals for the population as a whole, as well as for subsets of the population such as children and the elderly. In addition, 10 States have agreed to include their data in HCUPnet. About 4,000 visits are logged each month on HCUPnet, which can be found at

A second key component of our effort to facilitate researchers' access to HCUP data is the creation of a central distribution center for the State-level databases. Now researchers can go one-stop shopping instead of contacting each State on an individual basis.

Data from HCUP have been used to produce reports that answer questions on:

  • Reasons Americans are hospitalized.
  • How long they stay in the hospital.
  • The procedures they undergo.
  • How specific conditions are treated in hospitals.
  • The resulting outcomes.

In FY 2001, AHRQ launched an HCUP factbook series that is disseminated in print and through the AHRQ Web site. These factbooks were downloaded nearly 40,000 times in the first 6 months after they were posted on the Agency's Web site. Examples of information in the HCUP factbooks include:

  • The top five reasons for hospital admission are births, coronary arteriosclerosis, pneumonia, congestive heart failure, and heart attack.
  • Organ transplantation is associated with some of the longest and most expensive hospital stays.
  • Over one-third of all hospital admissions are through the emergency department.
  • The average charge for a hospital stay is over $110,000, and the average length of hospital stay is about 5 days.
  • Medicare and Medicaid are billed for about 54 percent of all hospital stays.

Return to Contents
Proceed to Next Section

The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care