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 It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work.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: https://info.ahrq.gov. Let us know the nature of the problem, the Web address of what you want, and your contact information.
Please go to www.ahrq.gov for current information.
John M. Eisenberg, M.D., Administrator, AHCPR
Before the Senate Labor and Human Resources Subcommittee on Public Health and Safety, April 30, 1998
Building Evidence on What Works
Measures for Choice
Measures for Improvement
Mr. Chairman, thank you for giving me the opportunity to address the Committee on the very timely and important issue of health care quality, and more specifically, AHCPR's role in helping to ensure that Americans get the best health care available.
We all know that the health care system is changing very rapidly and very dramatically. Change is unsettling, and questions and concerns have been raised about the quality of health care services provided in this country as a result of this change. We are all looking for answers. Patients and health care practitioners need the best scientific evidence so they can make informed choices about treatment alternatives. Health plan managers, purchasers, and policymakers need to know the evidence on the most effective ways to organize, finance, and manage the delivery of health care. The pharmaceutical, biotechnology, disease management, and equipment and device industries need to know how to best focus their research investments and design new products. And we all need to know how to recognize and choose high quality health plans and practitioners.
AHCPR addresses these concerns in two ways. One, we build the evidence of what works and doesn't work in everyday practice; and two, we help use this evidence to create the knowledge to measure and improve the quality of care.
Building Evidence on What Works
Our research complements the work of the National Institutes of Health (NIH) by enabling patients and society to reap the full benefits of our investment in biomedical research. Biomedical and health services research are inextricably linked in the continuum of health research. Biomedical research identifies the mechanisms of disease that can be interrupted by prevention, early diagnosis, and treatment. Health services research measures the effectiveness of the services that deliver the preventive, diagnostic, and therapeutic care; compares them with existing practice; and evaluates the ability of the health care system to deliver them efficiently. Health services research asks practical but tough questions:
- Are these new interventions more or less effective in day-to-day practice?
- How much does the improvement in clinical care cost?
- How will each of these alternative treatments affect the ability of patients to live their lives the way they want?
- How do these services and the way in which they are provided improve the quality of health care and how will they improve the quality of life of the people who receive them?
The history of our Nation's improvements in stroke prevention is just one example of this continuum of research. NIH-funded research at the University of Wisconsin led to the development of warfarin, an important drug that prevents strokes. Despite the great promise of this major biomedical development, AHCPR research showed that fewer than 25 percent of eligible patients actually receive the drug in practice because most doctors fear that their patients may experience bleeding complications. If the drug was used by patients at risk of stroke, we could prevent 40,000 strokes a year and reduce expenditures by hundreds of millions of dollars. AHCPR is now supporting a clinical trial to determine the most effective way to administer warfarin so that physicians have the confidence to prescribe it, and the right patients get it.
Another example is the effort to prevent low-birthweight infants, who are often born prematurely. There is very good evidence that administration of corticosteroids before birth is very effective in reducing illness in babies, yet as recently as 4 years ago, only 20 percent of women delivering premature babies were getting this treatment. To combat this problem, NIH sponsored a consensus conference to highlight the evidence. AHCPR-sponsored researchers not only identified why undertreatment was occurring (for instance, physicians tended to misunderstand the indications for treatment and withheld it, thinking it was not effective under those circumstances) and the poor outcomes associated with undertreatment, but clarified the actual indications for use of corticosteroid. After disseminating their findings widely, the researchers found that almost 70 percent of women delivering preterm births—a three-and-a-half-fold increase—had received corticosteroids to prevent illness in their babies after birth. Their findings provided some of the information that led to new recommendations by the Centers for Disease Control and Prevention (CDC) about screening and treating group B strep infections in pregnant women.
As these examples suggest, much of our research is directly relevant to improving the quality of care provided to Medicare and Medicaid patients. To make that link even clearer, we have begun discussions with the Health Care Financing Administration (HCFA) to identify better mechanisms for translating our Medicare relevant research findings into practice and ways that we can work more collaboratively with the Quality Improvement Organizations as laboratories for facilitating quality improvement strategies. We are also looking at ways we can expand our collaboration in the areas of quality measurement and assessment of Medicare beneficiary satisfaction, building upon our work on the Consumer Assessment of Health Plans study.
The real news story is that our research is continuing to demonstrate that better quality can cost less. I have included an addendum to my statement that provides a number of examples that are relevant to Medicare, Medicaid, and other programs. But let me mention just two more.
The first is the use of thrombolytic drugs in patients with a recent heart attack. Despite a number of clinical trials showing that this therapy was effective, our researchers found that it was underused in Medicare patients. Because a major reason for underuse was uncertainty by physicians about which patients would benefit most, our researchers developed an algorithm that is an accurate predictor of who is likely to benefit from thrombolytic therapy. Since thrombolytic therapy is often administered in the emergency room and physicians might not have time to actually go through the steps of the decision rule, AHCPR-sponsored researchers went one more step and actually programmed the algorithm into an electrocardiograph (EKG) device, so that when an emergency room physician reviews the EKG readout on a probable heart attack victim, the physician simultaneously receives a real-time prediction of whether the patient will benefit from immediate administration of thrombolytic therapy. That device is now awaiting approval by the Food and Drug Administration (FDA).
Last year one of our grantees developed a simple and accurate method to predict which patients with pneumonia may be treated at home rather than in a hospital. The prediction method—a clinical model used to help doctors assess the need for hospitalization—could reduce the over $4 billion spent annually for inpatient care. Within months, this finding and related work by this research team was being promoted by Washington area HMOs, was being used in 31 States by Medicare's Quality Improvement Organizations and several Veterans Administration hospitals. Such quick adoption of research findings on what works reflects the tremendous appetite of those who deliver care to do what is best for their patients. But we have only begun to scratch the surface in building the science base they need.
As these examples show, a major focus of our work is to close the gap between what we know and what is done in practice. Since my last appearance before you, our outcomes research projects on schizophrenia released its findings. The study will provide clinicians, patients, and their families the basic, evidence-based information they need to formulate a comprehensive treatment plan for their patients with schizophrenia. During my previous appearances before this Subcommittee, I described many related initiatives and will only briefly reference them here. They include:
- Our 12 Evidence-based Practice Centers, which will provide state-of-the-art scientific assessments of the evidence supporting the effectiveness of a particular treatment, technology, or procedure.
- Our National Guideline Clearinghouse™, a Web-based compendium of clinical practice guidelines which I demonstrated to you.
- Our new responsibility, granted by the Committee, to establish Centers for Education and Research Therapeutics. These Centers will increase the awareness of new uses and risks of medical products and provide better information about possible adverse effects.
Building the evidence base is only one step toward helping to ensure quality of care. AHCPR's second role is to use this evidence to create the knowledge to measure and improve the quality of care. This includes developing measures to enhance choice and improve quality. While the two are interconnected, the measures used to achieve quality may be very different from those that are used to inform choice.
Return to Contents
Measures for Choice
Our decisions about health care are very personal ones. Although many of our health care decisions are made with help—maybe from a doctor, a nurse or a loved one—they remain personal and individual decisions. But these decisions occur in the context of one of the most complex areas of modern life—a health care system where patients and clinicians often don't have the information they need to make the best choices among the alternatives available to them.
Measures for choice, such as report cards and satisfaction surveys, allow purchasers and patients to judge the quality of health care and to comparison-shop. When we decide on a health plan and a clinician or hospital we deserve the kind of information on quality we can get when choosing a car, a home, and most other products and services that we use our limited resources to buy. It is important that we offer people not just technical measures of quality, but measures that reflect the preferences and values of patients.
We need to develop more methods, and better methods, for measuring quality information for choice, and findings ways to report the results to consumers. AHCPR's Consumer Assessments of Health Plans survey is a case in point. CAHPS® is a series of questionnaires designed to be used by public- and private-sector health plans, employers, and other organizations to survey their members and employees. The information from CAHPS® questionnaires, presented in the CAHPS® reports, can help consumers and group purchasers compare health plans and make more informed choices based on quality.
CAHPS® has been tested by States, and by private employers, such as the Ford Motor Company. Ford's Department of Health Care Quality is testing CAHPS® in two markets, and CAHPS® is also being used by five large health plans. In the near future, you will get an opportunity to use CAHPS® to pick health care coverage for you and your family. The Office of Personnel Management will be using CAHPS® to help Federal employees select health plans based on information on quality beginning in 1999.
Return to Contents
Measures for Improvement
In addition to assisting consumers choose plans and patients choose doctors and health plans, information on the quality of care can help providers, health plans, and health systems identify opportunities for improvement. AHCPR plays a three-step role in the development of measures for quality improvement. One, we build the evidence base for measurement through our outcomes and effectiveness research. Two, we develop the measures ourselves, or work with other public or private sector measure developers, and three, we provide the means for disseminating the measures into practice.
For example, an AHCPR-supported study found that only 21 percent of eligible Medicare patients who suffered acute myocardial infarction (AMI) received beta blockers. Yet the mortality rate among those treated with beta blockers was 43 percent less than those not offered this treatment, and patients who receive beta blockers are rehospitalized for heart ailments 22 percent less often than those who do not get beta blockers. At HCFA's request, the National Committee for Quality Assurance used the findings of this study as the basis for changing the performance measurement for beta blocker use after AMI to include patients over 75 years of age in HEDIS 3.0. HEDIS is being used by HCFA to assess the quality provided by Medicare HMOs.
Another AHCPR-supported study developed measures to use specifically for getting self-reported health information from children, an area where quality measures are lacking. Dr. Sherrie Kaplan and her colleagues at the New England Medical Center in Boston developed interactive computer-based systems that use clear language that enable children with chronic disorders such as juvenile diabetes to report on how they're feeling in ways that can be used for research purposes.
We also are using the science we support to develop the "next generation" of quality measures. New measures must be developed constantly to assess the changing environment of health care. For both purposes of quality measurement—for choice and for improvement—it is important to adjust these measures for the severity of illness of the population being measured. The absence of such an adjustment can result in serious unintended consequences. High-quality health plans might attract the sickest patients and incur greater costs without recouping greater reimbursement. Equally as important, but less often recognized, is the need to adjust quality and outcome measures for severity and complexity of disease. Otherwise, poorer outcomes that are caused by a clinician's or a hospital's willingness to accept sicker patients could be misinterpreted as poorer quality care. Researchers are just beginning to discover ways of adjusting both costs and outcomes for severity, and recognizing that different factors may affect these two aspects of care.
In my view, AHCPR should, in partnership with the private sector, ensure that the science of performance measurement matures in a way that promotes effective, efficient, and reliable measurement and reporting. How do we hope to achieve this goals? By developing, testing, and using evidence-based measures and using the results to improve quality either through consumer choice strategies or quality improvement projects.
Mr. Chairman, we are making progress toward an effective system of quality measurement. We know that consumers, physicians, and the health care system as a whole need information on what works and what doesn't work in health care. This information is critical and essential to improving the quality of care provided in this Nation. Thank you.
Return to Contents
Current as of April 1998