Carolyn M. Clancy M.D.
Agency for Health care Research and Quality
U.S. Department of Health and Human Services
AHRQ’s Research Efforts in Comparative Effectiveness
Committee on Ways and Means
the Subcommittee on Health
U.S. House of Representatives
Tuesday June 12, 2007
Good morning, Mr. Chairman and members of the Subcommittee. I am Dr. Carolyn Clancy, the Director of the Agency for Healthcare Research and Quality (AHRQ), an agency of the U.S. Department of Health and Human Services (HHS). I am very pleased to testify before you on the exciting issue of comparative effectiveness. I am thrilled about the growing interest in, and attention to, enhancing the role of comparative effectiveness research in our health care system. And I am particularly pleased to be able to tell you about AHRQ’s important efforts in this area of research.
This is a very interesting time in the history of health and medicine. Our investments in biomedical research have resulted in many new diagnostic and therapeutic options. Clinicians and patients can often now choose among an expanded array of choices for treating hypertension, heart failure, HIV, mental illness, and other chronic illnesses, and unprecedented innovations in diagnosis and prediction bring us closer to a vision of personalized health care than ever.
We also are beginning to reap the benefits from the advances in health information technology (health IT) that can bring this information immediately to clinicians, patients, and others when and where they need it. Health IT also is enhancing our research capacity and our ability to diffuse breakthroughs quickly and efficiently throughout the health care system. Health IT can make research a natural by-product of delivering health care.
While this brave new world of health care presents wonderful opportunities, it also creates challenges. Chief among them is how to evaluate these innovations and determine which represent added value, which offer minimal enhancements to current choices, which fail to reach their potential, and which work for some patients and not for others. The need to develop better evidence about the benefits and risks of alternative choices is imperative.
The mission of AHRQ is to improve the quality, safety, efficiency, and effectiveness of health care for all Americans. Effectiveness sits squarely in our mission – what is the right treatment for the right patient at the right time.
Comparative effectiveness research is a means to an end. Our mission is fulfilled when health care decision makers – including patients, clinicians, purchasers, and policymakers – use up-to-date, evidence-based information about their treatment options to make informed health care decisions. This goal was the inspiration for the creation of AHRQ by those Members of Congress who grasped the power of information to improve the health care system and the health of Americans.
Effective Health Care Program
AHRQ was granted authority under Section 1013 of the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 (MMA) to conduct and support evidence syntheses and research on topics of highest priority to Medicare, Medicaid, and the State Children’s Health Insurance Program (SCHIP).
I would like to thank Congress for its recognition of the programs and infrastructure that AHRQ has established for providing the health care system with the scientific evidence that it needs to provide safe, high quality effective health care. AHRQ’s Evidence-based Practice Centers (EPCs), Centers for Education and Research on Therapeutics, and other research programs have become a trusted, unbiased national source of information on health care diagnostics and treatments
These programs are an integral part of the AHRQ’s Effective Health Care program, which was created under the authority of Section 1013. AHRQ was able to establish the Effective Health Care program and begin work very quickly because of our solid, existing research enterprise.
The Secretary of HHS, Michael Leavitt, has established priorities for research conducted in the Effective Health Care Program by establishing high priority conditions that have a major impact for the Medicare, Medicaid, and SCHIP programs. To be effective, comparative effectiveness research must be relevant to its users. Decision makers are often faced with situations for which multiple different treatments are relevant at different times. For example, should a fifty-five-year-old woman with a scan showing greatly decreased bone density take drugs, increase Vitamin D and calcium intake, focus on weight-bearing exercise, or watchfully wait? We know that drugs are effective, but there is limited information on their long-term effects. Some women will develop kidney stones after increased calcium; current evidence does not allow precise formulation of an effective exercise prescription; and many women will never experience a fracture.
The key to success for this research is that it provides evidence that informs the choices confronting clinicians and patients and, where possible, should closely align with the sequence of decisions they face. As MMA Section 1013 directs, we also need to ensure that findings are frequently revisited, so they remain relevant and up-to-date. New evidence, such as a genetic test that identifies people at increased risk of untoward outcomes, affects comparative effectiveness and should be incorporated into these reviews at the appropriate times.
Under the statute, the Secretary of HHS is required to establish priorities, informed by a transparent priority-setting process that includes all stakeholders. Priorities for the Effective Health Care Program therefore are set after receiving broad public input through Federal Register notices, public listening sessions, and other means.
There was much discussion within HHS about how to approach these priorities. During our discussion of research on diseases or conditions, for example, we debated specific questions about treating diabetes and heart disease, and whether our research should center on particular medications and interventions, such as stents or proton pump inhibitors.
We decided to take a disease- and condition-based approach because, at the end of the day, that is how health care decisions are made. A patient comes to the health care system with a condition or disease, and all decisions, including how best to treat it, follow.
In December 2004, based on input from stakeholders, the Secretary of HHS identified 10 priority conditions—all of special significance to the Medicare program—to be the first addressed by the Effective Health Care Program. These conditions are:
- Arthritis and nontraumatic joint disorders
- Chronic obstructive pulmonary disease/asthma
- Dementia, including Alzheimer’s disease
- Depression and other mood disorders
- Diabetes mellitus
- Ischemic heart disease
- Peptic ulcer/dyspepsia
- Stroke, including control of hypertension
AHRQ’s Effective Health Care Program comprises three parts. The first capitalizes on effectiveness research conducted by AHRQ’s existing 13 EPCs, which were created in 1997. The EPCs develop comparative effectiveness reviews which focus on treatments for the priority conditions. These reports synthesize currently available scientific evidence, including both published and unpublished studies, comparing treatments, including drugs, to determine relative benefits and risks, and wherever possible, measure these outcomes for subpopulation groups. In addition, the EPCs identify major gaps in the existing knowledge base.
To help fill these gaps, AHRQ established the second part of the Effective Health Care program, called theDEcIDE network (Developing Evidence to Inform Decisions about Effectiveness), which will focus on conducting rapid-cycle research to address specific issues that do not necessitate larger, more time-consuming randomized clinical trials. The DEcIDE network consists of 13 research centers that have access to databases that contain clinical information for more than 50 million patients but do not identify them individually.
The third part of the Effective Health Care program is the John M. Eisenberg Clinical Decisions and Communications Science Center, based at the Oregon Health and Science University’s Department of Medicine. The Eisenberg Center was established to ensure that the findings of our comparative effectiveness research are translated into formats that are understandable for all potential users. The center—named the John M. Eisenberg Center in honor of AHRQ’s late director—assists in ensuring that effectiveness research leads to real-world quality improvements by translating complex scientific findings into understandable language for different audiences. The Center will help assure that reports are presented in formats that make them useful to a wide range of audiences and also will develop tools that encourage and empower consumers to make informed health care decisions.
An important hallmark of the Effective Health Care program is transparency in all aspects of the process. The transparency begins with the open process for setting research priorities, described earlier. The public and all interested stakeholders also have the opportunity to comment on the framing of specific research questions, as well as commenting on draft reports. In addition to the open invitation to comment, manufacturers are notified when a study is begun on one of their products and are invited to submit relevant studies and data.
Draft research questions and reports are posted on AHRQ’s Effective Health Care Program Web site (www.effectivehealthcare.ahrq.gov), and the Web site has a listserv that automatically notifies interested parties when draft questions or draft reports are posted.
Comparative Effectiveness Reviews
To date, AHRQ has released seven comparative effectiveness reviews. These reviews can be found on the Effective Health Care Program Web site, discussed above. They are:
Gastroesophageal Reflux Disease (GERD)
For management of gastroesophageal reflux disease, medications called proton pump inhibitors can be as effective as surgery in relieving the symptoms and improving quality of life.
Breast Cancer Diagnosis
Among women who receive an abnormal mammography findings or physical exams, four common noninvasive tests (magnetic resonance imaging, ultrasonography, positron emission tomography scanning, and scintimammography) are not accurate enough to routinely replace biopsies.
Managing Anemia In Cancer Patients
Among cancer patients undergoing chemotherapy or radiation, there is no clinically significant difference between epoetin and darbepoetin in the management of anemia. The drugs show no clinically significant difference in improving hemoglobin concentration and reducing the need for transfusion.
Non-steroidal anti-inflammatory drugs (NSAIDs) and COX-2 inhibitors present similar increased risks of heart attacks while offering about the same level of pain relief for patients with osteoarthritis. The exception is naproxen, which presents a lower risk of heart attack for some patients than other NSAIDs or COX-2 inhibitors.
Renal Artery Stenosis
Increasing numbers of patients with narrowed kidney arteries are undergoing vessel-widening angioplasty and placement of a tubular stent, but evidence does not show a clear advantage of that treatment over prescription drug therapy.
Off-Label Use Of Atypical Antipsychotics
Some newer antipsychotic medications approved to treat schizophrenia and bipolar disorder are being prescribed for depression, dementia, and other psychiatric disorders without strong evidence that such off-label uses are effective. Research is urgently needed for new treatments of dementia patients with severe agitation.
Today's most commonly prescribed antidepressants are similarly effective to first-generation antidepressants and provide relief to about six in 10 patients, but current evidence is insufficient for clinicians to predict which medications will work best for individual patients. Six in 10 patients experience at least one side effect, ranging from nausea to sexual dysfunction.
In January 2007, AHRQ released the first summary guide for consumers and clinicians derived from a comparative effectiveness report by the Eisenberg Center. The consumer report, titled Choosing Pain Medicine For Osteoarthritis, translates the information from the comparative effectiveness report on osteoarthritis drugs into language that will help consumers choose among their treatment options. The companion guide, Choosing Non-Opioid Analgesics for Osteoarthritis, further synthesizes the evidence into a resource that can help clinicians work with their patients to make informed decisions about treatments for osteoarthritis.
AHRQ has a series of upcoming reports that deal with critically important issues facing the health care system. They include:
- Medications for type 2 diabetes
- ACEIs (Angiotensin-converting enzyme Inhibitors) vs. ARBs (angiotensin II receptor antagonists) for high blood pressure
- Surgery vs. stents coronary artery disease
- Medications and other treatments (e.g., diet, exercise) for low bone density
Health Information Technology
I would like to mention briefly the role of health IT, which will make it easier for researchers to gather information for their research and for users of research findings to get information in real time when they need it. The health care system’s growing investments in health IT provide us with an unprecedented opportunity for redefining the possibilities of observational studies, accelerating and targeting the uptake of relevant information, and providing feedback to the biomedical enterprise itself.
Health IT will make it possible for research to answer the pressing questions facing the health care system more quickly and efficiently. In the future, health IT will provide us with the vehicle for transforming our health services research enterprise so that we can evaluate the effectiveness of interventions and treatments in real time as a byproduct of providing care.
AHRQ’s Fiscal Year 2008 budget request includes $15 million for a personalized health care initiative that will begin the infrastructure for a federated system of databases that can help answer critical comparative effectiveness questions. This system would enable researchers to match treatments and outcomes, and in that way learn from the nation’s day-to-day medical practice and improve safety and effectiveness of medical treatments.
Health IT also will greatly improve the ability to diffuse evidence and information more quickly throughout the health care system. For example, clinical decision support tools will make it possible to deliver relevant information to clinicians and patients, at the point of decision making. Most commonly envisioned as a pop-up reminder on a screen, clinical decision support should include information communicated directly to patients and caregivers at home—by phone, computer, or by other means.
As AHRQ has implemented the Effective Health Care program, we have some significant observations:
Priority setting: It is important to set clear priorities that meet the needs of all of the stakeholders in the health care system. Therefore, end users and stakeholders must continuously provide input through an open and transparent process.
Framing the research questions: Research must track closely with how clinicians and patients make health care decisions every day. The Secretary’s decision to use a disease- and condition-based approach to priorities embodies this perspective. It is also very important to recognize the importance of revising findings frequently to incorporate new evidence that may change the conclusions of what works best and for whom.
Balancing benefits and harms: Comparative effectiveness research must provide information on benefits and harms of a particular medication or intervention. Evaluating the balance of harms and benefits is a critical component of informed decision making. Few interventions are risk free, and for many chronic conditions the therapeutic goal is management of symptoms and disease state rather than cure. Often times, the decision comes with some assumption of harm—by both patient and clinician—but with the understanding that the benefits are worth that risk.
Research is a means, not an end: The ultimate goal of our research efforts is the development of timely, relevant information for decision making. This requires us to go beyond the products of traditional research, namely scholarly articles, and translate findings into language and formats that are appropriate for different audiences. Creating evidence and information that is not useful and accessible, or that does not take advantage of the latest communication technologies and vehicles, is a missed opportunity.
Trust as a process, not a structure: As has been stated, comparative effectiveness research can be a risky business, and there are winners and losers. Therefore, it is important that there be a level playing field among stakeholders. AHRQ’s Effective Health Care program has adopted a policy of transparency and inclusion. Manufacturers are notified when a study is begun, are invited to submit relevant studies and data, and have the opportunity—along with any other interested party—to comment on the framing of the specific research questions as well as draft reports. In addition, it is clear that the program’s success is dependent on effective collaboration with scientists from industry as well as academia. At the same time, we ensure that the authors of the comparative effectiveness reports are free of conflict to make sure that the results are not perceived as being biased in any way.
The question of trust also extends to the integral role that patients play in research. Although government and the private sector pay for research, patients assume the risks and benefits of enrolling in clinical trials and other studies. A question that is the subject of debate is whether study findings can ethicallybe kept secret from other researchers and patients themselves. We all need to learn from the knowledge gained in research, but it can be a matter of life and death for patients. We must move to an atmosphere where it is unacceptable to hold back research findings that may have an impact on the care that patients receive.
In conclusion, the U.S. health care system is poised to take advantage of advances in science and health information and communications technology in ways that have previously only seemed like something out of science fiction.
The need for valid, reliable, and accessible information on the comparative benefits and potential harms of treatment options has gained an urgency due to recent policies to promote the adoption of interoperable health IT, continued expansion of diagnostic and treatment options, increased consumer interest in health and health care decisions, and broad interest in improving value.
AHRQ’s Effective Health Care program is a model for how this vision can be achieved: A transparent, participatory approach that is driven by the needs of users and encourages broad engagement of stakeholders to mitigate any expected controversies and to expand opportunities for diffusion of findings of comparative effectiveness research. The Effective Health Care Program represents a foundation in which a larger investment in comparative effectiveness can be built.
Thank you very much and I would be pleased to answer any questions.
Last revised: April 19, 2011