NINDS Stroke Disparities Advisory Panel Meeting
November 7-8, 2002
Table of Contents
In 2001, the NINDS charged its Stroke Progress Review Group (Stroke PRG) with the task of setting overall priorities for stroke research. Because racial and ethnic disparities in the stroke burden present a critical research and treatment challenge that must be addressed, the Stroke PRG in its April 2002 report discussed the need to develop a research agenda on stroke disparities. In response, NINDS formed a Stroke Disparities Advisory Panel, which convened a two-day meeting, on November 7-8, 2002. During this meeting, nine panels of experts were formed to discuss and consider recommendations for advancing research on stroke disparities.
Stroke ranks as the third leading cause of death and is the leading cause of serious, long-term disability in the United States. Racial and ethnic minority populations, in some age groups, have a higher relative risk of stroke death when compared with the U.S. non-Hispanic white population. For example, among African-Americans, the relative risk of stroke death is 4 times higher at ages 35-54, 3 times higher at ages 55-64, and almost 2 times higher at ages 65-74. The disparities in stroke mortality rates closely parallel racial and ethnic differences in the prevalence of stroke risk factors. For example, compared to white Americans, African-Americans have higher rates of hypertension, diabetes, smoking, obesity, and physical inactivity. These differences often persist even after statistical adjustments are made for differences in the socioeconomic status.
A number of possible mechanisms involved in stroke disparities have been hypothesized. They include a wide range of cultural and environmental factors, such as racial and ethnic variations in lifestyle, access to healthcare, quality of healthcare received, differences in health beliefs, religiosity, health literacy, adherence to prescribed therapy, stress, and exposures to environmental toxins. In addition, recent progress in stroke genetics has enabled exploration of possible racial and ethnic differences in genetic susceptibility to stroke or stroke risk factors. The panels reviewed the state of scientific evidence on the influence of environmental and genetic factors, and their interactions, on stroke incidence and stroke outcomes.
The discussions regarding current state of research on stroke disparities were conducted by nine panels, each focusing on one of the following topics:
1. Temporal Trends in Disparities in Stroke Incidence, Mortality, and Quality of Life
2. Disparities in Stroke Risk Factors and Mechanisms
3. Access to Health Care/Quality of Care
4. Genetics and Vascular Biology
5. Design of Clinical Trials Addressing Stroke Disparities
6. Bridges to the Community: Assuring Ethical Conduct of Studies and Data Integrity
7. Primary and Secondary Stroke Prevention
8. Acute Stroke Treatment
Each panel summarized the state of current knowledge within their field of expertise and identified barriers and challenges that hamper our understanding of stroke disparities. The panels specified data needs and scientific standards of evidence necessary to design effective interventions and to develop reliable methods of measuring progress toward the elimination of stroke disparities. Several common themes have emerged across the panel discussions:
- Data needs
The panels have consistently noted the urgent need to collect more data on racial and ethnic stroke disparities, going beyond the well-documented differences in stroke mortality. Racial and ethnic disparities in stroke mortality rates may reflect disparities in incidence rates or case-fatality rates, or both. However, stroke incidence and case fatality data from population-based studies that include substantial numbers of minorities are limited. Data regarding temporal trends in stroke incidence in minority populations are particularly important for measuring progress toward elimination of stroke disparities.
Evidence to date has suggested possible racial and ethnic differences in the incidence of various stroke subtypes. For example, existing data suggest greater risk of lacunar and intracranial atherosclerotic stroke in African-Americans and Hispanics. Therefore, the panels have emphasized the need to collect further data on racial and ethnic differences in the incidence of different stroke subtypes and the relative contribution of different risk factors to increased rates of various stroke subtypes.
There is strong and consistent evidence of low levels of implementation of established guidelines for primary and secondary stroke prevention by healthcare providers, and low levels of patient adherence to physician-prescribed medical therapy or changes in lifestyle. These shortcomings have greatly reduced the potential of existing evidence-based knowledge to reduce the stroke burden. However, little information is available on racial and ethnic differences in the quality of preventive care and patient compliance as mechanisms contributing to the disparities in stroke burden.
Numerous studies have documented racial and ethnic differences in medical care received for various conditions. For example, evidence shows that African-American cardiac patients are less likely than white American patients with the same condition and similar characteristics to receive diagnostic procedures, revascularization procedures, and thrombolytic therapy. In general, disparities in receipt of appropriate care remain after adjusting for factors known to affect care, such as age, sex, insurance status, co-morbidities, and disease severity. Despite the perception that there are racial disparities in acute stroke treatment, currently little is known about possible ethnic and racial differences in the treatment of acute stroke. For example, it is not known whether there are racial and ethnic differences in the use of thrombolytics for acute stroke treatment. Similarly, very little information exists on possible racial and ethnic differences in the rate of recovery from stroke, stroke recurrence, and functional outcomes and quality of life following stroke. Previous studies have shown that functional outcome after ischemic stroke is affected by many factors. Those most commonly reported include age, pre-stroke disability, severity of stroke and/or level of consciousness on presentation, presence or absence of urinary incontinence after stroke, previous stroke, diabetes or elevated serum glucose, cardiac disease, and degree of social support or marital status. However, in the majority of these studies, race was not considered as a variable potentially affecting post-stroke outcome.
- Challenges encountered in research on stroke disparities.
The panels consistently emphasized the need to improve and standardize race and ethnicity definitions and classifications. Definitions of race-ethnicity have been mandated by the government organizations, but may be inadequate and differ across available studies. These definitions were based on US census methods and do not take into account cultural distinctions, heterogeneity among race groups, ethnicity, heritage, and the effects of inter-marriage.
There are major confounding factors when examining racial/ethnic disparities in the burden of stroke, including socioeconomic status, education, religion, cultural factors, dietary patterns, geographic region, and gender. These factors complicate study design and limit our ability to generalize results from one study to the entire U.S. population.
Stroke is a heterogeneous disease and comprises a number of pathological conditions. The phenotypic characterization of stroke in research is a controversial issue that will require a diversity of approaches. While there is consensus that ischemic stroke, intracerebral hemorrhage, and subarachnoid hemorrhage should be addressed separately, an important question is whether there are subphenotypes or intermediate phenotypes that have a particular relevance to racial and ethnic differences in stroke risk.
There are many operational challenges to conducting research in different racial and ethnic groups. The lack of legal protection against genetic discrimination in the United States, and general mistrust among minorities toward the medical and research establishment, are substantial impediments to the inclusion of racial and ethnic minorities in biomedical, especially genetic research. Participation in clinical trials is declining in numerous disease areas. Unfortunately, recruitment strategies have focused on problem-solving strategies rather than on proactive, specific strategies targeted to reach historically underserved populations.
- Epidemiological research
Characterize temporal trends in racial and ethnic disparities for stroke mortality, incidence, and prevalence rates by age, gender, social class, social environment, and geography.
Identify population attributable risks and interactions among known risk factors and emerging stroke risk factors in multicultural populations, accounting for differences in sex and socioeconomic status.
Determine whether intermediate markers of stroke risk are valid markers in different racial and ethnic populations and potential targets for prevention trials.
- Research on genetic and environmental determinants of stroke risk
Provide characterization of variation in candidate genes and pathways within each racial and ethnic group, based on adequately powered studies within each racial and ethnic group, and with appropriate consideration of the potential for population stratification bias.
Examine how genetic and environmental factors and their interactions contribute to the incidence of stroke risk factors and subtype-specific strokes.
Study racial and ethnic differences in response to drugs used to treat acute stroke, as well as those used for primary and secondary prevention.
Focus translational research on similarities and differences between racial and ethnic groups in important vascular biology mechanisms, including endothelial function, biomarkers of inflammation, coagulation/fibrinolysis, angiogenesis, and oxidative stress. Further characterization of these phenotypes between racial and ethnic groups must accompany genetic studies of these biological mechanisms.
- Health services and patient management research
Evaluate the independent contributions of factors relating access to quality care and stroke health disparities, such as health insurance status, acculturation, religious and spiritual factors, and healthcare provider attitudes. Research focused on prejudice and cultural competency on the part of health care professionals and their influence on access to care and stroke disparities should be undertaken.
Evaluate barriers to adherence to stroke prevention strategies that are specific to minority groups. Develop and test for effectiveness of novel cost-effective and practical stroke prevention programs for multicultural groups that can be adapted for use within communities.
Collect data on access to acute stroke care among racial and ethnic subgroups. Evaluate the extent to which socioeconomic and cultural factors, level of public awareness, physician biases, and patient co-morbidities may influence acute stroke care.
Using multidisciplinary research and considering a variety of sociological, environmental, and access variables, conduct research to learn how race affects access to and quality of stroke rehabilitation services. Investigate whether racial differences occur in the recovery trajectory and in optimal recovery after stroke; and determine whether disparities exist in other long-term sequelae of stroke and what factors influence these outcomes.
Study diverse racial and ethnic groups in different geographic, social, and cultural environments. Cross-cultural studies or migration studies of racial/ethnic groups are useful scientific strategies to disentangle the role of race as a confounder reflecting both environmental and genetic differences. The effect of race/ethnicity on disease outcome should be examined after stratifying on numerous potential confounders, including measures of socioeconomic status, education, and health care access and utilization.
The need to increase minority representation in clinical trials remains critical. In addition to increasing sample size of minority groups, several strategies were suggested to increase the statistical power to characterize race as an effect modifier. For example, statistical power can be increased by the use of surrogate endpoints, such as a composite endpoint of clinical or silent (MRI) strokes. Another approach might be to expand the sample size to test race as an effect modifier, while minimizing the complexity of the study. In addition, the panels urged standardization of data elements on stroke risk factors and stroke outcomes, and data sharing across studies.
The foundation toward characterizing the public burden of stroke it to develop a national surveillance system to establish and compare incidence and prevalence rates for stroke, both overall and by stroke subtype, for all major racial and ethnic subgroups. This system should include population-based studies in multiple well-characterized communities in different geographic regions throughout the United States. The surveillance system should employ common methodology and definitions.
Develop additional measures of race and ethnicity that can complement current methods of self-reporting and ensure consistency in reporting race across epidemiologic studies.
Strengthen the regulatory environment that protects against genetic discrimination of individuals and named populations.
Develop Stroke Prevention Centers of Excellence, to test interventions aiming to improve screening for risk factors, increase implementation of risk factor management guidelines by healthcare organizations and healthcare providers, and to support patient adherence. Such interventions should be designed by multidisciplinary teams, drawing on experience from diverse fields of research on healthcare delivery, healthcare outcomes, and health behavior.
A more proactive approach to minority recruitment was recommended. Several strategies were suggested, including providing diversity training for investigators to increase their awareness of and sensitivity to cultural norms of target populations; creating a "special populations" advisory group to work with researchers; and making modifications in the NINDS guidelines to establish research requirements for community engagement.
Philip B. Gorelick, MD, MPH (Rush-Presbyterian-St. Lukes Medical Center)
Robert J. Adams, MD, MS (Medical College of Georgia)
Carol M. Allen, PhD, RN (Oakwood College, Huntsville, AL) not attending
Carol Ann Barch, MN (University of Pittsburgh)
Hal V. Barron, MD (UCSF/Genentech)
Ronny A. Bell, PhD, MS (Wake Forest University School of Medicine)
Oscar Benavente, MD (University of Texas)
Bernadette Boden-Albala, MPH, DrPH (Columbia University)
Eric A. Boerwinkle, PhD (University of Texas) not attending
Gloria J. Bonner, PhD, RN (University of Illinois)
Joseph P. Broderick, MD (University of Cincinnati)
Askiel Bruno, MD (Indiana University)
Michelle L. Casper, PhD (CDC) not attending
Patricia C. Clark, PhD, RN (Emory University)
Edward Cooper, MD (University of Pennsylvania)
Marco R. DiTullio, MD (Columbia University)
Pamela W. Duncan, PhD, PT (University of Florida)
Cheryl E. Easley, PhD, RN (Saginaw Valley State University)
Mitchell S.V. Elkind, MD, MS (Columbia University)
John M. Flack, MD, MPH (Wayne State University)
Karen L. Furie, MD, MPH (Harvard University)
James M. Galloway, MD (University of Arizona)
Moises Gaviria, MD (University of Illinois)
Gary H. Gibbons, MD (Morehouse School of Medicine)
Wayne H. Giles, PhD (CDC)
Daniel F. Hanley, MD (John Hopkins) not attending
Yvonne Harris, MPA, CCRA (Rush- Presbyterian- St. Lukes Medical Center)
David C. Hess, MD, MS (CDC)
George Howard, DrPH (University of Alabama at Birmingham)
Virginia J. Howard, MSPH (University of Alabama)
S. Claiborne Johnston, MD, PhD (UCSF)
Edgar J. Kenton, III, MD (Mainline/Jefferson Health System)
Brett M. Kissela, MD (University of Cincinnati) not attending
Steven J. Kittner, MD, MPH (University of Maryland)
Selma C. Kunitz, PhD (KAI)
Richard Levy, PhD (National Pharmaceutical Council)
George E. Locke, MD (King/Drew Medical Center)
Peter R. MacLeish, PhD (Morehouse School of Medicine)
Ann M. Malarcher, PhD (CDC)
James F. Meschia, MD (Mayo Clinic)
Elaine L. Miller, DNS, MN, BSN (University of Cincinnati)
Lewis B. Morgenstern, MD (University of Michigan)
Karen L. Parko, MD (US Public Health Service)
George W. Petty, MD (Mayo, Rochester)
David J. Pinsky, MD (Columbia University)
Beth E. Quill MPH (University of Texas, Houston)
DeJuran Richardson, PhD (Rush-Presbyterian- St. Lukes Medical Center)
Charles N. Rotimi, PhD (Howard University)
Ralph L. Sacco, MD, MS (Columbia University)
James P. Stansbury, PhD (Veterans Health Administration)
Deborah R. Summers, RN, CS, MSN (Mid America Brain and Stroke Institute)
Kim Dobson Sydnor, PhD (Morgan State University)
Herman A. Taylor, Jr., MD (University of Mississippi Medical Center)
David C. Tong, MD (Stanford University)
Stan Tuhrim, MD (Mount Sinai School of Medicine)
Linda S. Williams, MD (Indiana University School of Medicine)
Philip A. Wolf, MD (Boston University)
Lawrence K.S. Wong, MD (The Chinese University of Hong Kong)
Bradford Burke Worrall, MD, MSc (University of Virginia)
Audrey Penn, MD, Acting Director, NINDS
Constance Atwell, PhD, Director of Extramural Research, NINDS
Barbara Radziszewska, PhD, MPH, NINDS
Ronnie Horner, PhD, NINDS
Alfred Gordon, PhD, NINDS
Alison Baird, MD, NINDS
Jeffrey Cutler, MD, NHLBI
Paula Einhorn, MD, NHLBI
Steven Warach, MD, NINDS