Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “SEER Program”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 469 records · Page 26Linked to original sources
PubMed1999

Comparison of methods for classifying Hispanic ethnicity in a population-based cancer registry.

The accuracy of ethnic classification can substantially affect ethnic-specific cancer statistics. In the Greater Bay Area Cancer Registry, which is part of the Surveillance, Epidemiology, and End Results (SEER) Program and of the statewide California Cancer Registry, Hispanic ethnicity is determined by medical record review and by matching to surname lists. This study compared these classification methods with self-report. Ethnic self-identification was obtained by surveying 1,154 area residents aged 20-89 years who were diagnosed with cancer in 1990 and were reported to the registry as being Hispanic or White non-Hispanic. Predictive value positive, sensitivity, and relative bias were used to assess the accuracy of Hispanic classification by medical record and surname. Among those persons classified as Hispanic by either or both of these sources, only two-thirds agreed (predictive value positive = 66%), and many self-identified Hispanics were classified incorrectly (sensitivity = 68%). Classification based on either medical record or surname alone had a lower sensitivity (59% and 61%, respectively) but a higher predictive value positive (77% and 70%, respectively). Ethnic classification by medical record alone resulted in an underestimate of Hispanic cancer cases and incidence rates. Bias was reduced when medical records and surnames were used together to classify cancer cases as Hispanic.

California↗
PubMed1980

Epidemiologic relationships of the bovine population and human leukemia in Iowa.

Data from the Third National Cancer Survey and data collected under the auspices of the National Cancer Institute's Surveillance, Epidemiologic and End Results (SEER) Program were used to perform a descriptive epidemiologic study of leukemia in Iowa. Data were also collected on livestock population patterns in the state, and a survey was performed to locate cases of lymphosarcoma within the cattle population. Ecologic relationships between human leukemia, livestock populations and bovine lymphosarcoma were investigated. Iowa has higher rates than the national average for human leukemias. The lymphoid leukemias make up the majority of the excess leukemias. Excessive acute lymphoid leukemia (ALL) is seen in the ages under 20 years and over 60 years. The excess ALL is in males living in rural counties. There is no sex difference for chronic lymphoid leukemia (CLL), but there is a slight urban-rural differential. There is a high positive correlation between ALL in males and cattle density. This relationship is greater for daily cattle than for beef cattle. There is an additional positive relationship between counties with excessive ALL and the presence of dairy herds affected with bovine lymphosarcoma.

Adolescent↗
PubMed1983

Methods of cancer case selection: implications for research.

The race, age, and sex differences between clinically and microscopically diagnosed cancer cases for a total of 63,301 incident cases for 18 sites are reported. The cases, diagnosed during 1973-1978, were drawn from the Cancer Surveillance Section in the Division of Epidemiology at the Michigan Cancer Foundation in Detroit, a participant in the Surveillance, Epidemiology, and End Results (SEER) program. Although microscopically diagnosed cases and those diagnosed by other methods differed by race and sex for a few sites, the two case groups differed by age for 15 of the 18 sites examined. Cases without microscopic diagnoses (i.e., clinical diagnoses) tended to be older than microscopically confirmed cases. These findings indicate that in studies of certain cancer sites, the method of case selection could affect results pertaining to age differences in exposure to and effects of relevant risk factors.

Adolescent↗
PubMed1984

Assessing the validity of observed cancer incidence trends.

The available cancer incidence data from the National Cancer Institute surveys, the Surveillance, Epidemiology, and End Results (SEER) Program, and the Connecticut Tumor Registry are examined for the period from the late 1940s to the present. Two indicators of completeness and accuracy of reporting (percentage of cases reported only by death certificate and percentage of cases with histologic confirmation of the diagnosis) show considerable variation, particularly in the past, but no consistent strong patterns indicating noncomparability of the data are apparent. Incidence data for five geographic areas (Atlanta, Detroit, San Francisco-Oakland, Iowa, and Connecticut) are examined for several primary sites by area and age, and are compared with mortality data for the same areas and the entire country. A variety of patterns in the trends are apparent. For several sites, the incidence trends are similar to what would be predicted based on the mortality patterns, but differences are apparent for other sites. Mortality trends may be influenced by shifts in diagnostic specificity on the death certificates and by changes in survival rates, and both incidence and mortality rates may be influenced by increased casefinding, improvements in diagnostic procedures, expansion of the medical care delivery system, and real changes in the prevalence of risk factors. Therefore, both incidence and mortality data should be used when attempting to assess the real trends in cancer occurrence because sole reliance on one or the other may lead to erroneous conclusions.

Adult↗
PubMed1986

Choriocarcinoma incidence in the United States.

To determine the incidence of choriocarcinoma in the United States, the authors examined data from the Surveillance, Epidemiology, and End Results (SEER) Program, 1973-1982, the National Center for Health Statistics, and the Centers for Disease Control. They calculated an overall incidence of one per 24,096 pregnancies or one per 19,920 livebirths. Among whites, who comprised the majority of cases, the incidence was estimated to be one per 22,623 livebirths. After adjustment for age and birth effects, whites demonstrated the lowest rates, with blacks and other races at 2.1- and 1.8 fold elevated risk, respectively. The incidence among whites increased dramatically with maternal age, with women aged 40-44 years at an 8.6-fold excess risk compared with those aged 20-24 years. In addition, teenagers who gave birth were at higher risk. Possible reasons for these incidence patterns are discussed.

Adolescent↗
PubMed1987

Seasonal variation in the incidence of cutaneous malignant melanoma: an analysis by body site and histologic type.

A seasonal analysis of the incidence of cutaneous malignant melanoma was conducted to investigate the degree to which the increased frequency of cases diagnosed in the summer may be due to an artifact of increased ascertainment. The monthly incidence of 13,151 microscopically-confirmed cases of cutaneous malignant melanoma diagnosed between 1973 and 1982 among white residents of nine regions in the United States served by population-based cancer registries participating in the Surveillance, Epidemiology, and End Results (SEER) program was evaluated for the presence of seasonal trends. Except for nodular melanoma, there was substantial seasonal variation in the incidence of all histologic types. For the two most common categories, superficial spreading melanoma and melanoma not further classified as to histology (of which the majority are likely to be superficial spreading melanoma), the peaks in incidence at body sites which are likely to experience seasonal changes in clothing (upper limbs, trunk, legs) occurred in the summer and were of similar relative size to the peaks, located in the spring, at sites which are likely to experience little variation in clothing during the year (head, face, and neck). A summer peak was also observed for Hutchinson's melanotic freckle melanoma of the head, face, and neck. The presence of seasonal variation in the incidence of superficial spreading melanoma of the head, face, and neck suggests that increased ascertainment cannot by itself account for seasonal variation in the incidence of superficial spreading melanoma of other parts of the body.

Female↗
PubMed1992

The genetic epidemiology of second primary breast cancer.

It is well established that women with a family history of breast cancer run a higher risk of breast cancer than do women without a family history. The evidence, however, is less clear regarding a possible association between a family history of breast cancer and risk of second primaries. The purpose of this prospective study was to estimate the risk for second primary breast cancer associated with having a family history of breast, endometrial, and ovarian cancers. A cohort of 4,660 women with a first primary breast cancer diagnosed between 1980 and 1982 were interviewed as part of the Cancer and Steroid Hormone Study, a multi-center population-based case-control study, and followed through eight Surveillance, Epidemiology, and End Results (SEER) program registries for 4 to 6 years. Of these women, 136 developed a second primary breast cancer in the contralateral breast at least 6 months after diagnosis of the first primary. Cox proportional hazards modeling techniques were used to model the time to onset of second primary breast cancer while adjusting for multiple predictors. The risk of contralateral breast cancer was elevated among cohort members who reported a history of breast cancer in a first-degree relative (multivariable-adjusted rate ratio (RR) = 1.91, 95% confidence interval (CI) = 1.22-2.99). Early age at onset (< 46 years) in the relative further increased the risk of developing contralateral breast cancer (sister: multivariable-adjusted RR = 3.36, 95% CI 1.62-6.98; mother: multivariable-adjusted RR = 2.35, 95% CI 1.02-5.43). Bilateral breast cancer in mothers was also associated with more than a two and a half-fold increase in risk (multivariable-adjusted RR = 2.55, 95% CI 1.02-6.35). The association between family history of breast cancer and risk of contralateral breast cancer did not vary substantially according to age at onset of the first primary breast cancer. The age-adjusted rate ratio for development of a second primary breast cancer among women with a first-degree relative with endometrial cancer was 2.13 (95% CI 1.04-4.35), while the corresponding rate ratio among women with a family history of ovarian cancer was 1.69 (95% CI 0.42-6.83). There was little evidence that age at onset among the relatives with endometrial or ovarian cancer affected the risk. Some of these findings have not been previously reported and need replication in future studies.

Adult↗
PubMed1993

Parental smoking and risk of childhood brain tumors.

Data from a large, population-based, case-control study were analyzed to assess the role of parental smoking in childhood brain tumors. Parents of 361 cases, newly diagnosed between January 1, 1977 and December 31, 1981 and ascertained from eight Surveillance, Epidemiology, and End Results (SEER) program registries, and 1,083 controls had been interviewed. No significant differences in risks were found to be associated with maternal or paternal smoking at any time (odds ratio (OR) = 0.92 for mothers and 1.06 for fathers), during the year of birth of the child (which included both the prenatal and postnatal periods) (ORs = 0.84 for < 1 pack/day and 1.0 for > or = 1 pack/day for mothers, and 0.68 for < 1 pack/day and 1.07 for > or = 1 pack/day for fathers), or 2 years before the child was born, i.e., the pre-conception period (ORs = 0.75 for < 1 pack/day and 1.01 for > or = 1 pack/day for mothers, and 0.90 for < 1 pack/day and 1.15 for > or = 1 pack/day for fathers). Mothers were also specifically asked if they smoked during the pregnancy, and no association was found compared with never smokers (OR = 1.08, 95% confidence interval (CI) 0.80-1.45) or for ever-smokers who continued to smoke during pregnancy compared with those who stopped smoking during pregnancy (OR = 1.15, 95% CI 0.75-1.78). Finally, no significant increase in risk of brain tumors was found for the child's passive exposure to parental smoking during the period from birth to diagnosis of the brain tumor in the case. The lack of an effect of parental smoking was observed for both the major histologic types and locations of brain tumors. These findings and those from earlier studies provide no support for the hypothesis that parental cigarette smoking influences the risk of brain tumors in children.

Adult↗
PubMed1993

Are positive axillary nodes in breast cancer markers for incurable disease?

OBJECTIVE: This study was designed to determine if resection of positive regional nodes in patients with breast cancer provided a group of cured patients. SUMMARY BACKGROUND DATA: Previous studies of long-term follow-up of patients with breast cancer have demonstrated that 30% of patients with positive nodes may be alive at 10 or 20 years. Randomized, prospective studies have not shown a difference in survival between modified radical and total mastectomy. METHODS: The authors analyzed the statistical basis of randomized, prospective studies of breast cancer. Patients from University and Kings County Hospital with breast cancer were observed (n = 1024) and the 10-year survival curves were analyzed. Data from the Surveillance, Epidemiology, and End Results (SEER) program were evaluated to determine the prognosis of patients with positive-node breast cancer who survived 10 and 15 years. RESULTS: The authors found that no current randomized study of primary treatment of breast cancer has the power to determine the advantage of one treatment over another if the primary cancer is treated. Additionally, evaluation of the patients in this study demonstrated that those with regional disease usually had a recurrence by 7 years; however, at 10 years, 30% are still alive and have the same prognosis as patients with negative-node breast cancer. CONCLUSIONS: Regional breast cancer is a curable disease in some resected patients.

Adult↗
PubMed1997

Prostate cancer incidence and mortality rates among white and black men.

This paper presents prostate cancer incidence and mortality rates in the United States by factors associated with the disease at diagnosis and explores racial differences in these rates. The analysis is based on population-based cancer data from the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute. Incidence rates increased sharply from 1973 to 1992 (more so among whites), with the increase attributed to screening and preferential finding of low-risk, slow-growing tumors (that is, length bias). Mortality rates increased slightly (more so among blacks). The 1988-1992 white and black prostate cancer mortality rates comprise cases diagnosed as early as 1973. These rates remained fairly constant for each stage category at diagnosis, except for black men diagnosed with localized or unstaged disease, in whom the rates increased. Of the white prostate cancer cases, 33% are diagnosed with localized disease, 16% are diagnosed with regional disease, and 38% are diagnosed with distant disease. The corresponding percentage for black patients are 31, 12, and 45%. Prostate cancer mortality rates for blacks diagnosed with localized, regional, distant, and unstaged disease are 1.9, 1.5, 2.4, and 2.0 times those of whites. The time from diagnosis to death depends on the stage and age at diagnosis, with the time somewhat longer for whites than blacks. This longer time period is due, in part, to better staging and length bias. The effect of therapy on prostate cancer mortality rates is unclear.

Adult↗
PubMed1988

The probability that a specific cancer and a specified radiation exposure are causally related.

It is fundamental that a given cancer case cannot be attributed with absolute certainty to a prior ionizing radiation exposure, whatever the level of exposure. It is possible to estimate the probability of a causal relationship based on data and models that have been inferred from group statistics. Two types of information are needed to make these probability calculations: natural cancer incidence rates and risks of cancer induction from ionizing radiation. Cancer incidence rates for the United States are available in the report of the Surveillance, Epidemiology and End Results (SEER) program of the National Cancer Institute. Estimates of the risk of cancer induction from ionizing radiation have been published by the Advisory Committee on the Biological Effects of Ionizing Radiation (BEIR) of the National Academy of Sciences, the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR), and the International Commission on Radiological Protection (ICRP). Using the parameters discussed above, the probability of causation formulation estimates the probability that a person who develops a particular cancer after a known quantifiable radiation exposure has the cancer as a result of the exposure. In 1985, the National Institutes of Health, responding to a U.S. Congressional mandate, published radioepidemiologic tables using the probability-of-causation method.

Humans↗
PubMed2005

Comorbidity as a predictor of stage of illness for patients with breast cancer.

OBJECTIVE: The purpose of this research was to determine whether comorbidity affects the stage at which breast cancer is diagnosed. METHODS: Data from the Surveillance, Epidemiology and End Results (SEER) program of the National Cancer Institute (NCI) was merged with Medicare claims for 17,468 women diagnosed with breast cancer from 1993 to 1995. RESULTS: Women with cardiovascular disease, musculoskeletal disorders, mild-to-moderate gastrointestinal disease, and nonmalignant benign breast disease had a 13%, 7%, 14%, and 24% lower odds, respectively, of being diagnosed with advanced breast cancer. Women with diabetes, other endocrine disorders, psychiatric disorders, or hematologic disorders increased the odds of a late-stage diagnosis by 19%, 11%, 20%, and 19% respectively. Mammography screening and contact with the medical care system decreased the odds of late-stage diagnosis. DISCUSSION: Four hypotheses are suggested to explain this link between comorbid illness and stage at diagnosis: (1) the "surveillance" hypothesis, (2) the "physiological" hypothesis, (3) the "competing demand" hypothesis, and (4) the "death from other causes" hypothesis. CONCLUSIONS: Comorbidity may complicate the diagnostic decision-making process for breast cancer. The results suggest that contact with the medical care system improves the odds of early-stage diagnosis. Thus, barriers to access for people with chronic conditions may exacerbate those chronic conditions and increase the odds of late-stage breast cancer.

Breast Neoplasms↗
PubMed2000

Breast cancer stage at diagnosis and survival among patients with prior breast implants.

Longstanding concern exists regarding the potential for women with breast implants to experience delayed detection of breast cancer. Furthermore, survival among cosmetic breast implant patients who subsequently develop breast cancer is a concern. Since 1976, this institution has monitored cancer incidence in a cohort of 3182 women who underwent cosmetic breast augmentation between 1959 and 1981. The distributions of stage at diagnosis and survival of the 37 women who subsequently developed in situ or invasive breast cancer were compared with the observed population distributions. The distribution of stage at diagnosis for cosmetic breast implant patients who subsequently developed breast cancer was virtually identical to that of all breast cancer patients in Los Angeles County who were of the same age and race, and were diagnosed during the same time period. Furthermore, the 5-year survival rate of the 37 patients did not differ from that which would be expected based on rates established by the U.S. National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) program. These results suggest that cosmetic breast implant patients are not at increased risk of delayed detection of breast cancer, nor do they suffer a poorer prognosis when breast cancer does occur. Although the number of breast cancer patients in this study is small, the results are highly consistent with the existing epidemiologic evidence related to breast cancer detection and survival among breast implant patients. Although breast implant patients should continue appropriate breast cancer screening behavior, there seems to be no cause for alarm.

Adult↗
PubMed2001

The systematic relationship between melanomas diagnosed in situ and when invasive.

The Surveillance, Epidemiology and End Results (SEER) Program of the US National Cancer Institute has provided population-based, histologically confirmed incidence data for melanomas restricted to the epidermis (in situ) and invasive melanomas for representative areas of the US for the years 1975 to 1997. The incidence of in situ melanomas has increased more rapidly than that for invasive melanoma, but both appear to conform to the birth-cohort pattern. As both changed from year to year, the incidence of melanoma in situ and invasive melanoma at the time of diagnosis were linked by simple proportional relationships. The ratio of in situ to invasive tumours was similar in males and females, suggesting that the generally better female prognosis develops once the tumour has spread into the dermis. Across the age groups, the patients with in situ tumours were older than those with invasive tumours. This suggests slower growth, while histological confirmation of the diagnoses and the precise linkage between the incidence rates for in situ and invasive melanomas implies a close biological connection between tumours that progress and those that do not. Study of the in situ tumours could reveal what was holding them back.

Adolescent↗
PubMed2006

Incidence rates of acute promyelocytic leukemia among Hispanics, blacks, Asians, and non-Hispanic whites in the United States.

Despite significant improvements in the prognosis of acute promyelocytic leukemia brought about by therapeutic advances, understanding of the epidemiology of acute promyelocytic leukemia remains limited. Earlier reports have suggested that Hispanics may have an increased incidence of acute promyelocytic leukemia, but no systematic analysis of national data has yet been reported. We performed a retrospective cohort study, using data from the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute from 1992-2001 in order to compare leukemia incidence rates as a function of race and ethnicity. We identified 709 cases of acute promyelocytic leukemia and analyzed incidence rates by race and sex. Hispanics were not found to have greater lifetime incidence rates than whites, with an incidence relative rate (IRR) of 0.86 that of whites (P=0.17). The age distribution among Hispanics was significantly different from non-Hispanic whites, with greater incidence rates for children ages 1-19 years (IRR=1.9, P=0.02) and adult ages 20-44 years (IRR=1.6, P=0.004). Blacks had lower lifetime incidence rates than non-Hispanic whites (IRR=0.75, P=0.04), Hispanics (IRR=0.64, P=0.007), and Asians (IRR=0.67, P=0.03). Asians did not differ from non-Hispanic whites in lifetime or age-specific incidence rates. These results indicate that while US Hispanics do not have greater lifetime incidence rates of acute promyelocytic leukemia, blacks have lower incidence rates of acute promyelocytic leukemia than Hispanics, non-Hispanic whites, and Asians.

Adolescent↗
PubMed2002

Invasive papillary carcinomas of the extrahepatic bile ducts: a clinicopathologic and immunohistochemical study of 13 cases.

Carcinomas of the extrahepatic bile ducts are uncommon neoplasms that are morphologically heterogeneous and associated with a poor prognosis. We have previously shown that the noninvasive and minimally invasive papillary carcinomas of the extrahepatic bile ducts behave as in situ carcinomas and are associated with a better prognosis. We reviewed the clinical records of 13 patients with invasive papillary carcinomas of the extrahepatic bile ducts and analyzed the microscopic features and selected immunohistochemical reactivity (p53, Mib-1, and Dpc4) that might correlate with patient survival. In addition, we present the updated SEER (Surveillance, Epidemiology, and End Results) data of the National Cancer Institute for the invasive extrahepatic bile duct carcinomas compiled from 1975 to 1998. The 13 patients with papillary carcinoma had a male to female ratio of 1:1, and their ages ranged from 33 to 89 years. Painless jaundice and abdominal pain were the most common complaints. Five tumors were located in the distal portion, one in the mid portion, and six in the proximal portion of the common bile duct. One papillary carcinoma arose in the right hepatic duct. The Whipple procedure was performed in six patients, common bile duct resection in six, and right hepatic lobectomy in one. The cell phenotype of the papillary carcinomas was biliary in nine and intestinal in three. One tumor had both biliary and intestinal phenotypes. Four tumors dedifferentiated (two to undifferentiated small cell carcinomas, one to small [oat] cell carcinoma, and one to giant cell carcinoma). Two papillary carcinomas extended into the pancreas and three into the liver. Only one patient had lymph node metastases at presentation. Follow-up was available in 10 patients. Six patients died of disease from 2 weeks to 2 years and 1 month after surgery. Four patients are alive with no evidence of disease from 4 months to 8 years and 8 months after surgery. Of 174 invasive papillary carcinomas compiled by the SEER program, 71 were confined to the ductal wall, and 61 had regional lymph node metastases. Papillary carcinomas confined to the ductal wall have better 10-year relative survival rates than adenocarcinomas limited to the wall (21% versus 12%). Likewise papillary carcinomas with lymph node metastasis have better prognosis than adenocarcinoma with nodal metastases (10-y survival rate of 12% versus 5%). Currently, the histologic type and the stage of the disease are the most important prognostic factors in these papillary carcinomas. Separation of invasive and noninvasive or minimally invasive papillary carcinoma is critical in estimating the patient outcome. Our findings suggest that there is no correlation between p53, Ki-67, and Dpc4 expression in these tumors and survival of the patients.

Adult↗
PubMed2003

Outcomes of colorectal cancer in the United States: no change in survival (1986-1997).

OBJECTIVES: The past decade has witnessed important advances in colorectal cancer (CRC) screening and treatment. To assess the potential impact of these advances on CRC survival and incidence, we evaluated recent temporal trends in the United States. METHODS: Using the nine population-based cancer registries that constitute the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute, we identified primary CRCs diagnosed between 1986 and 1997. Temporal changes were evaluated for the periods 1986-1988, 1989-1991, 1992-1994, and 1995-1997. Age-adjusted incidence and relative 1-, 3-, and 5-yr survival were calculated, along with 95% CI. RESULTS: We identified 144,284 individuals with CRC during 1986-1997, 51.5% of whom were men. In individuals of white ethnicity, the age-adjusted incidence rates for rectosigmoid and left CRC fell over time. In addition, in white men (but not in white women), the age-adjusted incidence rates for right CRC declined. In individuals of African American/black ethnicity, the age-adjusted incidence rates for rectosigmoid and right CRC showed no significant change over time. For all sites, no significant differences in survival were seen over time in whites or blacks. CONCLUSIONS: The decrease in age-adjusted incidence rates over time is consistent with a decrease in CRC burden, especially among whites. However, the lack of improvement in survival over time in whites and blacks indicates that despite advances in screening and treatment, to date, no survival benefit has occurred.

Black or African American↗
PubMed2003

Survival of colorectal cancer patients hospitalized in the Veterans Affairs Health Care System.

OBJECTIVES: A recent analysis based on data from the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute indicates that no survival benefit occurred, for white or for black individuals, in colorectal cancer diagnosed during 1986-1997, and that blacks fared worse than whites. The objective of this research was to evaluate recent temporal trends in the survival of patients with colorectal cancer admitted to hospitals in the Veterans Affairs (VA) system, which offers equal access to care and facilitates systemwide implementation of prevention and treatment services. METHODS: This research was a cohort study of patients admitted to all VA hospitals with a new diagnosis of colorectal cancer between October, 1987, and September, 1998, and followed through September, 2001. Temporal changes in observed 5-yr survival were evaluated for the periods 1987-1989, 1990-1992, 1993-1995, and 1996-1998. Cumulative survival was obtained from Kaplan-Meier estimates, whereas adjusted risk of death was calculated using a Cox proportional hazards model. Adjustment was made for differences in patient characteristics, including comorbidity. RESULTS: We identified 46,044 individuals with colorectal cancer in VA hospitals during 1987-1998, 98.5% of whom were men. The mean age was 67.7 yr, and the two largest racial/ethnic groups were whites (76.5%) and blacks (17.1%). Significant differences in survival were seen over time (p < 0.001, log rank test) with longer survival in patients diagnosed in the more recent time periods. In the multivariable Cox model, survival showed an 18% increase over time (1987-1998) after adjusting for differences in age, race, comorbidity, cancer site, and extent of disease. There was a small but statistically significant decrease in chance of survival in blacks compared with whites (adjusted relative survival 0.96, 95% CI = 0.92-0.99). CONCLUSIONS: Recent non-VA studies have shown stable survival for colorectal cancer patients over time, coupled with significantly decreased survival for blacks compared with whites. In contrast, in the VA system, survival has improved for both white and black patients; in addition, the racial discrepancy in survival is markedly attenuated. These results suggest that the benefits of prevention and treatment advances may be more readily achieved in the VA's equal access, integrated health care system.

Aged↗