January 26, 2009
Congressman Charles W. Boustany, Jr., M.D.
1117 Longworth House Office Bldg
Washington, DC 20515-1807
RE: Comments on Proposed Amendment to H.R. 1: Sec. 9203 – Limitation on Application of Government Supported Comparative Effectiveness Research
Dear Congressman Boustany:
The National Minority Quality Forum (the Forum) supports your intent to assure that findings from comparative effectiveness research or cost-effectiveness analyses supported with Federal funds, including funds provided by the Agency for Healthcare Research and Quality or funds provided under relevant titles of the Social Security Act, not be used by the Administrator of the Centers for Medicare & Medicaid Services as the basis for determining that an item or service is not reasonable and necessary.
The Forum has established as a priority the advancement of strategies that incentivize innovation and investment to create a healthcare research, financing, and delivery system that is inclusive, and that assigns equal priority to the outcomes of care provided to all populations.
The Forum, therefore, supports the language you have included in the proposed Amendment to H.R. 1: Sec. 9203 – Limitation on Application of Government Supported Comparative Effectiveness Research.
The Forum strongly encourages you to further strengthen the protections for historically underserved patient populations by adding the following language to your amendment:
- All comparative effectiveness findings, cost-effectiveness analyses, and related publications shall contain a clear statement that defines the patient cohorts for whom the research findings are valid and applicable in order to assure that CER findings and associated recommendations are clearly supported, and framed within the available evidence. Extrapolation of findings to populations for which the highest level of, or comparable clinical evidence, is not available is not valid, and constitutes differential treatment.
- The Administrator of the Centers for Medicare & Medicaid services must collect data and report beneficial and adverse changes in outcomes for all coverage decisions in a manner that clearly identifies the effect of those decisions on different age, gender, racial/ethnic, geographic and economic cohorts.
- Comparative effectiveness research or cost-effectiveness analyses must be powered to detect variations in treatment and outcomes across and within the different populations that are likely to receive the treatment in clinical practice. Without this ability, assessments may lead to sub-optimal or contraindicated care for individuals who may respond differently, including children, the elderly, and racial/ethnic populations that are not among the majority in the United States.
The Forum is committed to assuring that efforts to improve the quality of healthcare provided to populations that are statistical minorities in the United States are not compromised or delayed as the United States invests in initiatives to reduce the escalating costs of healthcare. The healthcare of all Americans must be given equal weight in the metrics that are used to calculate the value of the healthcare that is purchased with public and private funds.
The Forum supports a systems-oriented approach to eliminating economic and scientific barriers to investment in the innovation that is required to provide optimal health care for all populations in the United States of America. The Forum lauds Congress for its recognition of the need to assure that medications and treatments that are provided are clinically effective and appropriately reimbursed. The Forum is concerned, however, that current proposals for CER are based upon documented gaps in clinical evidence,[i] and the flawed premise that there is a single population “norm” against which health care quality standards can be established, thus eliminating inappropriate practice variation, and regional cost variation.
The Forum believes that current and proposed efforts to norm and/or standardize the quality of care must proceed with caution in light of the following:
- Well-documented lack of racial, ethnic, and even gender inclusiveness in randomized clinical trial cohorts;
- Changing population demographics. It is projected that by the year 2020, approximately 40% of the American population will be composed, collectively, of population groups that are currently defined as racial/ethnic minority populations, increasing to 50% of the American population by 2050;[ii]
- Well-documented biases in the manner in which health care is delivered to different racial/ethnic groups;
- Well-documented disparities in access to treatment resulting from uninsurance and under-insurance for different racial/ethnic groups; and
- Ongoing research into genomic and biological markers for risk of disease as well as differential responses to pharmacological agents.
The Evidence Challenge for U.S. Populations
In order to define and purchase safe and effective primary, secondary, and preventive health care services for all who reside and work in this country, the U.S. health care delivery and financing system must have evidence that is both reliable and valid. Randomized controlled trials (RCTs) are considered the gold standard in evaluating medical interventions, but RCTs general rely upon samples of people who are as homogeneous as possible to reduce variability. As a result, RCTs are not powered to provide scientific findings for the full range of patients that clinicians encounter in the course of their daily practices: they generally lack age, gender, racial and ethnic diversity. The historic practice has been to extrapolate the findings from these trials to the larger population, leaving the physician to interpret the results to determine the appropriateness of care for the individual patient. In the absence of RCTs whose sample cohort would represent the full diversity of America, this practice of generalization has merit when it is understood that the clinician will use best judgment to determine where the results increase the likelihood that a particular patient will receive optimal care. The absence of diversity in RCTs translates to a lack of the most scientifically valid evidence to inform CER studies for the diverse American population.
The Forum’s concerns are reinforced by reports issued by the NHS R&D Health Technology Assessment Programme[iii] for the UK’s National Health Service, and the Baylor College of Medicine Eliminating Disparities in Clinical Trials (EDICT) Project[iv]. The HTA report, The causes and effects of socio-demographic exclusions from clinical trials notes that, “In general, the exclusion from trials of those who are seen as ‘different’ or would require increased resources to be included…cannot be defended ethically and is against the principle of wide inclusion criteria to maximize generalizability of trial findings.”[v] The HTA report also notes, “Under-representation occurs, but in drug trials at least this may not always affect the external validity of relative effect estimates. However, measures of absolute effectiveness, absolute harm and cost-effectiveness are associated with underlying risk levels in different socio-demographic groups. Under-representation will therefore bias absolute effect estimates.”[vi] Most recently, Major Deficiencies in the Design and Funding of Clinical Trials, issued by EDICT in April 2008, reported that, “When it comes to the makeup of clinical trials, the National Institutes of Health requires that women and members of minority groups be included in all NIH-supported biomedical and behavioral research projects involving human subjects. Despite this requirement, little measurable improvement has been made in increasing clinical trials participation in such populations.”[vii]
Most importantly, the EDICT report states that, “Under-representation of specific populations in clinical trials is also at direct odds with the current state of medical science and drug discovery. With the successful sequencing of the human genome, scientists are faced with the new challenge of documenting, describing, and understanding the non-random pattern of human genetic variation and its link to disease risk in different patient groups. Findings from the large amount of genetic data generated to date show that more than 90 percent of the observed genetic variations occur within rather than between groups. This underscores the fact that ethnicity — which incorporates multiple variables including genetics, economic, social, dietary, religious, and linguistic background — has biomedical consequences when studying health outcomes.”[viii]
Quality Health Care: The Sine Qua Non of Sustainable Cost-Containment
The Forum is further concerned that the majority of proposals currently being considered by public and private policymakers appear to prioritize cost-containment or cost-savings over improvement of both the inputs to, and outcomes of, health care and health services. According to the CBO, “Getting to the point where additional research on comparative effectiveness could have a significant impact on health spending would probably take many years…it would probably be a decade or more before new research on comparative effectiveness had the potential to reduce health care spending in a significant way.[ix] The Forum believes that the only way to generate real, sustained cost-benefit for purchasers and payers of health benefits, and for individual patients, is for stakeholders to invest in the health of the American population by supporting equitable, effective care for all U.S. subpopulations, not simply the largest subpopulation cohort.
Innovation and New Research Will Improve Health Care Quality
The Forum is particularly concerned that approaches to comparative effectiveness research that rely primarily upon the synthesis of currently available information will serve only to codify existing disparities in the quality of the care that is available to racial and ethnic population groups, or, indeed, to any groups that were not represented in clinical trials. The use of existing studies to develop purchasing guidelines and/or practice standards will result in orientating the nation’s health care system towards providing optimal care for those cohorts included in the RCTs, but will put others at risk. Additional, the vast majority of CER studies are retrospective, and assess the potential of new technologies based upon criteria that are driven by the performance of older technologies and practices.
Given the real need for innovation and scientific inquiry to fill the knowledge void that currently exists for those historically underrepresented in clinical trials and observational studies, there are genuine concerns that CER studies will codify inequities in care and the resulting health status disparities by cloaking extrapolations drawn from inadequately powered studies in the mantle of science. This may serve not only to put the health of many Americans at risk, but also to stifle future investigations into new or more effective treatments. The Forum is equally concerned that new clinical trials or research to support comparative effectiveness analyses, if designed within the policies and regulations that currently govern and/or guide clinical trials and research, will continue the abysmal lack of inclusiveness that is the Achilles heel of current clinical evidence.
AHRQ has issued a wide range of comparative effectiveness reviews (including for prostate cancer, hypertension, and diabetes) that acknowledge the absence of evidence from which can be drawn valid or clinically defensible conclusions about the effectiveness of therapies in “subpopulations”, presumably meaning those not included in the clinical trials — the elderly, women, and populations that are in the minority in the United States.[x],[xi],[xii] AHRQ, however, has made it a practice to issue press releases and other public pronouncements that have the unintended effect of obscuring these profound clinical limitations. The AHRQ draft methodological guidance for conducting comparative effectiveness reviews states that, “To assess effectiveness of other interventions, such as the efficacy of a drug, reviews may focus on the results of randomized controlled trials. For other questions, or to compare results of trials with those from everyday practice, observational studies may play a key role[xiii].” The Forum notes that, in order to be included in any type of observational study, it is a given that a patient must have access to the health care delivery system — access that is compromised by multiple factors for many of the populations that are not generally included in clinical trial cohorts. Further, a case can be made that researchers must have an incentive to, or interest in, studying inclusive or diverse patient cohorts. The lack of diversity in the researcher populations has been linked to lack of diversity in patient cohorts and areas of inquiry.
The Forum stands ready to assist you. If require additional information, please do not hesitate to contact me, or Ms. Gretchen C. Wartman, the Forum’s Vice President for Policy, at firstname.lastname@example.org or 202-223-7560.
Gary A. Puckrein, PhD
President and CEO
cc: Mike Thompson, Legislative Assistant, Office of Congressman Boustany
[i] Obtained either through randomized clinical trials or observational studies.
[ii] The demographics of the US population are shifting away from a single racial/ethnic group majority. According to the US Census Bureau’s 2005 mid-decade estimates, approximately 1-in-every-3 U.S. residents is part of a group other than single-race non-Hispanic White. In 2005, the nation’s minority populations totaled 98 million, or 33 percent, of the country’s total of 296.4 million. It is projected that by the year 2020, approximately 40% of the American population will be composed, collectively, of these racial/ethnic minority population groups, increasing to 50% of the American population by 2050. Given the relatively younger ages of the minority population groups, it can be assumed that an increasing percentage of the American work force and school age children will be members of population groups that are currently classified as non-Hispanic single race Caucasian.
[iii] One of the programs that informs the NICE decision-making process is the NHS R&D Health Technology Assessment Programme.
[iv] EDICT is a research partnership between the Intercultural Cancer Council and the Baylor College of Medicine in Houston that aims to improve the participation of minorities and undeserved patients in clinical research trails.
[v] Bartlett C, Doyal L, Ebrahim S, Davey P, Bachmann M, Egger M, et al. The causes and effects of socio-demographic exclusions from clinical trials. Health Technol Assess 2005; 9(38), page 100.
[vi] Ibid, page iv.
[vii] Chronic Disease Prevention and Control Research Center at Baylor College of Medicine, in collaboration with the Intercultural Cancer Council. Major Deficiencies in the Design and Funding of Clinical Trials: A Report to the Nation Improving on How Human Studies Are Conducted. Findings of the Eliminating Disparities in Clinical Trials Project (EDICT), April 1, 2008, page 8.
[viii] Ibid, page 9.
[ix] Op. cit. Peter R. Orszag. Letter to the Hon. Pete Stark.
[x] Wilt TJ, Shamliyan T, Taylor B, MacDonald R, Tacklind J, Rutks I, Koeneman K, Cho C-S, Kane RL. Comparative Effectiveness of Therapies for Clinically Localized Prostate Cancer. Comparative Effectiveness Review No. 13. (Prepared by Minnesota Evidence-based Practice Center under Contract No. 290-02-0009.) Rockville, MD: Agency for Healthcare Research and Quality, February 2008. Available at: www.effectivehealthcare.ahrq.gov/reports/final.cfm.
[xi] Matchar DB, McCrory DC, Orlando LA, Patel MR, Patel UD, Patwardhan MB, Powers B, Samsa GP, Gray RN. Comparative Effectiveness of Angiotensin-Converting Enzyme Inhibitors (ACEIs) and Angiotensin II Receptor Antagonists (ARBs) for Treating Essential Hypertension. Comparative Effectiveness Review No. 10. (Prepared by Duke Evidence-based Practice Center under Contract No. 290-02-0025.) Rockville, MD: Agency for Healthcare Research and Quality. November 2007. Available at: www.effectivehealthcare.ahrq.gov/reports/final.cfm.
[xii] Bolen S, Wilson L, Vassy J, Feldman L, Yeh J, Marinopoulos S, Wilson R, Cheng D, Wiley C, Selvin E, Malaka D, Akpala C, Brancati F, Bass E. Comparative Effectiveness and Safety of Oral Diabetes Medications for Adults With Type 2 Diabetes. Comparative Effectiveness Review No. 8 (Prepared by Johns Hopkins Evidence-based Practice Center under Contract No. 290-02-0018). Rockville, MD: Agency for Healthcare Research and Quality. July 2007. Available at: www.effectivehealthcare.ahrq.gov/reports/final.cfm.
[xiii] Agency for Healthcare Research and Quality. Methods Reference Guide for Effectiveness and Comparative Effectiveness Reviews, Version 1.0 [Draft posted Oct. 2007]. Rockville, MD. Accessed May 14, 2008 at http://effectivehealthcare.ahrq.gov/.