Table of Contents

• Acknowledgments

• Summary

• Introduction

  • Adverse Health Outcomes Defined - Incidence and Mortality

  • Purpose of the Report

  • Registry Background

  • Primary Goal of SRRHIS

• Findings

• The Five Most Common Cancer Site Reported in the SRRHIS Region, 1991-1993

• Cancer Incidence by Primary Site

  • All Cancers Sites Combined

  • Bladder

  • Brain and Other Nervous System

  • Breast

  • Cervix

  • Colon and Rectum

  • Esophagus

  • Kidney and Renal Pelvis

  • Leukemia

  • Lung and Bronchus

  • Lymphoma

  • Melanoma of the Skin

  • Multiple Myeloma

  • Oral Cavity and Pharynx

  • Ovary

  • Pancreas

  • Prostate

  • Stomach

  • Thyroid

  • Uterine Corpus

• Appendices

  • Appendix A: The Savannah River Site and Surrounding Area (with Figure 29)

  • Appendix B: Demographic Description of the SRRHIS Population (with Table 2)

  • Appendix C: Methodology

  • Appendix D: Cancer Mortality Studies in this Region

  • Appendix E: Completeness of Reporting and Quality Control Issues

    • Legislative Issues

    • SRRHIS Area Residents Diagnosed Outside Georgia and South Carolina

    • Use of Death Clearance Procedures to Identify Missed Cases

    • Under-Reporting of Cancer Cases

    • Confidentiality

  • Appendix F: Numbers of Cancers Reported, by Primary Site

    • Table 3: Invasive and Insitu Incidence Cases by Site, Race and Gender, SRRHIS, 1991-1993

    • Table 4: Invasive Incidence Cases by Site, Race and Gender, SRRHIS, 1991-1993

    • Table 5: Invasive Incidence Cases by Site, Race and Gender, Georgia, SRRHIS, 1991-1993

    • Table 6: Invasive Incidence Cases by Site, Race and Gender, South Carolina, SRRHIS, 1991-1993

  • Appendix G: Age-adjusted Incidence Rate for Invasive Cancers, by Primary Site

    • Table 7: Age-adjusted Incidence Rates for Invasive Cancers by Primary Site, Race and Gender, SRRHIS, 1991-1993

    • Table 8: Age-adjusted Incidence Rates for Invasive Cancers by Primary Site, Race and Gender, Georgia SRRHIS, 1991-1993

    • Table 9: Age-adjusted Incidence Rates for Invasive Cancers by Primary Site, Race and Gender, South Carolina,SRRHIS, 1991-1993

    • Table 10: Age-adjusted Incidence Rates for Invasive Cancers by Primary Site, Race and Gender, Atlanta SEER Area, 1990-1992

    • Table 11: Age-adjusted Incidence Rates for Invasive Cancers by Primary Site, Race and Gender, NCI SEER Program, 1990-1992

  • Appendix H: Summary Stage by Primary Site by Race, SRRHIS, 1991-1993

    • Table 12: Summary Stage by Site for Whites, SRRHIS, 1991-1993

    • Table 13: Summary Stage by Site for Black, SRRHIS, 1991-1993

  • Appendix I: Technical Terms

• References

This first report of the SRRHIS cancer registry is dedicated to the members of the Steering Committee, past and present, with enormous gratitude for their understanding and support over the years since SRRHIS's initiation in 1991. They have been steadfast in their belief in the effort, and they have affirmed their commitment to the project goals through attendance at the quarterly meetings over the years and by undertaking special jobs for the project. The present and past members of the Steering Committee are listed here so as to acknowledge them individually for their contributions to the SRRHIS program:

The staff wishes to convey special thanks to two former Chairs, Andrew Rea of Savannah, Georgia and Mildred Ford of Blackville, South Carolina. The success of a project such as SRRHIS requires the "voluntary" cooperation of many individuals, institutions and organizations, countless discussions and the collection of massive amounts of case data, long before the product becomes evident. Unusual fortitude and patience is required. Mr. Rea served the Steering Committee and staff for three years, 1992-1995, as the first Chair, and Ms. Ford served as Chair in 1995. Ms. Ford was the first member of the Steering Committee, appointed in 1992. She and Mr. Rea were always there and ready to work, providing encouragement and perseverance in the early critical years. These qualities contributed to both direction and staying power for the project. They were assisted by a loyal set of supporting members. This report is due in large measure to their belief in the ultimate benefit of the project to the citizens of the region.

Without the support of the SRRHIS area medical facilities, 17 in South Carolina and 24 in Georgia, the registry would not be possible. While it was necessary to justify to each facility the need for cooperation, and to convince all of them and their professional oversight committees that proper protections insuring confidentiality were in place, in the end the overwhelming majority joined with SRRHIS in the community interest. For their help and understanding no amount of gratitude is sufficient.

In order to help make the residents of the SRRHIS region aware of the registry, and to keep them informed about the findings, the staff appreciated the potential role of the media. To that end, the media were notified and kept informed of all SRRHIS activities. Staff were always available and accommodating for interviews. The extent to which the newspapers followed the project is revealed in part by the fact that two of them editorialized the need to evaluate potential adverse health effects in the area, with the role of SRRHIS emphasized. Their part in disseminating information to the public was critical.

The professional support and advice provided by the program and fiscal staff of the United States Department of Energy (DOE), the source of funding for this initiative also is gratefully acknowledged. They were always helpful, both in advising on compliance with Department policy, and for assisting SRRHIS in finding ways to get its job done. At no time did they try to direct or control the SRRHIS process. The DOE project officer for most of this time, Dr. Bonnie Richter, has especially impressed the SRRHIS staff and the Steering Committee with her devotion to quality and service. Her background as an epidemiologist, with significant research experience, has been helpful to SRRHIS as a source of technical as well as program advice. She has served as official program overseer for the Environmental Health Division of DOE, and her watchful eye has been evident, and welcome, from the beginning. Certainly none of this could have been accomplished without the support of the Department of Energy's top leaders, including Dr. Tara O'Toole who directs the Environmental Health Division under Secretary Hazel O'Leary.

Finally, we wish to acknowledge and thank all the staff of the SRRHIS program, both at Emory University in Atlanta, Georgia and at the Medical University of South Carolina in Charleston, South Carolina, for their continued dedication and hard work.

This document represents the first data report of the cancer registry of the Savannah River Region Health Information System (SRRHIS). SRRHIS, a joint project of the Medical University of South Carolina in Charleston, South Carolina, and Emory University in Atlanta, Georgia, is funded through a grant from the U.S. Department of Energy. The purpose of the cancer registry is to identify all residents of a 22-county region adjacent to and on both sides of the Savannah River, who were diagnosed with cancer since 1991. Through standard practices established by the National Cancer Institute (NCI), and endorsed by the North American Association of Central Cancer Registries (NAACCR), the SRRHIS cancer registry collects information from hospitals, laboratories and physicians, and edits and computerizes that data. Quality control checks of completeness and accuracy are performed to confirm that conclusions are valid.

This report defines the cancer registration project and presents data regarding invasive cancer for Georgia and South Carolina SRRHIS registrants, separately and combined, for the white and black, male and female populations. Age-adjusted cancer rates are the primary measure of cancer incidence, and equivalent national and Metropolitan Atlanta rates are provided for comparison. These rates are presented graphically and in tables, by type of cancer. In addition, tables indicating the percentage of subjects presenting with localized, regional or distant disease are included, as these are an excellent measure of access to care and use of clinical diagnostic services.

The report finds that overall, the rates of cancer in this region are lower than the national rates, or those of Metropolitan Atlanta. This is consistant with other findings that cancer rates tend to be lower in less metropolitan areas. Community concerns about excess rates in the region will be addressed by this report.

The primary exception to this pattern of lower rates is the 45% excess of invasive cervical cancer among black women in this region, compared to the national and Atlanta rates. Cervical cancer is preventable, by the early identification of cervical lesions before they become cancer, through Pap smear screening. The observed excess suggests the need in this population for better health education and access to affordable Pap smear screening programs.

In general, racial and gender cancer patterns were observed that are similar to those of the national and Metropolitan Atlanta data, with highest rates found in black males, followed by white males, white females and black females. The specific types of cancer for which racial differences were found are described. A number of these can be explained by known probable risk factors - diet, smoking and, in the case of melanoma, skin pigmentation.

In this population, cancers in blacks were diagnosed at later stages, suggesting, again, a need for better health education and access to care.

The continuation of the SRRHIS project will enable the pattern of cancer over time to be monitored in this region, will provide the local population with scientific evidence of the extent of the cancer problem in their community, and will allow the evaluation of success of cancer control programs in the area.

Introduction

This first report about the occurrence of cancer in the Savannah River region of South Carolina and Georgia is based on more than 13,000 cases of cancer identified as newly diagnosed among SRRHIS residents during the years 1991 through 1993. The number of cases is sufficient to provide a credible statistical estimate of cancer occurrence in the region for all cancers combined and for a number of specific cancer types of interest to the residents. This presentation will compare SRRHIS incidence rates with those representing metropolitan Atlanta and those representing nine large population based registries comprising the National Cancer Institute's Surveillance, Epidemiology and End Results (SEER) program. The Atlanta registry participates in the SEER program, and is the geographically nearest population based registry. The national SEER program includes high quality data on over ten percent of the United States population, and is often used to approximate a national rate.

The number of cases for this three year period, however, is not yet large enough for some important analyses. For instance, there are an insufficient number of cases to permit a tabulation of individual cancer types (for example, leukemia) for individual cities, counties or smaller units. Additional years of data collection will be required in order to support statistically robust tabulations of single cancer types in smaller geographic areas. This kind of "small area analysis" is necessary to determine whether cancer cases tend to cluster at certain places, or whether they are spread about the county more or less evenly. Finally, many more years of data collection will be necessary to provide adequate evaluation of time trends for the individual cancers. With three or even five years of data, one cannot determine with any degree of accuracy whether the rate for a given cancer, say breast cancer, is rising or falling in an area this size. To this end, the staff is continuing to collect 1994-95 data, and soon will begin abstracting 1996 cases. Even so, because a great deal of work is required to check, re-check, and assure the quality of all cases, an updated report, with additional years of data, is several years away.

The SRRHIS staff is enormously proud and delighted to present this first tabulation of cancer incidence data to the residents of the Savannah River region of Georgia and South Carolina. In addition, the information will be made available to several important groups: health authorities, scientists, and health practitioners. The amount of cancer being identified, the type, and the trends are of great importance to all these audiences, and for each it will have special meaning. How else can we know, for example, whether given cancers are more frequent here than elsewhere, or whether the cases are being found early or late in the disease cycle?

Adverse Health Outcomes Defined - Incidence and Mortality

When epidemiologists speak of adverse health outcomes, in relation to some known or perceived environmental hazard(s), they generally are referring to morbidity (occurrence of disease) and mortality (death) measures. If a person is exposed to a hazard and is significantly affected, he or she could become ill, as with a cancer, or in extreme cases, death could ensue. The first class of information is called incidence data; the second is mortality data. It is important to have access to both kinds of information when evaluating the environmental health of a population. The SRRHIS cancer registry program is making incidence data for the region available for the first time.

Before the SRRHIS registry was begun, the only way to evaluate the impact of cancer on this population was to review cancer mortality records. Mortality studies are important in understanding the toll of cancer in a population group, but they greatly underestimate the full impact of cancer on our community. The SRRHIS cases collectively permit the calculation of incidence rates (new cases per 100,000 population) of newly diagnosed cancer in a defined geographic area. For all cancers, the incidence rate is about double the mortality rate. Thus, looking only at mortality would greatly underestimate the true frequency. In addition, incident cases are verified in hospital and clinical records as cancer, and include specific pathologically-confirmed features, including the stage of the cancer, a measure of how advanced it is. This information is not available from death certificates. This report presents incidence data for the Savannah River region.

The major reason for starting the registry was to develop a mechanism to assess cancer occurrence in the region. On the other hand, the specific aims of this report, aside from a desire to inform the above-noted constituencies and readership, are of a statistical nature:

• to reinforce that the 22 county region has a shared interest in adverse health outcomes.
• to demonstrate the ability to collect and evaluate cancer incidence data in this unified geographic area, which includes counties in two states.
• to evaluate the completeness of reporting cases, and discuss barriers to completeness in case finding.
• to provide statistical tabulations of cancer data found in the SRRHIS area, Atlanta, and in the US overall.
• to inform the public about the occurrence of cancer in the area, and to make them aware of the role of population-based registries in evaluating their risks of getting cancer.

These individual issues will be addressed in a later section of this report.

Registry Background

The Savannah River Region Health Information System (SRRHIS) is a joint program of the Medical University of South Carolina, (College of Medicine, Department of Biometry and Epidemiology), and Emory University (Rollins School of Public Health, Department of Epidemiology). Its goal is to collect information about the occurrence of cancers in residents of counties adjacent to and downstream from the Savannah River and the Savannah River Site (SRS), and to report the findings to the residents in a timely and understandable fashion.

There had been attempts to inform the public about studies of cancer mortality in residents (Sauer, 1978) and workers (Cragle, 1988). The SRS also was holding town meetings to inform citizens about health risks and to answer their questions. However, citizens remained skeptical about the answers they were receiving. A summary of the findings of three mortality studies, the two noted previously plus another more recent one, is given in Appendix D.

It was this continuing concern which led researchers from the two universities to seek support for a population based registry, to be used to evaluate risk and to inform the community. To this end faculty members of the Medical University of South Carolina and Emory University met in Charleston in the fall of 1989 to begin planning the project. During the ensuing months they made five key decisions:

• The cancer registry would be compatible with the Nation's official registry, the "Surveillance Epidemiology and End Results" (SEER) program of the National Cancer Institute (a component of which is located at Emory University);
• Dissemination of the collected information to the residents would be given highest priority;
• To assist in gaining and keeping the public trust, several policies were adopted: openness would characterize every aspect of the operations; data quality controls would be scrupulous; and an advisory committee of professional and lay people would be constituted so as to reflect the different geographic, race, gender, and age groups in the SRRHIS region;
• The target region for the project would be those counties in Georgia and South Carolina adjacent to SRS and those downstream to the ocean.

The grant application was funded by DOE in April 1991, and cancer case collections began as of January 1, 1991.

A cancer reporting law, existed in Georgia prior to 1991; however, South Carolina did not have a law. In South Carolina, therefore, the goodwill of the hospitals, clinics, and laboratories made the registry possible. Considerable effort is made to feed back information and gratitude to these institutions for their invaluable voluntary assistance.

In 1996, South Carolina passed a cancer reporting law, to assist the Department of Health and Environmental Control in its effort to develop a statewide cancer registry. This new state effort is being undertaken with support from the Centers for Disease Control and Prevention (CDC), in full partnership and coordination with SRRHIS. DHEC-SRRHIS collaboration is a model program in that it serves the interests of both registries' activities equally, and avoids any duplication of effort.

While it was expected that many cancer cases of South Carolinians would be diagnosed in Georgia institutions, the large number observed came as a surprise. More than a third of cancers diagnosed in South Carolina SRRHIS registry residents were reported from Georgia institutions. The decision to make the registry a two-state project certainly was validated.

Priority Goal of SRRHIS

The primary goal of SRRHIS is to determine over time whether there is an excess of adverse health outcomes (such as cancer) in the region. Within this overall goal a continuing priority of the project is to inform the residents in the SRRHIS area. To this end a number of approaches have been tried. At first, the "community meeting" format held in Beaufort in 1991, worked modestly well. However, four others in 1992 were less successful. That model was laid aside to concentrate on more direct ways of addressing the citizens. Speeches given at meetings of civic clubs seemed more appealing to listeners. Of the 12 clubs addressed, from Aiken and Augusta to Beaufort and Savannah, and to Sylvania and Ridgeland, the response has been uniformly good. The audience questions have indicated a high level of interest, and word proceeded to other clubs which in turn called for a speaker. Television and radio interviews have been completed on six occasions. Judging by calls received by the interviewers, and by calls received at SRRHIS, the message about the existence of the registry and the program's aim to inform the public is spreading. These talk-to approaches are complemented by a quarterly newsletter with a mailing list of 3,800 names (including practicing physicians in the SRRHIS area). Anyone wishing to receive the newsletter is urged to contact the SRRHIS office.

Findings

This report indicates that, overall, the rates of cancer in the SRRHIS region are lower than metropolitan Atlanta and national U.S. SEER rates. This deficit is about 20% for males and black females, and 12% for white females. This is consistent with the lower rates seen in the rural Georgia SEER program and in other non-urban populations. Quality control studies for this period indicated that case identification and data quality were both quite good, so that the conclusion of lower rates in this area is supported. The overall rates of cancer also were quite consistent between Georgia and South Carolina.

The pattern was not uniform by cancer site, however. There were several cancers for which the rates in the SRRHIS area were higher than in the comparison populations. The greatest excess was for invasive cervical cancer in black women, which was elevated by over 40% compared to national or Atlanta rates. This indicates a clear need for intensified Pap smear screening in this population. The rates of uterine corpus cancer among black SRRHIS women also exceeded the national and Atlanta rates, but only by 7%.

The other cancers for which the SRRHIS rates matched or exceeded those of the comparison groups tend to have smoking or diet as risk factors. In white males, cancers of the esophagus and lung, both smoking related diseases, exceeded rates of metropolitan Atlanta. Lung cancer rates among black males were about the same as among whites, but somewhat lower than those of the Atlanta or national rates for blacks. The esophageal cancer rates for black males in the South Carolina SRRHIS region were especially high, exceeding the national or Atlanta rates by about 14%. In contrast, the Georgia rates were 16% lower than those of the comparison groups. Cancers of the oral cavity and pharynx were similar in the SRRHIS residents to national rates for women and white men. Stomach cancer rates in white men and black women exceeded the national rates, while remaining below the much higher Atlanta rates.

Especially clear is the confirmation that in this population, as in others, there is a racial disparity in cancer rates - black males in the SRRHIS area have an overall rate of cancer that is 14% greater than that of whites. These include cancers for which smoking and/or alcohol consumption probably play roles, such as cancer of the oral cavity, esophagus, lung and other respiratory system. In addition, blacks experience stomach cancer and multiple myeloma more often, and are diagnosed with prostate cancer, the most common cancer among men, at a rate 1.6 times that of whites. White males, however, experience skin cancers including melanoma, lymphomas and cancers of the urinary system (bladder, kidney, other) at a higher rate than do blacks.

In contrast to the experience of males, black women have much lower cancer rates than do whites, primarily because of considerably higher breast, melanoma and lung cancer rates among white female SRRHIS region residents. White women also experience higher rates of ovarian, bladder and brain tumors. Black women are far more likely to be diagnosed with cervical cancer, and with cancer of the oral cavity, pancreas or stomach.

Both black males and females are more likely to be diagnosed with their cancers at a late stage then are whites, and this is the greatest correlate with the poorer five year relative survival that black cancer patients experience.

Listing the most common cancers occurring in a population gives a quick overall picture of the cancers seen most frequently.* In the SRRHIS area, prostate cancer is the most common in men, and prostate and lung cancers together make up more than half of all cancers seen in men. Prostate cancers account for 37 percent of all cancers among black men, but only 27 percent among white men. Other frequent cancers for males include colon and rectum cancers, bladder cancers and cancers of the oral cavity. For black males, stomach cancers are more common than bladder tumors. This distribution is similar to that seen in the entire SEER population over a similar time period, with the exception of oral cavity cancers being more common in the SRRHIS region, and lymphomas more common nationwide.

For women, breast cancer is the most common cancer occurring in the SRRHIS population, making up almost 32 percent of all cancers reported. Colon and rectal cancers are the second most frequent overall, but lung cancer is second among white women. Cancers of the uterine corpus and ovary are fourth and fifth for all women in the region and among the SEER female population overall. However, cancer of the uterine cervix is the fourth most common among black women in the SRRHIS area.

While these rankings are of interest in illustrating the major cancer sites in the population, they are influenced by a number of factors, including differing age distributions of the population. More understanding of the SRRHIS area burden by race and gender can be gained by examining the data in the site specific sections of this report. Age-adjusted incidence rates and the role of risk factors help to elucidate specific cancer control issues that can then be addressed with public health programs.

* This discussion refers to invasive cancers only, except for bladder cancer which includes both in situ and invasive tumors in accordance with NCI guidelines.

References

• Breslow NE, Day NE. Statistical Methods in Cancer Research Volume II - the Design and Analysis of Cohort Studies. IARC Scientific pub. No. 82. Lyon, France, 1987.
• Cragle, Donna, American Journal of Industrial Medicine, Mortality Among Workers at a Nuclear Fuels Production Facility, Vol. 14, pgs. 379-401, 1988.
• Cragle, Donna, Press Conference at SRS March 20, 1995
• Jablon, Seymour, Cancer in Populations Living Near Nuclear Facilities, Three Volumes, National Cancer Institute, Bethesda, MD, July 1990.
• Kosary CL, Reis LAG, Miller BA, Hankey BF, Harras A, Edwards BK (eds). SEER Cancer Statistics Review, 1973-1992: Tables and Graphs, National Cancer Institute. NIH Pub. No. 96-2789. Bethesda, MD, 1995.
• Miller BA, Reis LAG, Hankey BF, Kosary CL, Harras A, Devesa SS, Edwards BK (eds). SEER Cancer Statistics Review: 1973-1990, National Cancer Institute. NIH Pub. No. 93-2789, 1993.
• Parker SL, Tong T, Bolden S, Wingo PA. Cancer statistics, 1996. CA 46:5-27, 1996.
• Sauer, Herbert I., Mortality Trends in Counties Near the Savannah River Plant, 1949-1978. Herb Sauer Associates, July 1984, 112 pp. (Prepared under contract with E. I. du Pont de Nemours and Co., Inc.)
• Seiffert JE, ed. Standards for Completeness, Quality, Analysis, and Management of Data. Standards for Cancer Registries Vol. III. American Association of Central Cancer Registries, 1994.
• SRS Environmental Report for 1991, Summary Pamphlet. Westinghouse Savannah River Company, Savannah River Site, Report Number WSRC-TR-92-186. 1992.

  Contents

  Acknowledgments

  Summary

  Introduction

  Findings

  Common_Site

  Cancer_Incidence

  Appendices

  Home