Risk factors for becoming homeless among a cohort of veterans who served in the era of the Iraq and Afghanistan conflicts.

OBJECTIVES: In this population-based cohort study, we assessed baseline risk factors for homelessness, including the role of service in the Iraq or Afghanistan conflicts, among a large cohort of recent veterans. METHODS: Data for this study came from administrative records for 310,685 veterans who separated from active military duty from July 1, 2005, to September 30, 2006. We used survival analysis methods to determine incidence rates and risk factors for homelessness, based on baseline data for military factors, demographic characteristics, and diagnoses of behavioral health disorders and traumatic brain injury. RESULTS: Service in Iraq or Afghanistan and, more specifically, posttraumatic stress disorder among veterans deployed there, were significant risk factors of modest magnitude for homelessness, and socioeconomic and behavioral health factors provided stronger indicators of risk. Gender was not a significant indicator of differential risk. CONCLUSIONS: Although service in Iraq and Afghanistan was significant, socioeconomic and behavioral health indicators show more promise in efforts to use administrative data to inform prevention efforts by identifying veterans who are at elevated risk for becoming homeless upon their return to civilian life.

Impact of socioeconomic status on cancer incidence and stage at diagnosis: selected findings from the surveillance, epidemiology, and end results: National Longitudinal Mortality Study.

BACKGROUND: Population-based cancer registry data from the Surveillance, Epidemiology, and End Results (SEER) Program at the National Cancer Institute (NCI) are mainly based on medical records and administrative information. Individual-level socioeconomic data are not routinely reported by cancer registries in the United States because they are not available in patient hospital records. The U.S. representative National Longitudinal Mortality Study (NLMS) data provide self-reported, detailed demographic and socioeconomic data from the Social and Economic Supplement to the Census Bureau's Current Population Survey (CPS). In 1999, the NCI initiated the SEER-NLMS study, linking the population-based SEER cancer registry data to NLMS data. The SEER-NLMS data provide a new unique research resource that is valuable for health disparity research on cancer burden. We describe the design, methods, and limitations of this data set. We also present findings on cancer-related health disparities according to individual-level socioeconomic status (SES) and demographic characteristics for all cancers combined and for cancers of the lung, breast, prostate, cervix, and melanoma. METHODS: Records of cancer patients diagnosed in 1973-2001 when residing 1 of 11 SEER registries were linked with 26 NLMS cohorts. The total number of SEER matched cancer patients that were also members of an NLMS cohort was 26,844. Of these 26,844 matched patients, 11,464 were included in the incidence analyses and 15,357 in the late-stage diagnosis analyses. Matched patients (used in the incidence analyses) and unmatched patients were compared by age group, sex, race, ethnicity, residence area, year of diagnosis, and cancer anatomic site. Cohort-based age-adjusted cancer incidence rates were computed. The impact of socioeconomic status on cancer incidence and stage of diagnosis was evaluated. RESULTS: Men and women with less than a high school education had elevated lung cancer rate ratios of 3.01 and 2.02, respectively, relative to their college educated counterparts. Those with family annual incomes less than $12,500 had incidence rates that were more than 1.7 times the lung cancer incidence rate of those with incomes $50,000 or higher. Lower income was also associated with a statistically significantly increased risk of distant-stage breast cancer among women and distant-stage prostate cancer among men. CONCLUSIONS: Socioeconomic patterns in incidence varied for specific cancers, while such patterns for stage were generally consistent across cancers, with late-stage diagnoses being associated with lower SES. These findings illustrate the potential for analyzing disparities in cancer outcomes according to a variety of individual-level socioeconomic, demographic, and health care characteristics, as well as by area measures available in the linked database.

Estimating average annual per cent change in trend analysis.

Trends in incidence or mortality rates over a specified time interval are usually described by the conventional annual per cent change (cAPC), under the assumption of a constant rate of change. When this assumption does not hold over the entire time interval, the trend may be characterized using the annual per cent changes from segmented analysis (sAPCs). This approach assumes that the change in rates is constant over each time partition defined by the transition points, but varies among different time partitions. Different groups (e.g. racial subgroups), however, may have different transition points and thus different time partitions over which they have constant rates of change, making comparison of sAPCs problematic across groups over a common time interval of interest (e.g. the past 10 years). We propose a new measure, the average annual per cent change (AAPC), which uses sAPCs to summarize and compare trends for a specific time period. The advantage of the proposed AAPC is that it takes into account the trend transitions, whereas cAPC does not and can lead to erroneous conclusions. In addition, when the trend is constant over the entire time interval of interest, the AAPC has the advantage of reducing to both cAPC and sAPC. Moreover, because the estimated AAPC is based on the segmented analysis over the entire data series, any selected subinterval within a single time partition will yield the same AAPC estimate--that is it will be equal to the estimated sAPC for that time partition. The cAPC, however, is re-estimated using data only from that selected subinterval; thus, its estimate may be sensitive to the subinterval selected. The AAPC estimation has been incorporated into the segmented regression (free) software Joinpoint, which is used by many registries throughout the world for characterizing trends in cancer rates.

Treatment and survival in a population-based sample of patients diagnosed with gastroesophageal adenocarcinoma.

AIM: To examine the extent of use of specific therapies in clinical practice, and their relationship to therapies validated in clinical trials. METHODS: The US National Cancer Institutes' Patterns of Care study was used to examine therapies and survival of patients diagnosed in 2001 with histologically-confirmed gastroesophageal adenocarcinoma (n = 1356). The study re-abstracted data and verified therapy with treating physicians for a population-based stratified random sample. RESULTS: Approximately 62% of patients had stomach adenocarcinoma (SAC), while 22% had gastric-cardia adenocarcinoma (GCA), and 16% lower esophageal adenocarcinoma (EAC). Stage IV/unstaged esophageal cancer patients were most likely and stage I-III stomach cancer patients least likely to receive chemotherapy as all or part of their therapy; gastric-cardia patients received chemotherapy at a rate between these two. In multivariable analysis by anatomic site, patients 70 years and older were significantly less likely than younger patients to receive chemotherapy alone or chemoradiation for all three anatomic sites. Among esophageal and stomach cancer patients, receipt of chemotherapy was associated with lower mortality; but no association was found among gastric-cardia patients. CONCLUSION: This study highlights the relatively low use of clinical trials-validated anti-cancer therapies in community practice. Use of chemotherapy-based treatment was associated with lower mortality, dependent on anatomic site. Findings suggest that physicians treat lower esophageal and SAC as two distinct entities, while gastric-cardia patients receive a mix of the treatment strategies employed for the two other sites.

Patterns of care for cervical cancer patients treated in the U.S. military health care system.

OBJECTIVE: To determine whether women diagnosed with cervical cancer within the U.S. Department of Defense (DOD) Military Health Care System received treatment within standard guidelines and whether survival was influenced by implementation of these guidelines. STUDY DESIGN: We identified 621 women treated from 1994 to 2002. Guideline therapy was defined as hysterectomy or radiation for FIGO stage < or = IB, and hysterectomy or chemoradiation for advanced localized FIGO stage > IB. Survival analysis was performed using Cox Proportional Hazards models. RESULTS: Of those 621 women, 25% received no surgery, while 41% received radiation therapy. 6% of all patients received chemoradiation therapy prior to the Clinical Announcement received versus 26% of patients after the announcement. Variables associated with significant increased risk of death were advanced age, advanced stage and poorly differentiated lesions, p < 0.01. The lack of guideline therapy was associated with an increased risk of death, p < 0.005. Those patients who did not receive guideline therapy were twice as likely to die (HR 2.11, p = 0.005). CONCLUSION: Our study suggests that chemoradiation was rapidly introduced in the DOD care system after the 1999 National Cancer Institute's Clinical Announcement. However, some patients did not receive guideline therapy.

Insurance status and the use of guideline therapy in the treatment of selected cancers.

PURPOSE: This study estimates the impact of type of insurance coverage on the receipt of guideline therapy in a population-based sample of cancer patients treated in the community. PATIENTS AND METHODS: Patients (n = 7,134) from the National Cancer Institute's Patterns of Care studies who were newly diagnosed with 11 different types of cancer were analyzed. The definition of guideline therapy was based on the National Comprehensive Cancer Network treatment recommendations. Insurance status was categorized as a mutually exclusive hierarchical variable (no insurance, any private insurance, any Medicaid, Medicare only, and all other). Multivariate analyses were used to examine the association between insurance and receipt of guideline therapy. RESULTS: Adjusting for clinical and nonclinical variables, insurance status was a modest, although statistically significant, determinant of receipt of guideline therapy, with 65% of the privately insured patients receiving recommended therapy compared with 60% of patients with Medicaid. Seventy percent of the uninsured patients received guideline therapy, which was nonsignificantly different compared with private insurance. When stratified by race, insurance was a statistically significant predictor of the receipt of guideline therapy only for non-Hispanic blacks. CONCLUSION: Overall, levels of guideline treatment were lower than expected and particularly low for patients with Medicaid or Medicare only. The use of guideline therapy for ovarian and cervical cancer patients and for patients with rectal cancers was unrelated to type of insurance. Of particular concern is the significantly lower use of guideline therapy for non-Hispanic black patients with Medicaid. After adjusting for other factors, only half of these patients received guideline therapy.

Efficient interval estimation for age-adjusted cancer rates.

The age-adjusted cancer rates are defined as the weighted average of the age-specific cancer rates, where the weights are positive, known, and normalized so that their sum is 1. Fay and Feuer developed a confidence interval for a single age-adjusted rate based on the gamma approximation. Fay used the gamma approximations to construct an F interval for the ratio of two age-adjusted rates. Modifications of the gamma and F intervals are proposed and a simulation study is carried out to show that these modified gamma and modified F intervals are more efficient than the gamma and F intervals, respectively, in the sense that the proposed intervals have empirical coverage probabilities less than or equal to their counterparts, and that they also retain the nominal level. The normal and beta confidence intervals for a single age-adjusted rate are also provided, but they are shown to be slightly liberal. Finally, for comparing two correlated age-adjusted rates, the confidence intervals for the difference and for the ratio of the two age-adjusted rates are derived incorporating the correlation between the two rates. The proposed gamma and F intervals and the normal intervals for the correlated age-adjusted rates are recommended to be implemented in the Surveillance, Epidemiology and End Results Program of the National Cancer Institute.

Estimating the variance of cancer prevalence from population-based registries.

Cancer prevalence is the proportion of people in a population diagnosed with cancer in the past and still alive. One way to estimate prevalence is via population-based registries, where data on diagnosis and life status of all incidence cases occurring in the covered population are collected. In this paper, a method to estimate the complete prevalence and its variance from population-based registries is presented. In order to obtain unbiased estimates of the complete prevalence, its calculation can be thought as made by three steps. Step 1 counts the incidence cases diagnosed during the period of registration and still alive. Step 2 estimates the expected number of survivors among cases lost to follow-up. Step 3 estimates the complete prevalence by taking into account cases diagnosed before the start of registration. The combination of steps 1+2 is defined as the counting method, to estimate the limited duration prevalence; step 3 is the completeness index method, to estimate the complete prevalence. For early established registries, steps 1+2 are more important than step 3, because observation time is long enough to include all past diagnosed cases still alive in the prevalence data. For more recently established registries, step 3 is by far the most critical because a large part of prevalence might have been diagnosed before the period of registration (Corazziari I, Mariotto A, Capocaccia R. Correcting the completeness bias of observed prevalence. Tumori 1999; 85: 370-81). The work by Clegg LX, Gail MH, Feuer EJ. Estimating the variance of disease-prevalence estimates from population-based registries. Biometrics 2002; 55: 1137-44. considers the problem of the variability of the estimated prevalence up to step 2. To our knowledge, no other work has considered the variability induced by correcting for the unobserved cases diagnosed before the period of registration, crucial to estimate the prevalence in recent registries. An analytic approach is considered to calculate the variance of step 3. A unified expression for the variance of the prevalence allowing for steps 1 through 3 is obtained. Some applications to cancer data are presented.

Impact of reporting delay and reporting error on cancer incidence rates and trends.

BACKGROUND: Cancer incidence rates and trends are a measure of the cancer burden in the general population. We studied the impact of reporting delay and reporting error on incidence rates and trends for cancers of the female breast, colorectal, lung/bronchus, prostate, and melanoma. METHODS: Based on statistical models, we obtained reporting-adjusted (i.e., adjusted for both reporting delay and reporting error) case counts for each diagnosis year beginning in 1981 using reporting information for patients diagnosed with cancer in 1981-1998 from nine cancer registries that participate in the Surveillance, Epidemiology, and End Results (SEER) program. Joinpoint linear regression was used for trend analysis. All statistical tests are two-sided. RESULTS: Initial incidence case counts (i.e., after the standard 2-year delay) accounted for only 88%-97% of the estimated final counts; it would take 4-17 years for 99% or more of the cancer cases to be reported. The percent change between reporting-adjusted and unadjusted cancer incidence rates for the 1998 diagnosis year ranged from 3% for colorectal cancers to 14% for melanoma in whites and for prostate cancer in black males. Reporting-adjusted current incidence trends for breast cancer and lung/bronchus cancer in white females showed statistically significant increases (estimated annual percent change [EAPC] = 0.6%, 95% confidence interval [CI] = 0.1% to 1.2%) and 1.2%, 95% CI = 0.7% to 1.6%, respectively), whereas trends for these cancers using unadjusted incidence rates were not statistically significantly different from zero (EAPC = 0.4%, 95% CI = -0.1% to 0.9% and 0.5%, 95% CI = -0.1% to 1.1%, respectively). Reporting-adjusted melanoma incidence rates for white males showed a statistically significant increase since 1981 (EAPC = 4.1%, 95% CI = 3.8% to 4.4%) in contrast to the unadjusted incidence rate, which was most consistent with a flat or downward trend (EAPC = -4.2%, 95% CI = -11.1% to 3.3%) after 1996. CONCLUSIONS: Reporting-adjusted cancer incidence rates are valuable in precisely determining current cancer incidence rates and trends and in monitoring the timeliness of data collection. Ignoring reporting delay and reporting error may produce downwardly biased cancer incidence trends, particularly in the most recent diagnosis years.

Quality of life following localized prostate cancer treated initially with androgen deprivation therapy or no therapy.

BACKGROUND: Many men diagnosed with clinically localized prostate cancer are initially treated conservatively, receiving neither surgery nor radiotherapy for the first year. Treatment patterns and quality-of-life outcomes have not been previously reported for a population-based sample of such men. METHODS: A population-based random sample of men (n = 661) from six geographic regions who had been newly diagnosed with clinically localized prostate cancer from 1994 through 1995 were followed for up to 1 year. Eligible subjects received neither surgery nor radiotherapy within 1 year of initial diagnosis. We assessed disease-specific and generic quality-of-life outcomes in men receiving androgen deprivation therapy (ADT) compared with men receiving no therapy. All statistical tests were two-sided. RESULTS: Two hundred and forty-five study patients received ADT and the remaining 416 patients received no therapy. Approximately two thirds of the patients (n = 159) receiving ADT had either baseline Gleason scores greater than six or serum prostate-specific antigen values above 20 ng/mL. Among men who were sexually potent before diagnosis (ADT = 88 patients; no therapy = 223 patients), 80% of those on ADT reported being impotent after 1 year compared with 30% of those receiving no treatment (P < .001). Patients receiving ADT reported more physical discomfort 1 year after diagnosis than did men who had received no therapy. However, patients receiving ADT, compared with those receiving no therapy, were more likely to be satisfied with their treatment decision (56% pleased versus 45.3%; P =.001). Patients on ADT also experienced a statistically significant decline in vitality, but not in physical function, after adjustment for the confounding factors (P =.05). CONCLUSION: ADT is a commonly used primary therapy for clinically localized prostate cancer. Therefore, men considering ADT as an initial treatment should be aware that sexual function and some aspects of physical well-being are likely to be affected in the first year following this treatment.

Trends in surgery and chemotherapy for women diagnosed with ovarian cancer in the United States.

PURPOSE: We examined patterns of care in a population-based sample of 601 ovarian cancer patients diagnosed in 1991, and a sample of 566 women was selected in 1996 to examine trends in care. PATIENTS AND METHODS: Patient cases were sampled from within the Surveillance, Epidemiology, and End Results program. Medical records were reabstracted, and treatment data were verified with the treating physician. RESULTS: Across these two time periods, the percentage of women with presumptive stage I, II, and IV disease who received lymph node dissection increased. However, a significant number still were not precisely staged. More than 65% of women with ovarian cancer were given cyclophosphamide in 1991 compared with about 14% in 1996. Paclitaxel increased from 1% to 62% during that time. After adjusting for age, race or ethnicity, registry, income, insurance status, Charlson score, residency training program, and marital status, women with early-stage disease were significantly more often given National Institutes of Health Consensus Development Conference guideline therapy in 1996 than in 1991. However, for women with stage III and IV disease, the use of guideline therapy did not significantly increase. Older women and minorities consistently received less guideline therapy, and the lack of private insurance was an impediment for both Hispanic and non-Hispanic black women. CONCLUSION: Despite guidelines presented by several organizations, significant numbers of women with ovarian cancer are not being provided with appropriate care. This is particularly true for older and minority women, especially those without private insurance. Educational strategies must be devised to increase the number of women receiving guideline therapy and decrease disparities across population groups.

CK20 and CK7 protein expression in colorectal cancer: demonstration of the utility of a population-based tissue microarray.

The ability to use archival tissue to test externally valid hypotheses of carcinogenesis is dependent on the availability of population-based samples of cancer tissue. Tissue microarrays (TMAs) provide an efficient format for developing population-based samples of tissue. A TMA was constructed consisting of archival tissue from patients diagnosed with invasive colorectal cancer in the state of Hawaii in 1995. The population representativeness of the TMA was evaluated by comparing patient and clinical characteristics of TMA cases to that of all cases of colorectal carcinoma diagnosed statewide in 1995. Cytokeratin 20 (CK20) and cytokeratin 7 (CK7) immunohistochemistry was used to validate the utility of the TMA, and the expression of these proteins was correlated with patient and tumor characteristics. The TMA comprised tissue specimens from 286 patients representing 47% of all invasive cases diagnosed statewide in 1995. TMA cases were comparable to all invasive colorectal cases statewide with respect to age, sex, race/ethnicity, anatomic site, and survival. There were some differences between TMA cases and all cases with respect to tumor stage, histological classification, and treatment. There were significant differences in the relative expression of CK20 and CK7 proteins between malignant and normal tissues and by tumor stage. Advanced cancers were more likely to have CK20+/cytokeratin 7+ (CK7+) profiles than early-stage cancers, which were predominantly CK20+/cytokeratin 7- (CK7-). CK7+ expression was not correlated with anatomic location of carcinomas. This well-characterized TMA offers a powerful tool for testing hypotheses regarding colorectal carcinogenesis, including the identification of potential markers of neoplastic development and progression.

Tissues from population-based cancer registries: a novel approach to increasing research potential.

Population-based cancer registries, such as those included in the Surveillance, Epidemiology, and End-Results (SEER) Program, offer tremendous research potential beyond traditional surveillance activities. We describe the expansion of SEER registries to gather formalin-fixed, paraffin-embedded tissue from cancer patients on a population basis. Population-based tissue banks have the advantage of providing an unbiased sampling frame for evaluating the public health impact of genes or protein targets that may be used for therapeutic or diagnostic purposes in defined communities. Such repositories provide a unique resource for testing new molecular classification schemes for cancer, validating new biologic markers of malignancy, prognosis and progression, assessing therapeutic targets, and measuring allele frequencies of cancer-associated genetic polymorphisms or germline mutations in representative samples. The assembly of tissue microarrays will allow for the use of rapid, large-scale protein-expression profiling of tumor samples while limiting depletion of this valuable resource. Access to biologic specimens through SEER registries will provide researchers with demographic, clinical, and risk factor information on cancer patients with assured data quality and completeness. Clinical outcome data, such as disease-free survival, can be correlated with previously validated prognostic markers. Furthermore, the anonymity of the study subject can be protected through rigorous standards of confidentiality. SEER-based tissue resources represent a step forward in true, population-based tissue repositories of tumors from US patients and may serve as a foundation for molecular epidemiology studies of cancer in this country.

Cancer survival among US whites and minorities: a SEER (Surveillance, Epidemiology, and End Results) Program population-based study.

BACKGROUND: Available cancer statistics pertain primarily to white and African American populations. This study describes racial or ethnic patterns of cancer-specific survival and relative risks (RRs) of cancer death for all cancers combined and for cancers of the colon and rectum, lung and bronchus, prostate, and female breast for the 6 major US racial or ethnic groups. METHODS: Cancer-specific survival rates were analyzed for more than 1.78 million patients who resided in the 9 SEER (Surveillance, Epidemiology, and End Results) Program geographic areas and were diagnosed between 1975 and 1997 as having an incident invasive cancer, by 6 racial or ethnic groups (non-Hispanic whites, Hispanic whites, African Americans, Asian Americans, Hawaiian natives, and American Indians and Alaskan natives). RESULTS: Survival rates improved between 1988 to 1997 for virtually all racial or ethnic groups. However, racial or ethnic differences in RRs of cancer death persisted after controlling for age for all cancers combined and for age and stage for specific cancer sites (P<.01). African American, American Indian and Alaskan native, and Hawaiian native patients tended to have higher RRs of cancer death than the other groups. American Indians and Alaskan natives generally exhibited the highest RRs of cancer death, except for colorectal cancer in males. CONCLUSIONS: Survival rates in patients with cancer have improved in recent years, but racial or ethnic differences in survival rates and in RRs of cancer death persist. Additional studies are needed to clarify the socioeconomic, medical, biological, cultural, and other determinants of these findings.

Treatment of a population based sample of men diagnosed with testicular cancer in the United States.

OBJECTIVES: Testicular cancer is the most common cancer in men age 25 to 35 years. We examined therapy, compliance with guidelines, and survival in a population based sample of men newly diagnosed with testicular cancer. MATERIALS AND METHODS: We analyzed the National Cancer Institute's (NCI) patterns of care data on 702 men diagnosed with testicular cancer in 1999. These studies supplement routine data collection by verifying therapy with the patients' treating physicians. Follow-up for vital status was available through December 31, 2004. RESULTS: The majority of the men with seminoma were diagnosed while their cancer was localized and more than 80% of received orchiectomy with radiation. For men with seminoma and nonseminoma germ cell tumors (NSGCT), the percent receiving chemotherapy increased markedly as stage increased. More than 90% of men with regional and distant NSGCT received chemotherapy. Less than 25% of men with localized NSGCT received orchiectomy and retroperitoneal lymph node dissection (RPLND), about 40% had surveillance following an orchiectomy alone, and 30% received orchiectomy and chemotherapy. CONCLUSIONS: The majority of these patients received therapy consistent with guidelines. While there was no significant difference in the use of RPLND in men with localized NSGCT by geographic region, chemotherapy use varied widely. Over 90% of men with localized or regional disease diagnosed in 1999 were alive at the end of 2004. The excellent survival rates point to the need to monitor for late effects of therapy.

A new method to evaluate the completeness of case ascertainment by a cancer registry.

BACKGROUND: Epidemiologic research into cancer and subsequent decision making to reduce the cancer burden in the population are dependent on the quality of available data. The more reliable the data, the more confident we can be that the decisions made would have the desired effect in the population. The North American Association of Central Cancer Registries (NAACCR) certifies population-based cancer registries, ensuring uniformity of data quality. An important assessment of registry quality is provided by the index of completeness of cancer case ascertainment. NAACCR currently computes this index assuming that the ratio of cancer incidence rates to cancer mortality rates is constant across geographic areas within cancer site, gender, and race groups. NAACCR does not incorporate the variability of this index into the certification process. METHODS: We propose an improved method for calculating this index based on a statistical model developed at the National Cancer Institute to predict expected incidence using demographic and lifestyle data. We calculate the variance of our index using statistical approximation. RESULTS: We use the incidence model to predict the number of new incident cases in each registry area, based on all available registry data. Then we adjust the registry-specific expected numbers for reporting delay and data corrections. The proposed completeness index is the ratio of the observed number to the adjusted prediction for each registry. We calculate the variance of the new index and propose a simple method of incorporating this variability into the certification process. CONCLUSIONS: Better modeling reduces the number of registries with unrealistically high completeness indices. We provide a fuller picture of registry performance by incorporating variability into the certification process.

Is mode of breast cancer detection associated with cancer treatment in the United States?

BACKGROUND: The prognosis for women who have breast cancer detected by mammography is more favorable than that for women who have breast cancer detected by other methods, even after controlling for tumor characteristics. In the current study, the authors explored whether detection by mammography was associated with greater use of guideline-consistent breast cancer treatment among patients with recently diagnosed breast cancer in the United States. METHODS: The authors evaluated the association between mode of breast cancer detection (mammography vs other) and use of guideline-consistent treatment in 1006 women aged > or =40 years who were diagnosed in 2000. These patients were sampled from the Surveillance, Epidemiology, and End Results Program as part of the Patterns of Care studies. The analyses controlled for the potential confounders of clinical, demographic, and health system characteristics in multivariate logistic regression models. RESULTS: Breast cancer patients who were diagnosed by mammography were more likely to be aged > or =55 years, to have lower stage disease, and to be treated in larger hospitals than patients who were diagnosed by other methods (P < .05). Women whose breast cancer was diagnosed by a method other than mammography were more likely to receive guideline-consistent treatment than women who were diagnosed by mammography in unadjusted (odds ratio, 1.39; 95% confidence interval, 1.07-1.80) and multivariate analyses (odds ratio, 1.43; 95% confidence interval, 1.05-1.95). CONCLUSIONS: The current results indicated that women who had breast cancer detected by methods other than mammography were slightly more likely to receive guideline-consistent therapy than women who had breast cancer detected by mammography. Future research exploring mode of detection, guideline-consistent treatment, and survival among patients with recently diagnosed breast cancer may inform understanding of factors associated with breast cancer prognosis.

Rising incidence rates of breast carcinoma with micrometastatic lymph node involvement.

We investigated the increased incidence of early-stage breast cancer with micrometastatic lymph node involvement. Breast cancer incidence trends from 1990 through 2002 in the US Surveillance, Epidemiology, and End Results Program catchment area were analyzed. Joinpoint regression was used to show the annual percentage change (APC) in breast cancer incidence trends. The overall incidence of breast cancer among women aged 50-64 years increased 1.8% (95% confidence interval [CI] = 1.4% to 2.2%) per annum from 1990 through 2002 but decreased in all other age groups. Stage IIA and stage IIB tumor incidence increased (APC for stage IIA from 1996 to 2002 = 61.9%, 95% CI = 51.1% to 73.4%, and APC for stage IIB from 1998 to 2002 = 53.7%, 95% CI = 20.6% to 96.0%). The incidence of micrometastatic lymph node involvement for stage IIA and stage IIB tumors increased during the 1990s, especially after 1997 (APC = 17.3% for both stages), more for estrogen receptor-positive than estrogen receptor-negative disease. Increased use of mammography screening partly explains the increased incidence of early-stage breast cancer. Increases in small tumors with micrometastatic lymph node involvement may be attributable to the increased use of the sentinel lymph node biopsy in community practice.

Quality of race, Hispanic ethnicity, and immigrant status in population-based cancer registry data: implications for health disparity studies.

Population-based cancer registry data from the Surveillance, Epidemiology, and End Results (SEER) Program at the National Cancer Institute are based on medical records and administrative information. Although SEER data have been used extensively in health disparities research, the quality of information concerning race, Hispanic ethnicity, and immigrant status has not been systematically evaluated. The quality of this information was determined by comparing SEER data with self-reported data among 13,538 cancer patients diagnosed between 1973-2001 in the SEER--National Longitudinal Mortality Study linked database. The overall agreement was excellent on race (kappa = 0.90, 95% CI = 0.88-0.91), moderate to substantial on Hispanic ethnicity (kappa = 0.61, 95% CI = 0.58-0.64), and low on immigrant status (kappa = 0.21. 95% CI = 0.10, 0.23). The effect of these disagreements was that SEER data tended to under-classify patient numbers when compared to self-identifications, except for the non-Hispanic group which was slightly over-classified. These disagreements translated into varying racial-, ethnic-, and immigrant status-specific cancer statistics, depending on whether self-reported or SEER data were used. In particular, the 5-year Kaplan-Meier survival and the median survival time from all causes for American Indians/Alaska Natives were substantially higher when based on self-classification (59% and 140 months, respectively) than when based on SEER classification (44% and 53 months, respectively), although the number of patients is small. These results can serve as a useful guide to researchers contemplating the use of population-based registry data to ascertain disparities in cancer burden. In particular, the study results caution against evaluating health disparities by using birthplace as a measure of immigrant status and race information for American Indians/Alaska Natives.