Excess Cervical Cancer Mortality A Marker for Low Access to Health Care in Poor Communities An Analysis Center to Reduce Cancer Health Disparities U. S. DEPARTMENT OF HEALTH AND HUMAN SERVICES National Institutes of Health National Cancer Institute Excess Cervical Cancer Mortality: A Marker for Low Access to Health Care in Poor Communities NCI Center to Reduce Cancer Health Disparities Harold P. Freeman, M.D., Director This publication is available on the Center to Reduce Cancer Health Disparities Web site: http://crchd.nci.nih.gov Suggested citation for the report: Freeman HP, Wingrove BK. Excess Cervical Cancer Mortality: A Marker for Low Access to Health Care in Poor Communities. Rockville, MD: National Cancer Institute, Center to Reduce Cancer Health Disparities, May 2005. NIH Pub. No. 05–5282. For additional copies, please contact: Center to Reduce Cancer Health Disparities National Cancer Institute 6116 Executive Boulevard, Suite 602 Bethesda, MD 20892 Telephone: 301–496–8589 Fax: 301–435–9225 Copyright information: All material in this report is in the public domain and may be reproduced or copied without permission; citation as to source, however, is appreciated. Acknowledgements The analysis presented in this report represents a synthesis of the findings of a Roundtable colloquium and two Think Tank meetings convened by the NCI Center to Reduce Cancer Health Disparities. We wish to acknowledge and thank the following people for their commitment, hard work, and assistance in the development of this report. Editor Barbara K. Wingrove, M.P.H. Chief, Health Policy Branch Planning, Development and Implementation of Meetings Patricia Newman, M.G.A. Manager, Communications Planning and Logistical Support NOVA Research Company (NCI contract number N02–CO–14231) Center Staff Involved in Program Development Nada Vydelingum, Ph.D. Deputy Director Barbara K. Wingrove, M.P.H. Chief, Health Policy Branch Jane MacDonald Daye, M.H.S. Special Assistant Susanne H. Reuben Progressive Health Systems Participants: Roundtable • Reducing Health Disparities in High Cervical Cancer Mortality Regions—Phase 1 November 28–30, 2001, Corpus Christi, TX: Appendix A Think Tanks • Regions With High Cervical Cancer Mortality—Phase 2 May 8, 2002, Bethesda, MD: Appendix B • Cervical Cancer Mortality—A Marker for the Health of Poor and Underserved Women: Toward an Interagency Collaboration to Reduce Disparities October 28–29, 2002, Bethesda, MD: Appendix C Special Acknowledgements Jon Kerner, Ph.D., for leading the organization of the Roundtable event and the concept mapping exercise. Susan Devesa, Ph.D., for the preparation of the national maps showing regions with high cervical cancer mortality. Suzanne H. Reuben, Progressive Health Systems, for significant help in formulating the Think Tanks, and for writing, revisions, and consultation on the manuscript. Director’s Message A recent report identifying priority areas of health requiring national action,1 including coordination of care, cancer screening, and self-management/health literacy, noted the stark fact that while “the United States spends more than $1 trillion on health care annually [and has] extraordinary knowledge and capacity to deliver the best care in the world…we repeatedly fail to translate that knowledge and capacity into clinical practice.” Nowhere is this failure of our health care system more apparent than in the disparities in cancer incidence and outcome, as well as in other health issues, suffered by members of particular racial and ethnic minority subgroups and other underserved populations. These disparities are grim realities resulting from the longstanding disconnect between (1) our extraordinary biomedical research discoveries and our ability to turn them into interventions that improve health and (2) our most distressing inability to deliver those interventions to all of the people who need them. It is in this context that the National Cancer Institute’s Center to Reduce Cancer Health Disparities (CRCHD) approached the problem of mortality from cervical cancer, a disease for which effective prevention—not just early detection—and treatment have existed for decades. Our failure to provide this lifesaving care to all women through appropriate infrastructure, information/communication systems, and adequate health care access highlights the urgent need to analyze our health care system—particularly publicly funded health services—and courageously craft the changes that will eliminate disparities and save lives. Harold P. Freeman, M.D. Director National Cancer Institute Center to Reduce Cancer Health Disparities Executive Summary Without question, cervical cancer is a success story in the history of cancer control. Since screening programs using the Papanicolaou test (Pap test) were implemented widely more than 50 years ago, cervical cancer deaths have declined 75 percent nationwide. Yet cervical cancer still takes the lives of approximately 4,000 women in the United States each year. This is particularly disturbing since virtually all cervical cancers should be avoidable with proper screening, and because effective treatment is available for precancerous lesions and for invasive cancers that are detected before they have spread. The National Cancer Institute (NCI) Center to Reduce Cancer Health Disparities (CRCHD) postulates that cervical cancer is an indicator of larger health system concerns such as: infrastructure, access, culturally competent communication, and patient/provider education deficits that disproportionately affect members of particular racial and ethnic minority subgroups and other underserved women who also are subject to the negative effects of poverty on health status. Following a review of the scientific literature and available data on persistent cervical cancer mortality, CRCHD convened more than 180 Federal, state, and local planning and program personnel, policy-makers, researchers, clinicians, advocates, educators, and communications specialists as participants in its Cervical Cancer Mortality Project (CCMP) to explore the components of the problem, identify critical needs, and suggest actions to meet those needs. An entrenched pattern of high cervical cancer mortality has existed for decades in distinct populations and geographic areas. Women suffering most severely from this disparity include African American women in the South, Latina women along the Texas-Mexico border, white women in Appalachia, American Indians of the Northern Plains, Vietnamese American women, and Alaska Natives. A more detailed analysis of two geographic regions where cervical cancer mortality is the greatest indicates that, in addition to needing targeted interventions and additional resources to reduce cervical cancer deaths, these communities also experience high mortality rates for other conditions and diseases for which screening and treatment are currently available. A recent Institute of Medicine report2 urges the Federal Government, using certain types of Federal health facilities as laboratories of innovation, to provide leadership in health care quality improvement efforts. In 2003, the Department of Health and Human Services (DHHS) chose to use the Progress Review Group (PRG) methodology to facilitate, promote, and coordinate partnerships among Federal agencies to address persistent cancer disparities, such as the excess cervical cancer mortality rates in geographic regions and populations, implement new initiatives, and evaluate progress over time. The NCI Center to Reduce Cancer Health Disparities recommends specific actions to eliminate cervical cancer mortality disparities suffered by women in identified geographic regions of the nation and to improve health care for all underserved women. Each major objective is listed with specific recommendations for reaching the goal. The recommendations are summarized on Table 1. In this report, the NCI Center to Reduce Cancer Health Disparities demonstrates that high rates of cervical cancer are an indicator of broader problems in access to health care. The report argues that a high rate of cervical cancer is a sentinel marker indicating larger, systemic health care issues that need to be addressed by cancer control and other strategies. It also illustrates how the recommendations of the Report of the Trans-HHS Cancer Health Disparities Progress Review Group (CHPRG), Making Cancer Health Disparities History (http://www.chdprg.omhrc.gov), can be implemented to improve women’s health in geographic areas experiencing excess cervical cancer mortality. The correspondence between our recommendations and those of the CHPRG recommendations are shown in Table 2. Table 1. Strategies for Reducing Excess Cervical Cancer Mortality Collaborations, Partnerships, and Advocacy Establish and strengthen partnerships to promote “whole woman” approach to care. Develop and implement an agenda to provide and sustain funding for coalitions, partnerships, and community-based quali- ty health services, education, and prevention programs. Research • Optimize HPV testing and HPV vaccine development to eliminate the primary biologic cause of cervical cancer. • Improve screening technologies and screening interventions to bring affordable screening to all women. • Conduct social/behavioral, health services, and intervention research to better understand high-risk populations and develop interventions to improve their care. • Improve data collection and surveillance activities related both to quantitative and to qualitative understanding of cervical cancer causes and control. Access—Outreach, Services, Navigation • Intensify outreach to women who have rarely or never been screened for cervical, breast, or colon cancer and other screenable/treatable diseases. • Enable women who rely on publicly funded health services to have a “medical home”—a usual source of health care. • Provide patient navigators to help eliminate the disconnects between screening and follow-up treatment. • Increase the number of female providers of the patients’ gender/race/ethnicity. • Improve coverage and reimbursement for cancer-related services. • Improve the quality of care in rural areas through telemedicine and multidisciplinary consultations. Communications and Information • Improve awareness and knowledge about cervical cancer through the development and provision of linguistically and culturally appropriate information. • Improve provider-patient communication through provider education and availability of language translation. • Provide central resource detailing best practices for cervical and other cancers including evidence-based interventions. • Improve medical records maintenance and retrieval systems through the use of rapidly evolving information technology. Table 2. Association of Eliminating Excess Cervical Cancer Mortality Recommendations With Priority Recommendations of the Trans-HHS Cancer Health Disparities PRG Recommendations for Eliminating Excess Cervical Cancer Mortality • Intensify outreach to women who have rarely or never been screened for cancer and other screenable/treatable diseases. • Provide patient navigators to help eliminate the disconnects between screening and follow-up treatment. • Improve screening technologies and screening interventions to bring affordable screening to all women. • Increase the number of female providers of patients’ gender/race/ethnicity. • Increase coverage and reimbursement for cancer-related services. • Improve the quality of care in rural areas through telemedicine and multidisciplinary consultations. • Improve awareness and knowledge through the development and provision of linguistically and culturally appropriate information. • Improve provider-patient communication through provider education and availability of language translation. • Provide a central resource detailing “best practices” evidence- based interventions. • Improve medical records maintenance and retrieval systems through the use of rapidly evolving information technology. • Improve data collection and surveillance activities related both to quantitative and to qualitative understanding of cervical cancer. • Enable women to have a “medical home”—a usual source of health care. • Establish and strengthen partnerships that promote a “whole woman” approach to care. • Develop and implement agenda to provide and sustain funding for coalitions, partnerships, and community-based quality health services, education, and prevention programs. • Optimize HPV testing and HPV vaccine development to eliminate the primary biologic cause of cervical cancer. • Conduct social/behavioral, health services, and intervention research to better understand high-risk populations and develop interventions to improve their care. Priority Recommendations of the Trans-HHS Cancer Health Disparities PRG • Ensure that populations at highest risk have access to age- and gender-appropriate screening and follow-up services. • Develop, implement, and evaluate education and training programs designed to create a diverse and culturally competent cancer care workforce. • Ensure that every cancer patient has access to “state-of-the- science” care. • Support culturally, linguistically, and literacy-specific approaches for eliminating cancer health disparities. These should include evidence-based “best practices,” proven interventions, and outreach strategies. • Establish new approaches for data collection and sharing to aid in the study of the effects of cancer and its relationship to variables such as race and socioeconomic status. • Collaborate with the private and voluntary health sectors to ensure that all Americans receive the full range of lifesaving information, services, and quality care. • Increase the proportion of HHS agency support targeted specifically to disease prevention, health promotion, evaluation, and translational research on cancer health disparities. • Establish partnerships for and support the development of sustainable community-based networks for participatory research in areas of high cancer disparities. Table of Contents Why Cervical Cancer Mortality Is Important . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 CRCHD’s Cervical Cancer Mortality Project (CCMP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Current Knowledge About Cervical Cancer Mortality—Overview . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Sources of Data on Cervical Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Cervical Cancer Mortality in America . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Identified Needs and Strategies to Reduce Cervical Cancer Mortality and Improve Women’s Health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Access—Outreach, Services, and Navigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Information and Communication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Collaborations, Partnerships, and Advocacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Research Needs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Cervical Cancer Mortality Is an Avoidable Cause of Death and a Marker for Conditions That Contribute to Health Disparities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Addressing Cervical Cancer Mortality Offers an Important Opportunity to Address the Nation’s Growing Concern About Persistent Health Disparities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Vulnerable Populations Must Be Provided With Necessary Preventive, Acute Care, and Disease Management Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Innovation, Commitment, and Creativity Are Crucial to Finding Ways to Use Available Resources More Efficiently and Effectively . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Leadership and Partnership Are Needed to Create Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Access—Outreach, Services, Navigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Information and Communication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Collaborations, Partnerships, and Advocacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Appendices Appendix A: Roundtable Agenda, Preliminary Think Tank Agenda, Participant Roster, and Executive Summary. Reducing Health Disparities in High Cervical Cancer Mortality Regions—Phase 1, November 28–30, 2001, Corpus Christi, TX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Appendix B: Think Tank Agenda and Participant Roster. Regions With High Cervical Cancer Mortality—Phase 2, May 8, 2002, Bethesda, MD . . . . . . . . . . . . . . . . . . . . . . . . 69 Appendix C: Think Tank Agenda and Participant Roster. Cervical Cancer Mortality— A Marker for the Health of Poor and Underserved Women: Toward an Interagency Collaboration to Reduce Disparities, October 28–29, 2002, Bethesda, MD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 1 Excess Cervical Cancer Mortality: A Marker for Low Access to Health Care in Poor Communities An Analysis from the Center to Reduce Cancer Health Disparities (CRCHD) Effectively addressing cervical cancer mortality can provide a model for action—an opportunity to address not only the health problems facing women who are dying from this disease but also the full set of human circumstances that lead to health disparities. Women suffering high cervical cancer mortality also: • Tend not to have a usual source of health care • Are less likely to receive preventive health services, including cancer screening • Have low incomes and educational attainment • Have high rates of breast cancer, colorectal cancer, cerebrovascular disease,and infant mortality CRCHD is the cornerstone and organizational locus of the National Cancer Institute’s efforts to reduce the unequal burden of cancer in society. CRCHD oversees and coordinates the NCI’s strategic plan to reduce cancer health disparities. Activities of the Center include health policy analyses and disparities research focusing on relationships among social, economic, cultural and environmental factors that cause or contribute to (1) the disproportionate cancer burden experienced by some populations and (2) the significant disconnect between research discoveries and the delivery of their benefits to all. Cervical cancer is unquestionably a success story in the history of cancer control. Since cervical cancer screening programs using the Papanicolaou test (Pap test/Pap smear) were introduced more than 50 years ago, age-adjusted mortality from cervical cancer overall has declined three-fold. Because of the Pap test, which is inexpensive, easily administered, and effective, and because proven treatment for precancerous cervical lesions and localized invasive cancers is available,3 virtually all cervical cancer deaths should be avoidable. Pap tests find precancerous lesions that are easily and effectively treated with colposcopy or simply watched, since not all precancerous lesions become cancer. When cervical cancer is detected before it has spread, it is one of the most successfully treated cancers,4 though patients may suffer adverse consequences from treatment, including infertility and late effects of radiation and/or chemotherapy. Recent research discoveries, including liquid-based cytology, a combined Pap test and test for the human papillomavirus (HPV) that causes most cervical cancers, and the development and testing of HPV vaccines promise to improve even further our ability to prevent or identify abnormalities of the uterine cervix long before they become cancerous. Still, in 2004 an estimated 3,900 women died from cervical cancer.5 Why Cervical Cancer Mortality Is Important Despite the consistent decline in cervical cancer mortality overall, an entrenched geographic pattern of deaths from this disease has persisted for decades. This ongoing disparity in mortality from a wholly preventable disease drew the interest of the National Cancer Institute (NCI) and led the NCI Center to Reduce Cancer Health Disparities (CRCHD/the Center) to lead the inquiry into underlying factors that may contribute to the disparity. The NCI explored the hypothesis that in addition to being a cause of concern, endemic elevated cervical cancer mortality may be a marker or an indicator of weaknesses in the health care system infrastructure, particularly with respect to medical care access, cultural issues, and health communication and education issues that disproportionately affect poor and other underserved women. Most women living in areas with high rates of cervical cancer mortality rely on publicly funded programs for their health care. A recent analysis of selected SEER areas confirms that late-stage cervical cancer diagnoses are more likely in areas that are economically or socially distressed. The authors recommend that all distressed areas should automatically receive public funding.6 Poverty, in turn, is a human condition marked by substandard housing, lower educational attainment, subsistence-level employment, high unemployment, greater exposure to environmental toxins, and reduced access to health care. These conditions, occurring in a variety of urban and rural settings, also may affect health status significantly. The Center further postulated that addressing issues related to cervical cancer in areas with high mortality from the disease also should result in improved overall health status and reduced mortality in these geographic regions. Women living in areas characterized by excess cervical cancer mortality also experience mortality rates above the national average for breast cancer, colon cancer, heart disease, stroke, and other conditions whose outcomes improve with regular screening or early intervention. Applying appropriate system improvements throughout publicly funded health services could have a broad-reaching effect on women’s health nationwide. CRCHD’S Cervical Cancer Mortality Project (CCMP) This project was conducted in two major phases. The first phase focused on collecting and analyzing both historical and current data on cervical cancer incidence, screening, treatment, and mortality in the United States. In addition, a review of the literature published between 1966 and May 2001 (and some studies published between 1950 and 1965) on cervical cancer mortality among rural women was commissioned and is detailed in a report.7 From November 28–30, 2001, a Roundtable meeting was held in Corpus Christi, TX. The 144 participants (see Appendix A) included Federal, state, and local planning and program personnel; researchers from several disciplines with an interest in cervical cancer; clinicians; 2 advocates; educators; communications specialists; and NCI planning, cancer control, and CRCHD personnel. Findings from the data and literature reviews were shared with the participants. Prior to the meeting, attendees were asked to participate in an online “concept mapping” exercise in which they submitted ideas for actions that would reduce cervical cancer mortality in their state or region. Some of the suggestions also were for actions at the national level. All of the ideas submitted were collated, redundancies were eliminated, and the ideas were sorted into conceptual categories. The condensed data were displayed on a “concept map” that identified clusters of related ideas. This map was presented to Roundtable attendees, who were asked to use the possible focus areas suggested by the clusters as the nucleus for one-year state action plans. On November 30, 2001, core members of the Cervical Cancer Think Tank met at the conclusion of the Roundtable meeting to review the suggested activities, actions, interventions, and policy changes in each of the four identified focus areas. The Think Tank members discussed possible priorities among the suggested actions, primarily at national and regional levels. All of the Phase 1 activities and outcomes were described in an interim report8 distributed to Roundtable participants, NCI staff, and others. The presentations made at that Roundtable meeting can be viewed at: http://www.dccps.cancer.gov/d4d/info. html#conferences. An Executive Summary of the Roundtable meeting is contained in Appendix A. Following the completion of Phase 1 activities, it was decided that further exploration was needed to gain a better understanding of demographic, cultural, and environmental characteristics of specific populations in geographic areas experiencing high cervical cancer mortality in order to assess the potential impact of these factors on the burden from this disease. A Think Tank meeting was held in Bethesda, MD, in May 2002 (see Appendix B) bringing together several members of the group that convened in November 2001; added to this diverse group were other participants with specific experience and expertise in cervical cancer and with the populations most at risk. The group discussed factors specific to Appalachian and other rural whites; rural African Americans, particularly those in the Deep South; Latinas living near the Texas- Mexico border; and Vietnamese American and other Asian women, particularly those in California. Though little data was available and less is known about the causes of cervical cancer mortality disparities among Native Americans in the Northern Plains and among Alaska Natives, it was also acknowledged that these populations have higher than average cervical cancer death rates. Data from the 2001 California Health Interview Survey (CHIS) have been released and confirm lower rates of cervical cancer screening among Asian women, including Vietnamese (http://www.healthpolicy.ucla.edu/pubs/ publication.asp?pubID=85).9 In October 2002, CRCHD staff and selected participants in the prior Think Tank and Roundtable meetings presented key project findings and issues to a group of senior officials 3 4 of Federal agencies that either administer or finance health services for underserved women (see agenda and participant roster, Appendix C). To illustrate the issues of geographic disparities in cervical cancer mortality, data specific to the Deep South and Appalachian regions were presented, but the discussion included all known affected populations. The focus of the meeting was to identify ways in which these agencies could better work together through collaborations and partnerships to provide improved and more consistent care to women needing (1) cervical cancer screening, diagnosis, and treatment services, and (2) care for other health conditions to which they are particularly vulnerable. The discussion also emphasized identifying interventions, health services, and policies that could be implemented using existing resources or, at most, minimal additional funding. Current Knowledge About Cervical Cancer Mortality: Overview Available statistics show clearly that while all women in America have benefited from the overall cervical cancer mortality reductions achieved over the past few decades, much remains unclear regarding the reasons for continuing mortality disparities and the best ways to address them. Data on the disparities and factors contributing to them are accumulating; however, much important information has yet to be collected. Sources of Data on Cervical Cancer Data on cervical cancer incidence and mortality are available from a number of national sources, including NCI’s Surveillance, Epidemiology, and End Results (SEER) cancer registry program, the National Program of Cancer Registries (NPCR) of the Centers for Disease Control and Prevention (CDC), NCI’s Atlas of Cancer Mortality in the United States, and the Medicare database. National studies such as the National Health Interview Survey (NHIS) conducted by the National Center for Health Statistics (NCHS), CDC’s Behavioral Risk Factor Surveillance System (BRFSS), the American College of Surgeons (ACoS) National Cancer Data Base, and American Hospital Association (AHA) surveys provide additional information on risk factors, screening, and treatment. In addition, the medical literature documents studies that explore biologic, socioeconomic, cultural, environmental, and other factors that may affect cervical cancer incidence and mortality. Some of these studies address selected regional, state, or local populations. Other studies on topics relevant but not specific to cervical cancer may be found in the social and behavioral sciences literature. Several meta-analyses have examined specific aspects of cervical cancer prevention and care. However, as Yabroff et al.10 note, the published studies have such varied study designs 5 (e.g., sample selection and characteristics, response rates, method of ascertainment, definition of rural population) that data cannot be compared with confidence. In addition, few studies of screening distinguished between screening and diagnostic Pap tests or were corrected for hysterectomy or tubal ligation. As a result, the independent effects of age, social class, race, education, and geography on screening and mortality cannot be disaggregated, an endeavor further complicated by the racial, ethnic, age, and cultural heterogeneity of the affected populations. Cervical Cancer Mortality in America An overview of the available literature and data provide the following picture of cervical cancer in America: Geographic Disparities Women living in largely rural and suburban counties in states stretching from northern New England through Appalachia, in the Deep South, along the Texas-Mexico border, and in parts of the central valley of California have consistently higher rates of cervical cancer mortality than do women in other parts of the country (Map 1). Mortality rates have remained substantially higher in these areas over the past few decades, but rates have fallen somewhat even in these high mortality areas as rates have declined nationwide. Sufficient information exists to pinpoint, by county, areas of high cervical cancer mortality for white and African American women. For African American women, high mortality areas tend to be rural. The geographic mortality pattern is similar among white women, but with a somewhat more suburban pattern. The data suggest an urban-rural gradient (i.e., mortality lower in urban areas, higher in rural areas) reflecting less access to care and poorer outcomes, though this has not been demonstrated clearly.11 Racial/Ethnic Mortality Disparities African American women suffer more than twice the number of cervical cancer deaths per 100,000 population compared with white women (Maps 2 and 3, pooled white and black data, 1970–1998). In fact, National Center for Health Statistics data now available for 1996–2001, adjusted to the 2000 U.S. standard million population, indicate that while the cervical cancer mortality rate for African American women has declined considerably to 5.7 per 100,000 population, it remains the highest rate of the major (census-defined) racial/ethnic populations and almost twice the rate of white women (3.4 per 100,000).12 Rates are particularly high among black women in the rural South, but not among black women in the West.13 Cervical cancer mortality is higher than average among Hispanic/Latina women living on the Texas-Mexico border, and among white women in Appalachia, rural New York State, and northern New England. It also is known that American Indians of the Northern Plains and Alaska Native women have high cervical cancer mortality rates, but due to the small size of these populations and small number of cases relative to the total population, these differences are less well documented, and maps depicting mortality patterns currently are unavailable. Cervical cancer incidence rates are five times higher among Vietnamese American women than white women.14 Data are beginning to emerge suggesting that overall rates of cervical cancer mortality among native-born women are declining, while rates among foreign-born women are increasing, particularly in the South. The reasons for the persistent disparities in cervical cancer mortality experienced by these populations have yet to be elucidated fully. Socioeconomic Status (SES)-Related Mortality Disparities As with other cancers, the risk of dying from cervical cancer increases with later stage at diag- nosis. Available data, though limited, indicate that higher mortality is associated with lower income, less education, and lower SES overall. Among women diagnosed with stage I cervical cancer, only about 20 percent are those having lower educational attainment. Similarly, cervical cancer incidence rises with increasing poverty and decreasing SES across all racial ethnic 6 Map 1. Cancer Mortality Rates by County (Age-adjusted 1970 U.S. Population) Cervix Uteri: All Races, Females, 1970–1998 U.S. = 3.56/100,000 6.32–14.66 (highest 10%) 5.30–6.31 4.79–5.29 4.31–4.78 3.95–4.30 3.64–3.94 3.27–3.63 2.95–3.26 2.52–2.94 1.18–2.51 (lowest 10%) Sparse data (<12 observed deaths; 1,264 counties; 4.85% of deaths) Ç }ÀœÕ«Ã°£x >Ì>ʜ˜ÊëiVˆvˆVÊVœ“«œ˜i˜ÌÃʜvÊ- -Ê>Ài Liˆ˜}ÊVœiVÌi`Ê̜Ê>Ê}Ài>ÌiÀÊ`i}ÀiiÊ̅>˜ÊiÛiÀ LivœÀi]ÊLÕÌʈ˜Ê“œÃÌÊ>Û>ˆ>LiÊV>˜ViÀÊÀi}ˆÃÌÀÞÊ>˜` œÌ…iÀÊVˆ˜ˆV‡L>Ãi`Ê`>Ì>ÊÃiÌÃ]ÊÀ>ViÉi̅˜ˆVˆÌÞÊÃ̈ÊˆÃ ˆŽiÞÊ>Ê«ÀœÝÞʓi>ÃÕÀiʜvÊ- -°Ê-ÕÀÛiÞÃÊŜÜÊ̅>Ì œÜÊ- -ʈÃÊ>ÃÜVˆ>Ìi`Ê܈̅ʏ>VŽÊœvʅi>Ì…ʈ˜ÃÕÀ‡ >˜ViÊ>˜`ʜ̅iÀʅi>Ì…ÊV>ÀiÊ>VViÃÃÊv>V̜Àð /…iÊÃ>“iÊ«œ«Õ>̈œ˜ÃʜvÊܜ“i˜Ê̅>ÌÊ>Ài iÝ«iÀˆi˜Vˆ˜}ʅˆ}…ʏiÛiÃʜvÊViÀۈV>ÊV>˜ViÀ “œÀÌ>ˆÌÞÊ>ÃœÊ>ÀiÊ>ÌÊÀˆÃŽÊvœÀʜ̅iÀÊÃVÀii˜>Li >˜`ÊÌÀi>Ì>LiÊVœ˜`ˆÌˆœ˜Ã°Ê>«ÃÊ{]Êx]ÊÈ]Ê>˜`ÊÇ ˆÕÃÌÀ>ÌiÊ̅ˆÃÊÈÌÕ>̈œ˜ÊVi>ÀÞÊ>ÃʈÌÊ>««ˆiÃÊ̜
vÀˆV>˜Ê
“iÀˆV>˜Êܜ“i˜Êˆ˜Ê̅iÊii«Ê-œÕ̅ ­
>L>“>Ê>˜`ʈÃÈÃÈ««ˆ®]Ê>˜`Ê>«ÃÊn]ʙ]Ê£ä] >˜`Ê££ÊˆÕÃÌÀ>ÌiÊVœÕ˜ÌˆiÃʜvÊiÝViÃÃʓœÀÌ>ˆÌÞÊvœÀ ܅ˆÌiÊܜ“i˜Êˆ˜Ê«>ÀÌÃʜvÊ
««>>V…ˆ>Ê­7iÃÌ 6ˆÀ}ˆ˜ˆ>Ê>˜`Êi˜ÌÕVŽÞ®ÊÀiëiV̈ÛiÞ°£È ˜ÊLœÌ…Ê>Ài>Ã]Ê«œÛiÀÌÞʏiÛiÃÊ>ÀiÊv>ÀÊ>LœÛi ̅iʘ>̈œ˜>Ê>ÛiÀ>}i]Ê>˜`Ê̅iÃiÊ«œ«Õ>̈œ˜ÃÊ>Ài V…>À>VÌiÀˆâi`ÊLÞʏœÜiÀÊi`ÕV>̈œ˜>Ê>ÌÌ>ˆ˜“i˜Ì iÛiÃ]Ê}iœ}À>«…ˆVʈ܏>̈œ˜]ʏ>VŽÊœvÊÌÀ>˜Ã‡ «œÀÌ>̈œ˜]Ê}Ài>ÌiÀÊiÝ«œÃÕÀiÊ̜Êi˜ÛˆÀœ˜“i˜Ì> >˜`ʜVVÕ«>̈œ˜>Ê…>â>À`Ã]Ê>˜`Ê«œœÀʅœÕȘ}°£Ç /…iÊܜ“i˜Êˆ˜Ê̅iÃiʅˆ}…ÊViÀۈV>ÊV>˜ViÀ “œÀÌ>ˆÌÞÊ>Ài>ÃÊ>ÃœÊiÝ«iÀˆi˜ViÊiÝViÃÈÛi >«ÊÓ°Ê >˜ViÀÊœÀÌ>ˆÌÞÊ,>ÌiÃÊLÞÊ œÕ˜ÌÞÊ­
}i‡>`ÕÃÌi`Ê£™ÇäÊ1°-°Ê*œ«Õ>̈œ˜®Ê iÀۈÝÊ1ÌiÀˆ\ >VŽÊi“>iÃ]Ê£™Çäq£™™nÆÊ*œœi`Ê7…ˆÌiÊ>˜`Ê >VŽÊ,>Ìià ./4% 3HADES□OF□PURPLE□^□□COUNTIES□EACH 3HADES□OF□TAN□□□□□□□^□□COUNTIES□EACH 53□□ n n n n n n n n n n 3PARSE□DATA□□OBSERVED□DEATHS□□COUNTIES□□OF□DEATHS “œÀÌ>ˆÌÞÊvÀœ“ÊLÀi>ÃÌÊ>˜`ÊVœœÀiVÌ>ÊV>˜ViÀÃ] …i>ÀÌÊ>˜`ÊViÀiLÀœÛ>ÃVՏ>ÀÊ`ˆÃi>ÃiÃ]Ê>˜`ʅˆ}… À>ÌiÃʜvʈ˜v>˜ÌʓœÀÌ>ˆÌÞ°£n
Ì…œÕ}…ÊVœ“«>À>LiÊ`>Ì>ÊVÕÀÀi˜ÌÞÊ>Ài ˜œÌÊ>Û>ˆ>LiÊvœÀʜ̅iÀÊ«œ«Õ>̈œ˜ÃÊ܈̅ʎ˜œÜ˜ …ˆ}…ÊViÀۈV>ÊV>˜ViÀʓœÀÌ>ˆÌÞÊÀ>ÌiÃ]Ê̅iˆÀÊ`ˆÃi>Ãi «>ÌÌiÀ˜ÃÊ>ÀiÊLiˆiÛi`Ê̜ÊLiÊÃÕLÃÌ>˜Ìˆ>ÞÊȓˆ>À° iÊ/Þ«i œÃÌÊViÀۈV>ÊV>˜ViÀÃÊ>ÀiÊõÕ>“œÕÃÊVi V>ÀVˆ˜œ“>Ã°Ê œ˜ÃµÕ>“œÕÃÊViÊV>˜ViÀà ­>`i˜œV>ÀVˆ˜œ“>îʘœÜÊ>VVœÕ˜ÌÊvœÀÊ>LœÕÌÊÎä «iÀVi˜ÌʜvÊV>ÃiðÊ/…iÊ«Àœ«œÀ̈œ˜ÊœvÊ> ViÀۈV>Ê>`i˜œV>ÀVˆ˜œ“>Ãʅ>Ãʈ˜VÀi>Ãi`Ê>ÃÊ̅i À>ÌiÃÊvœÀÊõÕ>“œÕÃÊViÊV>˜ViÀÃʅ>ÛiÊ`iVˆ˜i`° ,i`ÕVˆ˜}ʓœÀÌ>ˆÌÞÊvÀœ“ÊViÀۈV>Ê>`i˜œ‡ V>ÀVˆ˜œ“>Ãʅ>ÃÊLii˜ÊiÃÃÊÃÕVViÃÃvՏÊ̅>˜ ܈̅ÊõÕ>“œÕÃÊViÊV>˜ViÀðÊÌʈÃʘœÌÊÞiÌ Ž˜œÜ˜Ê܅i̅iÀÊ̅iÀiʈÃÊ>ÊÀi>̈œ˜Ã…ˆ« LiÌÜii˜Ê…ˆ}…Ê“œÀÌ>ˆÌÞÊ}iœ}À>«…ˆVÊ>Ài>ÃÊ>˜` ̅iÊëiVˆvˆVʅˆÃ̜œ}ÞʜvÊ̅iÊViÀۈV>ÊV>˜ViÀà vœÕ˜`Ê̅iÀi° n >«ÊÎ°Ê >˜ViÀÊœÀÌ>ˆÌÞÊ,>ÌiÃÊLÞÊ œÕ˜ÌÞÊ­
}i‡>`ÕÃÌi`Ê£™ÇäÊ1°-°Ê*œ«Õ>̈œ˜®Ê iÀۈÝÊ1ÌiÀˆ\ 7…ˆÌiÊi“>iÃ]Ê£™Çäq£™™nÆÊ*œœi`Ê7…ˆÌiÊ>˜`Ê >VŽÊ,>Ìià ./4% 3HADES□OF□PURPLE□^□□COUNTIES□EACH 3HADES□OF□TAN□□□□□□□^□□COUNTIES□EACH 53□□ n n n n n n n n n n 3PARSE□DATA□□OBSERVED□DEATHS□□COUNTIES□□OF□DEATHS ™ >«Ê{°Ê iÀۈV>Ê >˜ViÀÊœÀÌ>ˆÌÞÊ,>Ìi\Ê >VŽÊi“>i]Ê£™È™q£™™™Ê­
>L>“>Ê>˜`ʈÃÈÃÈ««ˆ®Ê !UTAUGA "ALDWIN "ARBOUR "IBB "LOUNT "ULLOCK "UTLER #ALHOUN #HAMBERS #HEROKEE #HILTON #HOCTAW #LARKE #LAY #LEBURNE #OFFEE #OLBERT #ONECUH #OOSA #OVINGTON #RENSHAW #ULLMAN $ALE $ALLAS $E□+ALB %LMORE %SCAMBIA %TOWAH &AYETTE &RANKLIN 'ENEVA 'REENE (ALE (ENRY (OUSTON *ACKSON *EFFERSON ,AMAR ,AUDERDALE ,AWRENCE ,EE ,IMESTONE ,OWNDES -ACON -ADISON -ARENGO -ARION -ARSHALL -OBILE -ONROE -ONTGOMERY -ORGAN 0ERRY 0ICKENS 0IKE 2ANDOLPH 2USSELL 3T□#LAIR 3HELBY 3UMTER 4ALLADEGA 4ALLAPOOSA 4USCALOOSA 7ALKER 7ASHINGTON 7ILCOX 7INSTON !DAMS !LCORN !MITE !TTALA "ENTON "OLIVAR #ALHOUN #ARROLL #HICKASAW #HOCTAW #LAIBORNE #LARKE #LAY #OAHOMA #OPIAH #OVINGTON $E□3OTO &ORREST &RANKLIN 'EORGE 'REENE 'RENADA (ANCOCK (ARRISON (INDS (OLMES (UMPHREYS )SSAQUENA )TAWAMBA *ACKSON *ASPER *EFFERSON $AVIS *ONES +EMPER ,AFAYETTE ,AMAR ,AUDERDALE ,AWRENCE ,EAKE ,EE ,EFLORE ,INCOLN ,OWNDES -ADISON -ARION -ARSHALL -ONROE -ONTGOMERY .ESHOBA .EWTON .OXUBEE /KTIBBEHA 0ANOLA 0EARL□2IVER 0ERRY 0IKE 0ONTOTOC 0RENTISS 1UITMAN 2ANKIN 3COTT 3HARKEY 3IMPSON 3MITH 3TONE 3UNFLOWER 4ALLAHATCHIE 4ATE 4IPPAH 4ISHOMINGO 4UNICA 5NION 7ALTHALL 7ARREN 7ASHINGTON 7AYNE 7EBSTER 7ILKINSON 7INSTON 9ALOBUSHA 9AZOO *EFFERSON □TO□ □TO□ □TO□e □TO□ □TO□ □TO□ □TO□ p 3UPPRESSED  □ □!LL□RACES□NATIONAL□AVERAGE□ □PER□ □e□"LACK□NATIONAL□AVERAGE□ □PER□ □p□7HITE□NATIONAL□AVERAGE□ □PER□  □□□$ATA□HAVE□BEEN□SUPPRESSED□TO□ENSURE□ □□□CONFIDENTIALITY□AND□STABILITY□OF□RATE□ESTIMATES >«ÃÊ«Ài«>Ài`ÊvÀœ“Ê`>Ì>Ê«ÀœÛˆ`i`ÊLÞ\ -ÕÀÛiˆ>˜Vi]Ê «ˆ`i“ˆœœ}Þ]Ê>˜`Ê ˜`Ê,iÃՏÌÃÊ­- ,®Ê*Àœ}À>“Ê­ÜÜÜ°ÃiiÀ°V>˜ViÀ°}œÛ®Ê- ,I-Ì>Ì >Ì>L>Ãi\ œÀÌ>ˆÌÞq
Ê "]Ê*ÕLˆV‡1ÃiÊ7ˆÌ…Ê œÕ˜ÌÞ]Ê/œÌ>Ê1°-°Ê­£™È™q£™™™®]Ê >̈œ˜>Ê >˜ViÀʘÃ̈ÌÕÌi]Ê *-]Ê-ÕÀÛiˆ>˜ViÊ,iÃi>ÀV… *Àœ}À>“°Ê >˜ViÀÊ-Ì>̈Ã̈VÃÊ À>˜V…°Ê1˜`iÀÞˆ˜}ʓœÀÌ>ˆÌÞÊ`>Ì>° >«Êx°Ê œœ˜Ê >˜ViÀÊœÀÌ>ˆÌÞÊ,>Ìi\Ê >VŽÊi“>i]Ê£™È™q£™™™Ê­
>L>“>Ê>˜`ʈÃÈÃÈ««ˆ® !UTAUGA "ALDWIN "ARBOUR "IBB "LOUNT "ULLOCK "UTLER #ALHOUN #HAMBERS #HEROKEE #HILTON #HOCTAW #LARKE #LAY #LEBURNE #OFFEE #OLBERT #ONECUH #OOSA #OVINGTON #RENSHAW #ULLMAN $ALE $ALLAS $E□+ALB %LMORE %SCAMBIA %TOWAH &AYETTE &RANKLIN 'ENEVA 'REENE (ALE (ENRY (OUSTON *ACKSON *EFFERSON ,AMAR ,AUDERDALE ,AWRENCE ,EE ,IMESTONE ,OWNDES -ACON -ADISON -ARENGO -ARION -ARSHALL -OBILE -ONROE -ONTGOMERY -ORGAN 0ERRY 0ICKENS 0IKE 2ANDOLPH 2USSELL 3T□#LAIR 3HELBY 3UMTER 4ALLADEGA 4ALLAPOOSA 4USCALOOSA 7ALKER 7ASHINGTON 7ILCOX 7INSTON !DAMS !LCORN !MITE !TTALA "ENTON "OLIVAR #ALHOUN #ARROLL #HICKASAW #HOCTAW #LAIBORNE #LARKE #LAY #OAHOMA #OPIAH #OVINGTON $E□3OTO &ORREST &RANKLIN 'EORGE 'REENE 'RENADA (ANCOCK (ARRISON (INDS (OLMES (UMPHREYS )SSAQUENA )TAWAMBA *ACKSON *ASPER *EFFERSON $AVIS *ONES +EMPER ,AFAYETTE ,AMAR ,AUDERDALE ,AWRENCE ,EAKE ,EE ,EFLORE ,INCOLN ,OWNDES -ADISON -ARION -ARSHALL -ONROE -ONTGOMERY .ESHOBA .EWTON .OXUBEE /KTIBBEHA 0ANOLA 0EARL□2IVER 0ERRY 0IKE 0ONTOTOC 0RENTISS 1UITMAN 2ANKIN 3COTT 3HARKEY 3IMPSON 3MITH 3TONE 3UNFLOWER 4ALLAHATCHIE 4ATE 4IPPAH 4ISHOMINGO 4UNICA 5NION 7ALTHALL 7ARREN 7ASHINGTON 7AYNE 7EBSTER 7ILKINSON 7INSTON 9ALOBUSHA 9AZOO *EFFERSON □TO□□ □TO□□ □TO□ □TO□e □TO□ □TO□ p □TO□ □TO□ 3UPPRESSED  □ □!LL□RACES□NATIONAL□AVERAGE□ □PER□ □e□"LACK□NATIONAL□AVERAGE□ □PER□ □p□7HITE□NATIONAL□AVERAGE□ □PER□  □□□$ATA□HAVE□BEEN□SUPPRESSED□TO□ENSURE□ □□□CONFIDENTIALITY□AND□STABILITY□OF□RATE□ESTIMATES £ä >«ÊÈ°Ê Ài>ÃÌÊ >˜ViÀÊœÀÌ>ˆÌÞÊ,>Ìi\Ê >VŽÊi“>i]Ê£™È™q£™™™Ê­
>L>“>Ê>˜`ʈÃÈÃÈ««ˆ®Ê !UTAUGA "ALDWIN "ARBOUR "IBB "LOUNT "ULLOCK "UTLER #ALHOUN #HAMBERS #HEROKEE #HILTON #HOCTAW #LARKE #LAY #LEBURNE #OFFEE #OLBERT #ONECUH #OOSA #OVINGTON #RENSHAW #ULLMAN $ALE $ALLAS $E□+ALB %LMORE %SCAMBIA %TOWAH &AYETTE &RANKLIN 'ENEVA 'REENE (ALE (ENRY (OUSTON *ACKSON *EFFERSON ,AMAR ,AUDERDALE ,AWRENCE ,EE ,IMESTONE ,OWNDES -ACON -ADISON -ARENGO -ARION -ARSHALL -OBILE -ONROE -ONTGOMERY -ORGAN 0ERRY 0ICKENS 0IKE 2ANDOLPH 2USSELL 3T□#LAIR 3HELBY 3UMTER 4ALLADEGA 4ALLAPOOSA 4USCALOOSA 7ALKER 7ASHINGTON 7ILCOX 7INSTON !DAMS !LCORN !MITE !TTALA "ENTON "OLIVAR #ALHOUN #ARROLL #HICKASAW #HOCTAW #LAIBORNE #LARKE #LAY #OAHOMA #OPIAH #OVINGTON $E□3OTO &ORREST &RANKLIN 'EORGE 'REENE 'RENADA (ANCOCK (ARRISON (INDS (OLMES (UMPHREYS )SSAQUENA )TAWAMBA *ACKSON *ASPER *EFFERSON $AVIS *ONES +EMPER ,AFAYETTE ,AMAR ,AUDERDALE ,AWRENCE ,EAKE ,EE ,EFLORE ,INCOLN ,OWNDES -ADISON -ARION -ARSHALL -ONROE -ONTGOMERY .ESHOBA .EWTON .OXUBEE /KTIBBEHA 0ANOLA 0EARL□2IVER 0ERRY 0IKE 0ONTOTOC 0RENTISS 1UITMAN 2ANKIN 3COTT 3HARKEY 3IMPSON 3MITH 3TONE 3UNFLOWER 4ALLAHATCHIE 4ATE 4IPPAH 4ISHOMINGO 4UNICA 5NION 7ALTHALL 7ARREN 7ASHINGTON 7AYNE 7EBSTER 7ILKINSON 7INSTON 9ALOBUSHA 9AZOO *EFFERSON □TO□□ □TO□□ □TO□ □TO□ □TO□e □TO□ □TO□p □TO□ □ □!LL□RACES□NATIONAL□AVERAGE□ □PER□ □e□"LACK□NATIONAL□AVERAGE□ □PER□ □p□7HITE□NATIONAL□AVERAGE□ □PER□ >«ÃÊ«Ài«>Ài`ÊvÀœ“Ê`>Ì>Ê«ÀœÛˆ`i`ÊLÞ\ -ÕÀÛiˆ>˜Vi]Ê «ˆ`i“ˆœœ}Þ]Ê>˜`Ê ˜`Ê,iÃՏÌÃÊ­- ,®Ê*Àœ}À>“Ê­ÜÜÜ°ÃiiÀ°V>˜ViÀ°}œÛ®Ê- ,I-Ì>Ì >Ì>L>Ãi\ œÀÌ>ˆÌÞq
Ê "]Ê*ÕLˆV‡1ÃiÊ7ˆÌ…Ê œÕ˜ÌÞ]Ê/œÌ>Ê1°-°Ê­£™È™q£™™™®]Ê >̈œ˜>Ê >˜ViÀʘÃ̈ÌÕÌi]Ê *-]Ê-ÕÀÛiˆ>˜ViÊ,iÃi>ÀV… *Àœ}À>“°Ê >˜ViÀÊ-Ì>̈Ã̈VÃÊ À>˜V…°Ê1˜`iÀÞˆ˜}ʓœÀÌ>ˆÌÞÊ`>Ì>° >«ÊÇ°ÊˆÃi>ÃiʜvÊ̅iÊi>ÀÌÊœÀÌ>ˆÌÞÊ,>Ìi\Ê >VŽÊi“>i]Ê£™È™q£™™™Ê­
>L>“>Ê>˜`ʈÃÈÃÈ««ˆ® !UTAUGA "ALDWIN "ARBOUR "IBB "LOUNT "ULLOCK "UTLER #ALHOUN #HAMBERS #HEROKEE #HILTON #HOCTAW #LARKE #LAY #LEBURNE #OFFEE #OLBERT #ONECUH #OOSA #OVINGTON #RENSHAW #ULLMAN $ALE $ALLAS $E□+ALB %LMORE %SCAMBIA %TOWAH &AYETTE &RANKLIN 'ENEVA 'REENE (ALE (ENRY (OUSTON *ACKSON *EFFERSON ,AMAR ,AUDERDALE ,AWRENCE ,EE ,IMESTONE ,OWNDES -ACON -ADISON -ARENGO -ARION -ARSHALL -OBILE -ONROE -ONTGOMERY -ORGAN 0ERRY 0ICKENS 0IKE 2ANDOLPH 2USSELL 3T□#LAIR 3HELBY 3UMTER 4ALLADEGA 4ALLAPOOSA 4USCALOOSA 7ALKER 7ASHINGTON 7ILCOX 7INSTON !DAMS !LCORN !MITE !TTALA "ENTON "OLIVAR #ALHOUN #ARROLL #HICKASAW #HOCTAW #LAIBORNE #LARKE #LAY #OAHOMA #OPIAH #OVINGTON $E□3OTO &ORREST &RANKLIN 'EORGE 'REENE 'RENADA (ANCOCK (ARRISON (INDS (OLMES (UMPHREYS )SSAQUENA )TAWAMBA *ACKSON *ASPER *EFFERSON $AVIS *ONES +EMPER ,AFAYETTE ,AMAR ,AUDERDALE ,AWRENCE ,EAKE ,EE ,EFLORE ,INCOLN ,OWNDES -ADISON -ARION -ARSHALL -ONROE -ONTGOMERY .ESHOBA .EWTON .OXUBEE /KTIBBEHA 0ANOLA 0EARL□2IVER 0ERRY 0IKE 0ONTOTOC 0RENTISS 1UITMAN 2ANKIN 3COTT 3HARKEY 3IMPSON 3MITH 3TONE 3UNFLOWER 4ALLAHATCHIE 4ATE 4IPPAH 4ISHOMINGO 4UNICA 5NION 7ALTHALL 7ARREN 7ASHINGTON 7AYNE 7EBSTER 7ILKINSON 7INSTON 9ALOBUSHA 9AZOO *EFFERSON □TO□□ □TO□□ □TO□ □TO□e □TO□ □TO□ □TO□ □TO□ p □ □!LL□RACES□NATIONAL□AVERAGE□ □PER□ □e□"LACK□NATIONAL□AVERAGE□ □PER□ □p□7HITE□NATIONAL□AVERAGE□ □PER□ 3UPPRESSED   □□□$ATA□HAVE□BEEN□SUPPRESSED□TO□ENSURE□ □□□CONFIDENTIALITY□AND□STABILITY□OF□RATE□ESTIMATES 3UPPRESSED   □□□$ATA□HAVE□BEEN□SUPPRESSED□TO□ENSURE□ □□□CONFIDENTIALITY□AND□STABILITY□OF□RATE□ESTIMATES ££ >«Ên°Ê iÀۈV>Ê >˜ViÀÊœÀÌ>ˆÌÞÊ,>Ìi\Ê7…ˆÌiÊi“>i]Ê£™È™q£™™™Ê­i˜ÌÕVŽÞÊ>˜`Ê7iÃÌÊ6ˆÀ}ˆ˜ˆ>®Ê !DAIR !LLEN !NDERSON "ALLARD "ARREN "ATH "ELL "OONE "OURBON "OYD "OYLE "RACKEN "REATHITT "RECKINRIDGE "ULLITT "UTLER #ALDWELL #ALLOWAY #AMPBELL #ARLISLE #ARROLL #ARTER #ASEY #HRISTIAN #LARK #LAY #LINTON #RITTENDEN #UMBERLAND $AVIESS %DMONSON %LLIOTT %STILL &AYETTE &LEMING &LOYD &RANKLIN &ULTON 'ALLATIN 'ARRARD 'RANT 'RAVES 'RAYSON 'REEN 'REENUP (ANCOCK (ARDIN (ARLAN (ARRISON (ART (ENDERSON (ENRY (ICKMAN (OPKINS *ACKSON *EFFERSON *ESSAMINE *OHNSON +ENTON +NOTT +NOX ,ARUE ,AUREL ,AWRENCE ,EE ,ESLIE ,ETCHER ,EWIS ,INCOLN ,IVINGSTON ,OGAN ,YON -C#RACKEN -C#REARY -C,EAN -ADISON -AGOFFIN -ARION -ARSHALL -ARTIN -ASON -EADE -ENIFEE -ERCER -ETCALFE -ONROE -ONTGOMERY -ORGAN -UHLENBERG .ELSON .ICHOLAS /HIO /LDHAM /WEN /WSLEY 0ENDLETON 0ERRY 0IKE 0OWELL 0ULASKI 2OBERTSON 2OCKCASTLE 2OWAN 2USSELL 3COTT 3HELBY 3IMPSON 3PENCER 4AYLOR 4ODD 4RIGG 4RIMBLE 5NION 7ARREN 7ASHINGTON 7AYNE 7EBSTER 7HITLEY 7OLFE 7OODFORD "ARBOUR "ERKELEY "OONE "RAXTON "ROOKE #ABELL #ALHOUN #LAY $ODDRIDGE &AYETTE 'ILMER 'RANT 'REENBRIER (AMPSHIRE (ANCOCK (ARDY (ARRISON *EFFERSON ,EWIS ,INCOLN ,OGAN -C$OWELL -ARION -ARSHALL -ERCER -INERAL -INGO -ONONGALIA -ONROE -ORGAN .ICHOLAS /HIO 0ENDLETON 0LEASANTS 0OCAHONTAS 0RESTON 2ALEIGH 2ANDOLPH 2ITCHIE 3UMMERS 4AYLOR 4UCKER 4YLER 5PSHUR 7AYNE 7EBSTER 7ETZEL 7YOMING *ACKSON +ANAWHA -ASON 0UTNAM 2OANE 7IRT 7OOD □TO□ □TO□ □□□TO□e □□□TO□ □□□TO□ □□□TO□ □□□TO□ p □ □!LL□RACES□NATIONAL□AVERAGE□ □PER□ □e□"LACK□NATIONAL□AVERAGE□ □PER□ □p□7HITE□NATIONAL□AVERAGE□ □PER□ >«ÃÊ«Ài«>Ài`ÊvÀœ“Ê`>Ì>Ê«ÀœÛˆ`i`ÊLÞ\ -ÕÀÛiˆ>˜Vi]Ê «ˆ`i“ˆœœ}Þ]Ê>˜`Ê ˜`Ê,iÃՏÌÃÊ­- ,®Ê*Àœ}À>“Ê­ÜÜÜ°ÃiiÀ°V>˜ViÀ°}œÛ®Ê- ,I-Ì>Ì >Ì>L>Ãi\ œÀÌ>ˆÌÞq
Ê "]Ê*ÕLˆV‡1ÃiÊ7ˆÌ…Ê œÕ˜ÌÞ]Ê/œÌ>Ê1°-°Ê­£™È™q£™™™®]Ê >̈œ˜>Ê >˜ViÀʘÃ̈ÌÕÌi]Ê *-]Ê-ÕÀÛiˆ>˜ViÊ,iÃi>ÀV… *Àœ}À>“°Ê >˜ViÀÊ-Ì>̈Ã̈VÃÊ À>˜V…°Ê1˜`iÀÞˆ˜}ʓœÀÌ>ˆÌÞÊ`>Ì>° >«Ê™°Ê œœ˜Ê >˜ViÀÊœÀÌ>ˆÌÞÊ,>Ìi\Ê7…ˆÌiÊi“>i]Ê£™È™q£™™™Ê­i˜ÌÕVŽÞÊ>˜`Ê7iÃÌÊ6ˆÀ}ˆ˜ˆ>® !DAIR !LLEN !NDERSON "ALLARD "ARREN "ATH "ELL "OONE "OURBON "OYD "OYLE "RACKEN "REATHITT "RECKINRIDGE "ULLITT "UTLER #ALDWELL #ALLOWAY #AMPBELL #ARLISLE #ARROLL #ARTER #ASEY #HRISTIAN #LARK #LAY #LINTON #RITTENDEN #UMBERLAND $AVIESS %DMONSON %LLIOTT %STILL &AYETTE &LEMING &LOYD &RANKLIN &ULTON 'ALLATIN 'ARRARD 'RANT 'RAVES 'RAYSON 'REEN 'REENUP (ANCOCK (ARDIN (ARLAN (ARRISON (ART (ENDERSON (ENRY (ICKMAN (OPKINS *ACKSON *EFFERSON *ESSAMINE *OHNSON +ENTON +NOTT +NOX ,ARUE ,AUREL ,AWRENCE ,EE ,ESLIE ,ETCHER ,EWIS ,INCOLN ,IVINGSTON ,OGAN ,YON -C#RACKEN -C#REARY -C,EAN -ADISON -AGOFFIN -ARION -ARSHALL -ARTIN -ASON -EADE -ENIFEE -ERCER -ETCALFE -ONROE -ONTGOMERY -ORGAN -UHLENBERG .ELSON .ICHOLAS /HIO /LDHAM /WEN /WSLEY 0ENDLETON 0ERRY 0IKE 0OWELL 0ULASKI 2OBERTSON 2OCKCASTLE 2OWAN 2USSELL 3COTT 3HELBY 3IMPSON 3PENCER 4AYLOR 4ODD 4RIGG 4RIMBLE 5NION 7ARREN 7ASHINGTON 7AYNE 7EBSTER 7HITLEY 7OLFE 7OODFORD "ARBOUR "ERKELEY "OONE "RAXTON "ROOKE #ABELL #ALHOUN #LAY $ODDRIDGE &AYETTE 'ILMER 'RANT 'REENBRIER (AMPSHIRE (ANCOCK (ARDY (ARRISON *EFFERSON ,EWIS ,INCOLN ,OGAN -C$OWELL -ARION -ARSHALL -ERCER -INERAL -INGO -ONONGALIA -ONROE -ORGAN .ICHOLAS /HIO 0ENDLETON 0LEASANTS 0OCAHONTAS 0RESTON 2ALEIGH 2ANDOLPH 2ITCHIE 3UMMERS 4AYLOR 4UCKER 4YLER 5PSHUR 7AYNE 7EBSTER 7ETZEL 7YOMING *ACKSON +ANAWHA -ASON 0UTNAM 2OANE 7IRT 7OOD □TO□□ □TO□□ □TO□ □TO□ □TO□e □TO□ p □TO□ □TO□ □ □!LL□RACES□NATIONAL□AVERAGE□ □PER□ □e□"LACK□NATIONAL□AVERAGE□ □PER□ □p□7HITE□NATIONAL□AVERAGE□ □PER□ 3UPPRESSED   □□□$ATA□HAVE□BEEN□SUPPRESSED□TO□ENSURE□ □□□CONFIDENTIALITY□AND□STABILITY□OF□RATE□ESTIMATES 3UPPRESSED   □□□$ATA□HAVE□BEEN□SUPPRESSED□TO□ENSURE□ □□□CONFIDENTIALITY□AND□STABILITY□OF□RATE□ESTIMATES £Ó >«Ê£ä°Ê Ài>ÃÌÊ >˜ViÀÊœÀÌ>ˆÌÞÊ,>Ìi\Ê7…ˆÌiÊi“>i]Ê£™È™q£™™™Ê­i˜ÌÕVŽÞÊ>˜`Ê7iÃÌÊ6ˆÀ}ˆ˜ˆ>®Ê !DAIR !LLEN !NDERSON "ALLARD "ARREN "ATH "ELL "OONE "OURBON "OYD "OYLE "RACKEN "REATHITT "RECKINRIDGE "ULLITT "UTLER #ALDWELL #ALLOWAY #AMPBELL #ARLISLE #ARROLL #ARTER #ASEY #HRISTIAN #LARK #LAY #LINTON #RITTENDEN #UMBERLAND $AVIESS %DMONSON %LLIOTT %STILL &AYETTE &LEMING &LOYD &RANKLIN &ULTON 'ALLATIN 'ARRARD 'RANT 'RAVES 'RAYSON 'REEN 'REENUP (ANCOCK (ARDIN (ARLAN (ARRISON (ART (ENDERSON (ENRY (ICKMAN (OPKINS *ACKSON *EFFERSON *ESSAMINE *OHNSON +ENTON +NOTT +NOX ,ARUE ,AUREL ,AWRENCE ,EE ,ESLIE ,ETCHER ,EWIS ,INCOLN ,IVINGSTON ,OGAN ,YON -C#RACKEN -C#REARY -C,EAN -ADISON -AGOFFIN -ARION -ARSHALL -ARTIN -ASON -EADE -ENIFEE -ERCER -ETCALFE -ONROE -ONTGOMERY -ORGAN -UHLENBERG .ELSON .ICHOLAS /HIO /LDHAM /WEN /WSLEY 0ENDLETON 0ERRY 0IKE 0OWELL 0ULASKI 2OBERTSON 2OCKCASTLE 2OWAN 2USSELL 3COTT 3HELBY 3IMPSON 3PENCER 4AYLOR 4ODD 4RIGG 4RIMBLE 5NION 7ARREN 7ASHINGTON 7AYNE 7EBSTER 7HITLEY 7OLFE 7OODFORD "ARBOUR "ERKELEY "OONE "RAXTON "ROOKE #ABELL #ALHOUN #LAY $ODDRIDGE &AYETTE 'ILMER 'RANT 'REENBRIER (AMPSHIRE (ANCOCK (ARDY (ARRISON *EFFERSON ,EWIS ,INCOLN ,OGAN -C$OWELL -ARION -ARSHALL -ERCER -INERAL -INGO -ONONGALIA -ONROE -ORGAN .ICHOLAS /HIO 0ENDLETON 0LEASANTS 0OCAHONTAS 0RESTON 2ALEIGH 2ANDOLPH 2ITCHIE 3UMMERS 4AYLOR 4UCKER 4YLER 5PSHUR 7AYNE 7EBSTER 7ETZEL 7YOMING *ACKSON +ANAWHA -ASON 0UTNAM 2OANE 7IRT 7OOD □TO□□ □TO□□ □TO□ □TO□ □TO□e □TO□ □TO□p □TO□ □ □!LL□RACES□NATIONAL□AVERAGE□ □PER□ □e□"LACK□NATIONAL□AVERAGE□ □PER□ □p□7HITE□NATIONAL□AVERAGE□ □PER□ >«ÃÊ«Ài«>Ài`ÊvÀœ“Ê`>Ì>Ê«ÀœÛˆ`i`ÊLÞ\ -ÕÀÛiˆ>˜Vi]Ê «ˆ`i“ˆœœ}Þ]Ê>˜`Ê ˜`Ê,iÃՏÌÃÊ­- ,®Ê*Àœ}À>“Ê­ÜÜÜ°ÃiiÀ°V>˜ViÀ°}œÛ®Ê- ,I-Ì>Ì >Ì>L>Ãi\ œÀÌ>ˆÌÞq
Ê "]Ê*ÕLˆV‡1ÃiÊ7ˆÌ…Ê œÕ˜ÌÞ]Ê/œÌ>Ê1°-°Ê­£™È™q£™™™®]Ê >̈œ˜>Ê >˜ViÀʘÃ̈ÌÕÌi]Ê *-]Ê-ÕÀÛiˆ>˜ViÊ,iÃi>ÀV… *Àœ}À>“°Ê >˜ViÀÊ-Ì>̈Ã̈VÃÊ À>˜V…°Ê1˜`iÀÞˆ˜}ʓœÀÌ>ˆÌÞÊ`>Ì>° >«Ê££°ÊˆÃi>ÃiʜvÊ̅iÊi>ÀÌÊœÀÌ>ˆÌÞÊ,>Ìi\Ê7…ˆÌiÊi“>i]Ê£™È™q£™™™Ê­i˜ÌÕVŽÞÊ>˜`Ê7iÃÌÊ6ˆÀ}ˆ˜ˆ>® !DAIR !LLEN !NDERSON "ALLARD "ARREN "ATH "ELL "OONE "OURBON "OYD "OYLE "RACKEN "REATHITT "RECKINRIDGE "ULLITT "UTLER #ALDWELL #ALLOWAY #AMPBELL #ARLISLE #ARROLL #ARTER #ASEY #HRISTIAN #LARK #LAY #LINTON #RITTENDEN #UMBERLAND $AVIESS %DMONSON %LLIOTT %STILL &AYETTE &LEMING &LOYD &RANKLIN &ULTON 'ALLATIN 'ARRARD 'RANT 'RAVES 'RAYSON 'REEN 'REENUP (ANCOCK (ARDIN (ARLAN (ARRISON (ART (ENDERSON (ENRY (ICKMAN (OPKINS *ACKSON *EFFERSON *ESSAMINE *OHNSON +ENTON +NOTT +NOX ,ARUE ,AUREL ,AWRENCE ,EE ,ESLIE ,ETCHER ,EWIS ,INCOLN ,IVINGSTON ,OGAN ,YON -C#RACKEN -C#REARY -C,EAN -ADISON -AGOFFIN -ARION -ARSHALL -ARTIN -ASON -EADE -ENIFEE -ERCER -ETCALFE -ONROE -ONTGOMERY -ORGAN -UHLENBERG .ELSON .ICHOLAS /HIO /LDHAM /WEN /WSLEY 0ENDLETON 0ERRY 0IKE 0OWELL 0ULASKI 2OBERTSON 2OCKCASTLE 2OWAN 2USSELL 3COTT 3HELBY 3IMPSON 3PENCER 4AYLOR 4ODD 4RIGG 4RIMBLE 5NION 7ARREN 7ASHINGTON 7AYNE 7EBSTER 7HITLEY 7OLFE 7OODFORD "ARBOUR "ERKELEY "OONE "RAXTON "ROOKE #ABELL #ALHOUN #LAY $ODDRIDGE &AYETTE 'ILMER 'RANT 'REENBRIER (AMPSHIRE (ANCOCK (ARDY (ARRISON *EFFERSON ,EWIS ,INCOLN ,OGAN -C$OWELL -ARION -ARSHALL -ERCER -INERAL -INGO -ONONGALIA -ONROE -ORGAN .ICHOLAS /HIO 0ENDLETON 0LEASANTS 0OCAHONTAS 0RESTON 2ALEIGH 2ANDOLPH 2ITCHIE 3UMMERS 4AYLOR 4UCKER 4YLER 5PSHUR 7AYNE 7EBSTER 7ETZEL 7YOMING *ACKSON +ANAWHA -ASON 0UTNAM 2OANE 7IRT 7OOD □TO□□ □TO□□ □TO□ □TO□e □TO□ □TO□ □TO□ □TO□ p □ □!LL□RACES□NATIONAL□AVERAGE□ □PER□ □e□"LACK□NATIONAL□AVERAGE□ □PER□ □p□7HITE□NATIONAL□AVERAGE□ □PER□ 3UPPRESSED   □□□$ATA□HAVE□BEEN□SUPPRESSED□TO□ENSURE□ □□□CONFIDENTIALITY□AND□STABILITY□OF□RATE□ESTIMATES 3UPPRESSED   □□□$ATA□HAVE□BEEN□SUPPRESSED□TO□ENSURE□ □□□CONFIDENTIALITY□AND□STABILITY□OF□RATE□ESTIMATES 13 Human Papillomavirus (HPV) Sexually transmitted HPV infection is believed to be responsible for 90 to 95 percent of all cervical cancers. The risk of contracting HPV is influenced by a number of factors, including age, lifetime number of sexual partners, number of recent sexual partners, early age at first sexual contact, and race/ethnicity.19 Many HPV infections regress spontaneously. Research suggests that persistent infections are the most critical for later development of cervical cancer. Of the more than four dozen strains of HPV, 15 types appear to be most strongly implicated in cervical cancer, but it is not clear that these differences are important to mortality. HPV–16 accounts for about 50 percent of all cervical cancers. HPV types 18, 31, 33, and 44 together account for an estimated 20 percent.20 A recent study suggests that 13 percent of the U.S. population aged 12 to 59 years may have serum antibodies to HPV–16, demonstrating exposure and infection. In addition, women are more than twice as likely as men to have antibodies to the virus (17.9 percent versus 7.9 percent).21 However, the existence of HPV does not predict cervical cancer. A recent international study suggests an association between the number of full-term pregnancies and increased risk of squamous cell cervical cancer among HPV-positive women.22 A second study using the same data indicates that HPV infection together with oral contraceptive use for five or more years significantly increases cervical cancer risk, but no increased risk was found for HPV-free oral contraceptive users.23 At this time, HPV testing is not part of routine gynecologic examination, and current surveillance systems do not collect data on geographic variations in persistent HPV infection that would make it possible to study the relationship between geographic variations in HPV infection and variations in cervical cancer incidence. A combined Pap/HPV screening test recently was approved by the Food and Drug Administration (FDA) as a primary screen for cervical cancer and its precursors in women aged 30 and older, and as a followup to any abnormal Pap test result.24,25 The sensitivity of HPV testing, particularly in conjunction with cytology, has been confirmed in studies26,27 conducted through the NCI-funded ASCUS/LSIL Triage Study (ALTS), which is comparing alternative strategies for initial management of mildly abnormal Pap test results. However, the U.S. Preventive Service Task Force (USPSTF) concludes that “the evidence is insufficient to recommend for or against the routine use of HPV testing as a primary screening test for cervical cancer.” The USPSTF found poor evidence to determine the benefits and potential harms of HPV screening as an adjunct or alternative to regular Pap test screening.28 14 Screening Rates Since 1987, screening rates for all age and ethnic groups have been increasing,29 but screening rates remain higher among urban women compared with those living in rural areas30 and among younger women compared with older women.31 As Table 3 indicates, screening rates also vary by racial/ethnic group and by level of educational attainment. Retrospective reviews of the screening his- tories of women diagnosed with cervical cancer suggest that 50 to 70 percent of women devel- oping invasive cervical cancer either did not have a Pap test within the five years prior to diagnosis or had never been screened.32,33 BRFSS data (2002) indicated that, nationally, more than 85 percent of women had received a Pap test within the previous three years,34 though screening rates were higher among younger women and lower among older women (Figure 1). Factors that may contribute to lower rates among older women are related to less frequent contact with gynecologists after the childbearing years and/or after tubal ligation or hysterecto- my. Such lower rates may be appropriate in older women who have had hysterectomies and several subsequent negative Pap tests. Indeed, while the USPSTF strongly recommends screen- ing for cervical cancer in women who have been sexually active and have a cervix, it recom- mends against routinely screening women older than age 65 for cervical cancer if they have been recently screened, the findings were normal with Pap tests, and the women are not other- wise at high risk for cervical cancer.35 Lower rates may also be related to the failure of providers to recommend or perform the test on many older women as part of their primary care, particularly when the patient has comorbidities such as hypertension or diabetes that are perceived to pose a more immediate health threat.36 A special analysis of BRFSS data (1995–1999) conducted by CDC staff for the Roundtable and the Think Tanks also indicates that, among white women, those who are Hispanic, older, less educated, and living in high mortality counties have the lowest screening prevalence. Among black women, the only discernable difference, due to small sample size, is lower screening rates among those aged 40–64 years. Data available at the time of the Roundtable and Think Tanks were too sparse to conduct similar analyses of Asian, Native American, or other groups. NHIS data (1990–1999) show that Pap test use declines with age (while cervical cancer incidence and mortality risk increase with age). In addition, these data indicate that women Table 3. Screening Rates by Population Characteristic, Women 18 and Older, U.S. 2000 Race/Ethnicity Rate White (non-Hispanic) 82 Black (non-Hispanic) 84 Hispanic 77 American Indian/Alaska Native 77 Asian/Pacific Islander 67 Education (years, women 25 and older) 11 or fewer 74 12 81 13 or more 86 Source: 2000 National Health Interview Survey data, NCHS, CDC, July 2002. 15 with less education, less income, those uninsured, and those without a usual source of care are less likely to be screened. However, other studies37,38 also indicated that a substantial percentage of women in managed-care plans who developed cervical cancer had not been screened in the three years prior to diagnosis. Conversely, there also appear to be populations of women who may be getting screened more often than necessary (e.g., those with consistently negative Pap tests who continue to be screened annually). Screening rates also are affected by cultural factors (e.g., modesty, fatalism, prohibitions against examination by male health care providers) and competing life priorities (i.e., screening may not be perceived as important compared to income generation and fulfilling work and family responsibilities). Several meta-analyses indicate that patient- oriented interventions such as reminder letters, patient education, and financial incentives39,40,41 can be effective in increasing Pap test screening and followup after abnormal test results.42 These interventions have been especially successful in increasing Pap test use among historically underserved women.43 Likewise, provider- oriented interventions, including reminders and education, can increase screening rates.44 Provider interventions such as these are particularly important since primary care providers have reported not notifying patients Figure 1. Percentage of Women Responding to the Question: “How Long Has It Been Since Your Last Pap Smear?” by Age Group Respondents (%) 5+ years ago 18–39 40–49 50–59 Age (y) 60–64 65+ 100 80 60 40 20 0 past year past 2 years past 3 years past 5 years Source: Behavioral Risk Factor Surveillance System, 2002; survey data, Centers for Disease Control and Prevention, access at: http://www.cdc.gov/brfss/index.htm. of normal screening results45 and sometimes failing to notify patients about abnormal findings.46 Other Cervical Cancer Risk Factors Smoking has been identified as a contributing cause of cervical cancer.47 In addition, HIV- positive women are at increased risk for cervical intraepithelial neoplasia (CIN) and cervical cancer, which tend to be more progressive and aggressive in these women.48 The CDC lists cervical cancer as an AIDS-defining condition.49 Diagnosis and Treatment Patterns Available studies on diagnosis and treatment patterns for cervical cancer do not paint a clear picture of this aspect of care, particularly with regard to insurance status. Not surprisingly, data from ACoS-accredited hospitals, which tend to serve more affluent and better-educated populations,50 suggest that lower-income women are more likely than higher-income women to be diagnosed at a later stage of disease. Uninsured women are more likely to be diagnosed with late-stage cervical cancer than those with private insurance.51 Women in Medicare managed-care plans and HMO enrollees under age 65 tend to be diagnosed at earlier stages of disease than are women in traditional fee-for-service (FFS) plans.52,53 Limited state-level data provide additional information. For example, research in Michigan linking Medicaid, cancer registry, and death certificate data54 in that state indicate that about the same percentage of women in Medicaid HMO/managed-care and fee-for-service plans are diagnosed at early stages, but, as in other populations, later-stage diagnosis increases with age. Notably, older Medicaid patients in long- term care facilities are more likely to be diagnosed at later stages than women not in institutionalized settings. It is unclear what portion of this problem reflects comorbidities versus the place of residence. The Michigan research also suggests that poor patients with no health coverage prior to diagnosis and sub- sequent Medicaid coverage, and those without continuous coverage of any kind, are more likely to be diagnosed at later stages and die. The ACoS data indicate that black women are more likely than white women to receive no treatment after a diagnosis of cervical cancer, regardless of disease stage. The differences detected in this sample are likely to be magnified in rural and other medically underserved and poor populations. AHA data include non-ACoS hospitals; these hospitals are more likely to be smaller and have fewer full- time medical personnel. They more often are located in rural counties with low income and educational attainment, and in counties with higher cervical cancer mortality. They also are less likely to have oncology and radiation services. Eighteen percent of all U.S. counties have no hospital, and cervical cancer mortality is markedly higher in these counties compared to those with hospitals. SEER data for 1992–1997 indicate that patients with stage I disease (more than half of all cases), particularly those under age 50, are more likely to be treated with surgery alone than are older women, who are more likely to 16 17 be treated with surgery plus radiation. Women with more advanced disease are most likely to be treated with radiation as the primary modality.55 Adding chemotherapy to the treatment regimen lengthens survival but provides only modest reductions in mortality. A small population of women exists who are diagnosed but whose disease appears to be unstaged and untreated. Data on this population show no significant differences in treatment by race, income, or educational level at any disease stage. Overall, 5 percent of patients do not receive treatment, but up to 20 percent of women over age 65 with stage II–IV disease are receiving no treatment.56 Identified Needs and Strategies to Reduce Cervical Cancer Mortality And Improve Women’s Health CCMP participants identified four major areas of emphasis in which policy changes and interventions at Federal, state, and local levels could significantly impact women’s health and reduce cervical cancer mortality, particularly in high mortality geographic areas: • Access, including services, outreach, and navigation • Information and communication • Collaborations, partnerships, and advocacy • Research To reduce excess cervical cancer mortality, it is essential to understand the diverse cultures and risk factors of affected populations and the barriers to care they face. Populations in the geographic areas with highest cervical cancer mortality are quite heterogeneous, even within major racial/ethnic groups. Further, factors that appear to be patient group characteristics may in fact be system or infrastructure characteristics. However, some characteristics are shared by most women at high risk of cervical cancer mortality: poverty, lack of insurance, distance from health care, modesty (particularly among older women), fatalism concerning cancer, patriarchal cultures, a distrust of government and mainstream medicine, and a resilience that enables these populations to survive under harsh living conditions. Access—Outreach, Services, and Navigation Access includes both financial and physical access to necessary services as well as outreach and navigation services that enable access. CCMP participants identified seven (7) high- priority issues that address access to quality health care: 1. Enabling Patients to Have a “Medical Home” Data from the 2000 National Health Interview Survey57 show that the greatest disparities in Pap test use (during the previous three years), defined as differences between the highest and lowest groups for all age groups combined, were associated with having or not having a usual source of care (26 percentage point difference), age (a significant downward gradient with increasing age), and immigration within the past 10 years compared with U.S.-born or less recently immigrated women. Disparities also 18 were observed by level of education, family income, chronic disability, race, and ethnicity, but these differences were not as great as those due to health care access, age, and immigration. CCMP participants emphasized strongly the need for women at risk for cervical cancer, other cancers, and other chronic conditions to have a “medical home”—a usual source of medical care where the patient can receive screening and counseling, experience continuity of care, and develop trusting relationships with the medical staff. The medical home also can serve as a hub with linkages to other community services. Among the suggestions for improving access to health services were: • Leveraging resources by adding cervical health and other screening services to the services offered at existing clinics, health centers, or other provider sites. • Building necessary infrastructure where it does not yet exist. • Basing resource distribution and redistribution on small-area analyses of areas of greatest need. • Encouraging state health departments to take the lead in efforts to improve services for high-risk populations. CCMP participants suggested that access issues in high mortality regions may have more to do with inadequate primary, tertiary, and support service infrastructure and problems linking primary care clinics to the hospital system than factors such as race or ethnicity. Primary health services available in regions experiencing high rates of cervical cancer and other disease mortality offer the possibility of coordination of a more extensive range of social and medical services. These include Health Resources and Services Administration (HRSA)- funded health centers, which include community health centers (CHCs), migrant health centers, homeless health centers, and public housing primary care centers.58 For more than 30 years, HRSA-funded centers have been a safety net for underserved communities, providing primary health care for some of the nation’s most vulnerable populations. These centers also receive a substantial portion of their revenues from the Centers for Medicare and Medicaid Services (CMS) through Medicaid funding. Approximately 40 percent of CHC patients are uninsured, and almost 90 percent are low income. More than two-thirds of patients have incomes below the poverty level. More than 10 million people receive medical services at almost 4,000 sites. In some communities, CHCs are the predominant source of care.59 In addition to the network of HRSA-funded centers, a patchwork of approximately 3,300 rural health providers serves many of the underserved women in areas of high cancer mortality. Many of these rural health providers are private practices, clinics within rural hospitals, or small part-time clinics in remote areas. Those certified as rural health centers by CMS receive reimbursements from Medicaid and Medicare that include an added reimbursement factor to help them remain viable in their locations. Because of financial pressures, however, the number of these rural health centers is declining in many areas, including Appalachia. To maintain and better integrate health services in rural areas, it will be important to include this group of providers. Though the rural providers are not obligated to participate in a demonstration project or other initiative to establish an integrated rural health system, they may be able to play a role in cervical cancer education and counseling. A sampling of the range of types of services that could be organized to play a role in cervical cancer education and counseling. is presented in the following paragraphs. Other nonfederally funded health centers exist that are structurally similar to HRSA centers. Like HRSA and Indian Health Service (IHS) or tribally operated primary care centers, many have been designated Federally Qualified Health Centers (FQHCs) and, as such, receive cost-based reimbursements from Medicaid and Medicare.60 As in federally funded programs, the percentage of patients who are enrolled in Medicaid managed-care plans has increased in recent years. In some locations, state and county health department facilities and private primary care practices also provide primary care. States and counties may provide categorical maternal and child health care, immunizations, STD (sexually transmitted disease) treatment, or other targeted programs through Federal or other grants. Some patients who have no primary care facility in their local area or who choose not to seek care at available clinics may rely on emergency rooms for their primary care needs. In a project currently under way in Texas and funded by the Robert Wood Johnson Foundation, researchers are attempting to link medical and community services to provide a medical home with culturally responsive health care for women at high risk for cervical cancer who do not have a physician or health insurance. Representatives from area primary and specialty care facilities, public housing health service programs, other system components, academia, and the business community are participating on the project steering committee. 2. Eliminating the Disconnects Between Primary Care, Screening Services, and Followup/Treatment at Hospitals Linkages between HRSA primary care centers and area freestanding screening and hospital services often are tenuous. Similarly weak connections, if any, occur between other primary care providers and local hospitals in areas with high cervical cancer mortality. Moreover, these services may be geographically distant from each other and from patients living in rural areas. As a result, established procedures for rapidly referring women who have abnormal Pap tests may not be in place. In such situations, women may become lost in or drop out of the system and not receive the diagnostic, follow-up, or treatment services they 19 need. In addition, it is sometimes unclear whether the primary care center or an oncologist should manage the care of women with advanced cervical cancer. Currently 50 states, 4 U.S. territories, 13 American Indian/Alaska Native organizations, and the District of Columbia have elected to participate in the Breast and Cervical Cancer Mortality Prevention Act of 1990.61 In 2000, Congress passed the Breast and Cervical Cancer Treatment and Prevention Act to help make treatment services more accessible to women enrolled in the National Breast and Cervical Cancer Early Detection Program (NBCCEDP). As of January 1, 2005, 49 states and the District of Columbia have received approved Medicaid amendments to participate in this program. Under it, states extend Medicaid coverage to women whose cervical abnormality is detected through the CDC’s NBCCEDP. The Federal med- ical assistance percentage (FMAP) rate covers 65 to 85 percent of the cost of treatment, with state Medicaid dollars covering the remainder. While an important step in linking screening with care, the early detection program covers only 12 to 15 percent of eligible women (usually women with family incomes 250 percent or less of the Federal poverty threshold).62 Disconnects between primary care, screening, and treatment often are compounded by transportation/distance from care, payment and childcare issues, and lack of understanding by patients of the need to pursue follow-up care in the event of an abnormal Pap test. Difficulty in navigating the health system appears to be a major barrier. Eliminating these disconnects for both physicians and patients was seen as a crucial step in improving care for women with cervical abnormalities and other health conditions for which continuity of care is essential. The need to coordinate federally funded medical and social services in rural areas was emphasized repeatedly. The HRSA, NCI, and CDC pilot Health Disparities Collaboratives are addressing many of these issues in a pilot project being conducted in several HRSA centers. Among the goals of this project is a plan to perform Pap tests on 90 percent of female primary care patients, with test result notification within 30 days. Follow-up colposcopy is provided for women needing further evaluation; the program seeks to perform at least 80 percent of necessary follow-up procedures within 90 days. For women diagnosed with early cervical cancer, the goal is to have treatment begin within 90 days of diagnosis. The program is being documented and evaluated to determine if it can be expanded to the entire network of HRSA pri- mary care clinics. A description of these cancer health disparities collaboratives can be found at the Web site: http://healthdisparities.net. 3. Reaching Women Who Have Rarely or Never Been Screened for Cervical Cancer and Other Diseases To reduce cervical cancer mortality, it is critical to reach rarely or never screened women, since half of all invasive cervical cancer cases are found in women either who have not received a Pap test in the previous five years or who have never had a Pap test.63,64 20 21 The CDC NBCCEDP was an important step in attempting to reach more low-income women with breast and cervical cancer screening, particularly those in rural areas. On average, however, the NBCCEDP sites reach only 15 percent of eligible low-income women, and it appears that some programs may be screening some women more frequently than is necessary. ACS and USPSTF guidelines recommend Pap tests every three years for women with previous normal test results; the NBCCEDP screens all women annually. Some CDC program sites have been more successful in reaching eligible women than others. CDC is conducting evaluations to determine why these programs are succeeding and how contributing procedures, activities, and other factors can be incorporated to strengthen performance at other program sites. Greater numbers of outreach workers are needed, both in the CDC screening programs and in other community health programs, to locate rarely or never-screened women and facilitate their entry into screening programs. Outreach workers are particularly needed in rural areas, especially in counties with limited or no county health services. Such workers, typically members of the community who receive training to fulfill this role, are sometimes referred to as lay educators, community health advisors, community health representatives, promotoras, navigators, or by other titles. They are key players in attracting high-risk women into the health care system through culturally appropriate education, lay counseling, and other interventions. These workers also may arrange medical appointments, accompany women to screening and follow-up appointments, and help with transportation and childcare needs. CCMP participants suggested that outreach and screening be expanded to include emergency rooms, other providers of women’s health services, and work sites. To the extent possible, the number of mobile units providing screening, follow-up, and at least limited treatment services should be increased. In addition, screening services should be made available in the evening and on weekends. Free screening could be provided to groups with limited health care access, such as migrant health workers and the poor. CCMP participants reported that, for example, a pilot program in Santa Clara County, CA, is reaching out through media and other communication channels to hard-to-reach Asian women, many of whom are relatively recent immigrants, to provide Pap tests at a free clinic. The screening program also is proving successful as a way to draw women into the system for the care of other health problems. However, when cancer is diagnosed, unless a woman is Medicaid eligible, the program currently faces the same lack of referral resources as HRSA clinics. In another pilot effort reported by the CCMP participants, NCI, CDC, the American Cancer Society, the Centers for Medicare and Medicaid Services (CMS), and the Department of Agriculture Extension Service are testing the benefit of using Extension Service workers to reach rural women in eight states with evidence- based breast and cervical cancer screening promotion program materials. Some of the actions proposed by CCMP participants may require supportive policy, modified programs, and funding. For example, they recommended that lay health workers, many of whom are volunteers, should be paid, particularly when they are fulfilling a role the physician is unable to undertake due to time constraints. Moreover, to sustain the effort, such workers should become part of an institution’s economic infrastructure rather than being supported by grant funding. CMS representatives indicated that Medicaid currently can provide matching funds for community health advisors if the state requests a Section 1115 waiver. Section 1115 of the Social Security Act permits authorization of experimental, pilot, or demonstration projects that the Secretary, HHS, deems likely to assist in promoting the objectives of the Medicaid statute.65 Medicaid also provides funding through disease management organizations (DMOs) that could provide an option for supporting this type of staffing within a system demonstration project in a specific community. While cancer currently is not one of the diseases specified for this type of funding, it was suggested that NCI work with CMS leadership to include cancer in the categories of disease supported via this mechanism. Some CCMP participants, however, maintained that further research is needed to confirm the efficacy of the community health advisor role. In addition, the recently revised cervical cancer screening guidelines66 may enable NBCCEDP and other program resources to be redirected to outreach activities and staffing. The American Cancer Society recommends: 1. Cervical cancer screening should begin approximately three years after a woman begins having vaginal intercourse or no later than by the time she is 21 years of age. 2. Cervical screening should be done annually with regular Pap tests or every two years using liquid-based tests. At or after age 30, women who have had three normal test results in a row may get screened every two to three years. But the doctor may suggest getting the test more often if a woman has certain risk factors such as HIV infection, a weak immune system, or multiple partners. 3. Women 70 years of age and older who have had three or more normal Pap tests and no abnormal Pap tests in the last 10 years may choose to stop cervical cancer screening. 4. Screening after total hysterectomy (with removal of the cervix) is not necessary unless the surgery was done as a treatment for cervical cancer or precancer. Women who have had a hysterectomy without removal of the cervix should continue cervical cancer screening at least until age 70. Primary care physicians and other providers also play a vital role in reaching rarely or never-screened women. These providers need to be encouraged to ask women patients about screening history, provide cancer education, and perform screening when indicated on women who have come in for other reasons. For this strategy to be successful, particularly with 22 providers who must see 50 to 60 patients daily, reimbursement would have to be available for such “opportunistic” screening (see also Improving Coverage and Reimbursements, on page 24). 4. Providing Patient Navigators to Help Women Through the Health System Once an Abnormality Has Been Detected Some women who are screened and found to have an abnormality never receive follow-up diagnostic and treatment services because they are fearful or ashamed of a possible diagnosis, have no telephone or fixed address (e.g., homeless or migrant women), or do not understand the meaning of the test result and the need for additional care. Some patients do not follow up on abnormal test results because they lack transportation or child care needed to attend medical appointments. For all of these reasons, culturally competent patient navigators or case managers are needed to ensure that women enter the health system and receive the guidance and assistance needed to receive necessary treatment. Patient navigators typically assist women with scheduling and keeping appointments; explain medical terminology and the importance of treatment; and help women obtain financial, medical, transportation, child care, and other assistance they may need. It is essential that navigators be highly knowledgeable about the local health and supportive care systems and resources to fulfill this role. A number of such programs, tailored to local needs, already are under way or planned in communities across the country. One such program now operating in nine sites (though not limited to cervical cancer) is sponsored by CRCHD. http://healthdisparities.net. The need for navigators was considered a high priority by CCMP participants. Funding to support them could come from Federal, state, or local sources. A representative from CMS indicated that Medicaid can pay for case management (patient navigator) services through the 1115 waiver mechanism as well as in conjunction with “hybrid” proposals for expanded Medicaid coverage. For patient navigators to be optimally effective, an infrastructure needs to be in place. Physicians must be up to date on screening and treatment guidelines, and patient reminder and tracking systems must be in place to ensure that patients are offered information and receive all of the services needed in a timely manner. Similarly, patient navigator programs should include an evaluation component to ensure that the program is effective. As with Community Health Advisors, further research on the costs and benefits of patient navigator programs was urged. 5. Increasing the Number of Providers of the Patient’s Gender/Race/Ethnicity Modesty, distrust, and cultural taboos may cause women in high cervical cancer mortality areas to avoid screening and follow-up care if only male physicians are available to conduct the examination. Increasing the number of female providers, particularly providers of the patient’s race/ethnicity, is an essential step in breaking down resistance to screening and saving 23 women’s lives. African Americans, Hispanics/Latinos, and Native Americans together account for approximately 21 percent of the U.S. population, but only 8.6 percent of physicians.67 The proportion of women physicians in these groups is even lower. While increasing the number of women physicians from these racial-ethnic groups is an important long-term strategy, CCMP participants also urged the training of nurses and other health care providers to perform Pap tests and colposcopy. Women in the above-listed racial- ethnic groups are much better represented among nurses and other types of providers, making this a more feasible short-term strategy. National support to achieve this objective could include national training programs and scholarships, National Health Service Corps (NHSC) placements, and provider incentives. 6. Improving Coverage and Reimbursements Coverage and reimbursement issues permeated the discussions of access to cervical cancer screening and treatment. Physicians are reducing the number of new Medicare and Medicaid patients they accept because of low reimbursements, and reduced reimbursement rates are straining the budgets of publicly funded health centers that count on Medicare and Medicaid programs for major portions of total patient care revenues. Reimbursement is needed to specifically encourage primary care and other providers to ask about Pap screening history, provide cervical cancer education, and perform screening when indicated on women who have come in for other reasons. Unfortunately, categorical funding grants often constrain providers’ ability to provide multiple services during a single medical encounter. CCMP participants assigned a high priority to the suggestion that over a 10-year period, the CDC National Breast and Cervical Cancer Early Detection Program and Treatment Act should be fully funded to provide 100 percent coverage of eligible women, including patient navigator services. In addition, it was recommended that Medicaid or Medicare coverage for at least one year should be provided to any uninsured or underinsured woman (living in poverty) with cervical cancer whose disease was not detected through the NBCCEDP. According to a CMS representative, a woman who receives cancer treatment through Medicaid remains eligible for necessary follow- up care and additional treatment as long as she has no other insurance coverage, is under age 65, and continues to require treatment. If a patient reaches age 65 after her initial course of treatment is complete, she will be dropped from the Medicaid rolls. Older women who are not eligible for Medicare (e.g., recent immigrants) but later need additional treatment currently must find another source of payment. Paperwork related to federally funded programs should be easy to read and complete, and be streamlined to encourage provider participation. Recent legislation68 includes a provision to help increase and equalize reimbursements between urban and rural areas. This, however, may not be sufficient to overcome rural payment issues. 24 7. Improving Telemedicine and Multidisciplinary Consultations Multidisciplinary consultation prior to cancer treatment selection is considered a key element in optimal cancer care, but this option currently is unavailable to most women in high cervical cancer mortality areas. Telemedicine may provide a means of bringing review of diagnostic test results and treatment recommendations to areas without cancer specialists. CMS currently provides reimbursement for these services. Information and Communication Current efforts at information dissemination and communication about cervical cancer risk, screening, and treatment are not reaching high- risk populations effectively. Possible strategies in seven (7) key areas have potential to improve information outreach both to patients and to health professionals: 1. Improving Awareness and Knowledge Levels About Cervical, Breast, Colorectal Cancer and Other Screenable/Treatable Disease Mortality The Institute of Medicine (IOM) has underscored the need for patient education programs to increase patients’ knowledge of how to access care and participate in treatment decisions.69 A second Institute report in 200370 reiterated the need for public and private organizations to work to improve the public’s understanding of cancer prevention and early detection and their importance for reducing cancer mortality. CCMP participants suggested that public and private agencies develop and implement a cervical cancer awareness and education media campaign that would be funded nationally but implemented regionally so that it could be tailored to high-risk populations and geographic areas. In addition to providing information about cervical cancer, its prevention and treatment, and available services, a key focus of these campaigns would be to raise public awareness of the longstanding disparities in cervical cancer mortality and to motivate action. 2. Developing and Providing Linguistically and Culturally Appropriate Information The Federal Government established standards for Culturally and Linguistically Appropriate Services (CLAS) in health care in recognition of the impact of limited English proficiency and linguistic isolation on health disparities. The proportion of linguistically isolated households (i.e., in which no person over 14 years old in the household can communicate in English) remained at approximately 5 percent of all U.S. households for the past decade. However, in addition to the traditional destination states of many immigrants (e.g., California, Texas, New York), some states have seen dramatic increases in this population (e.g., 10 percent of Asian households in Kentucky, 18.6 percent of Spanish-speaking households in North Carolina).71 It is therefore essential to address both health literacy and overall literacy issues that affect communication with populations at high risk of death from cervical cancer, including both native- and foreign-born individuals. 25 26 Native-born populations in high cervical cancer mortality areas tend to have limited educational attainment and require materials written at a level that matches their reading levels. For Asian, Hispanic/Latina, and other immigrant populations with high rates of cervical cancer incidence and mortality, materials must address not only limited or complete lack of literacy in English but also, in many cases, limited literacy in their native language. Low literacy among women of all races and ethnicities has been shown to be directly linked to lower cervical cancer screening knowledge.72 Materials designed for visual learners (e.g., pictographs) and the oral transfer of information (e.g., videos, story telling, talking circles) are needed to reach women at all levels of literacy with essential information on cervical cancer prevention and treatment. Few such materials are available and more need to be developed. Educational materials must be both understandable and culturally appropriate. The continuing inflow of immigrants from numerous countries, the diverse cultures that exist even among immigrants from the same nation, and the regional cultural differences among native-born women create a massive challenge to meeting their information needs. Experience with Native American populations, for example, indicates that materials developed locally or regionally for specific populations often are the most effective, but this approach is costly and time consuming. Regional efforts can, however, be supported at the national level. The Native CIRCLE (Cancer Information Resources Center and Learning Exchange) program, cosponsored by NCI and the Mayo Comprehensive Cancer Center, stimulates local development of cancer education materials for diverse Native American and Alaska Native audiences. It has developed materials with Native American themes that have proven useful for a number of tribes or have been adapted to reflect specific Native cultures (http://mayoresearch.mayo.edu/mayo/ research/cancercenter/native.cfm). Similarly, an NCI-supported study demonstrated that culturally and linguistically tailored educational materials increased cervical cancer screening among Chinese American women in Seattle.73 Evaluation of these programs is ongoing to determine their efficacy in reaching specific groups of women. Translation needs are another critical component of information services for women at high risk of cervical cancer mortality. Literal translations often do not increase the readability and comprehension levels. Among others, the IOM74 has noted that unaddressed language barriers can affect the delivery of adequate health care due to poor information exchange, misunderstanding of physician instructions, inadequate shared decision-making, loss of important cultural information, or ethical compromises. In addition, language barriers can negatively affect adherence to medication regimes, appointment attendance, and patient satisfaction. However, due to the lack of translators in the variety of languages required, particularly translators who are sufficiently familiar with medical terminology to translate information about treatments and procedures with accuracy, health care providers in many areas must rely on family members, including children, or persons who are strangers to the patient to provide translation services. Innovative approaches to developing interpreter networks tailored to the populations in a given geographic location are sorely needed. NCI recently has begun a shift from translation of cancer education materials to “trans-creation,” which often renders a different but more culturally appropriate framing of information. Of note, the California legislature passed SB 853, Culturally and Linguistically Appropriate Services Bill, in 2003.75 This bill requires the Department of Managed Health Care to adopt, not later than January 1, 2006, regulations establishing standards and requirements to provide health care service plan enrollees with access to language assistance in obtaining health care services. The regulations would require standard and specialized health care service plans to implement programs to assess enrollee needs, to provide translation and interpretation for medical services and translation of vital documents to enrollees, and to report to the Department regarding internal policies and procedures related to cultural appropriateness. The bill requires the regulations to determine that a health care service plan is compliant with requirements if it meets the same or similar standards imposed by the Medi- Cal program. Opponents of the bill fear that interpreter costs could exceed provider reimbursements under the state’s Medicaid plan, causing physicians to cease providing services to this largely underserved population. However, according to one review of funding options for language services and examples from the field,76 all states can obtain a 50 percent Federal match for language services provided to Medicaid and State Child Health Insurance Program (SCHIP) recipients, or a greater percentage if language assistance is adopted as a covered service under the state health plan. 3. Enhancing Health Professional Education The literature shows that systems, including reminder systems and patient tracking with feedback to physicians, are far more effective (than Continuing Medical Education [CME] or other educational or information) to ensure that physicians provide preventive services. Some evidence also suggests that systems can be set up to promote culturally appropriate care and that these are more effective than individual- based education efforts. Participants emphasized the need to educate practicing physicians and those still in training in the following high- priority areas: (1) encouraging physicians to review screening history and advise women, in- cluding Medicare beneficiaries, about screening at each medical encounter, (2) communicating the importance of Pap test screening to women seeking treatment for sexually transmitted diseases, and (3) ensuring culturally competent care in cervical cancer primary, secondary, and tertiary prevention and treatment. With regard to cultural competency, the American Medical Association has produced a video and accompanying workbook for use by physicians to sensitize them to the importance of 27 accommodating cultural differences when providing care.77 In addition, a recent study suggests that physicians in health centers need assistance in recognizing and accommodating patients with limited literacy to ensure the transfer of cervical cancer-related information.78 Participants cited the need for unambiguous and simply stated guidelines on cervical cancer screening, abnormal cytology, and treatment. Efforts are needed to improve providers’ knowledge about screening and follow-up guidelines and about how to take advantage of opportunities to discuss screening when patients come in for other reasons. In California, for example, a short CME course targeting Vietnamese physicians was highly successful in raising their awareness and knowledge about cervical cancer and screening and follow-up guidelines. The Health Disparities Collaboratives are working to enhance provider knowledge in these areas and to coach health center providers on how best to provide these services to diverse populations. Training also is being provided on cultural competence, and retraining to accommodate staff turnover is being addressed. Further, providers are being trained to use a cancer registry system, the Patient Electronic Care System (PECS), that should assist them in monitoring patient progress. Funding is needed for scholarships to increase the number of female and minority health providers, including physicians and other facilitators, serving underserved populations. Training funds also are needed for training community health advisors, case managers, and patient navigators. 4. Enhancing Provider-Patient Communication Communication between patient and provider was identified as an area needing substantial improvement. For example, reminder systems are needed to help inform both patients and providers when screening and follow-up care is due. CHC and other programs currently are experimenting with different types of reminder systems. The care model being implemented in the HRSA/NCI/CDC cancer control Health Disparities Collaborative pilot centers includes provider-patient shared decision-making. Under this model, the provider is to assess and ask the patient about her conviction to continue screening, identify barriers to screening, and inform the patient about screening and followup. The information is to be tailored to address identified barriers, and a written plan for screening and followup is to be provided to the patient before she leaves the provider’s office. Tools to assist providers in this communication are being designed and tested. In addition, patients are to be given self- management worksheets that help them keep track of when they are due for screening and other care. 5. Using Lay Health Workers to Increase Population Awareness of the Importance of Screening Volunteer and paid lay health workers are a crucial component of the cervical cancer education network. These workers, drawn from the community, have credibility and a trust relationship with patients that may take 28 physicians weeks or months to cultivate. Unlike most physicians, these workers also have the time to explain information in detail and to meet with target-population women in familiar settings including churches, health fairs, other community events, and women’s homes. Geographic areas with high cervical cancer mortality are populated primarily by mar- ginalized populations with high rates of diabetes and other health problems. Educational program funding for these conditions is scarce; some participants felt that additional evidence may be needed to persuade local resource allocators to fund lay health worker programs targeting cervical cancer. 6. Providing a Central Resource for Cervical Cancer and Other Disease Information and Intervention Strategies A national cervical cancer Web site was suggested that would link national organiza- tions and their informational and other resources so that states and localities could easily obtain publications and materials. Native CIRCLE helps to disseminate cancer education materials for American Indian and Alaska Native populations but does not reach all populations with high cervical cancer mortality. NCI, in collaboration with CDC, the Substance Abuse and Mental Health Services Administration (SAMHSA), and the American Cancer Society (ACS), has launched Cancer Control PLANET (Plan, Link, Act, Network with Evidence-based Tools, http://cancercontrolplanet.cancer.gov) to assemble in one place information on evidence-based interventions and tools that can be downloaded by provider organizations and used in or adapted to varied care settings. If this or a similar Web site were established as a clearinghouse to include cervical cancer control programs along with other evidence-based cancer control programs, targeted efforts would be needed to make state and local health administrative and provider organizations, voluntary and advocacy organizations, and the private provider network aware of this resource. 7. Improving Medical Records Maintenance and Retrieval Systems Improved medical records maintenance and retrieval systems were cited as important for retaining patients in the health system, helping providers to rapidly access and review patients’ medical information, and reducing cost by eliminating the need for repeat procedures. Critical to the success of using medical records to improve services is regular and compas- sionate feedback to physicians. As advances in information technology become more widely used by providers, monitoring of the impact of these technologies on health disparities will be needed. Collaborations, Partnerships, and Advocacy To reduce health system fragmentation, improve continuity and quality of care, and make the most efficient use of available resources, expanded, new, and innovative collaborations and partnerships are needed among national, regional, and local providers and other stakeholders. In addition, increased advocacy concerning cervical cancer is needed. 29