The impact of National Death Index linkages on population-based cancer survival rates in the United States.

BACKGROUND: In order to ensure accurate survival estimates, population-based cancer registries must ascertain all, or nearly all, patients diagnosed with cancer in their catchment area, and obtain complete follow-up information on all deaths that occurred among registered cancer patients. In the US, linkage with state death records may not be sufficient to ascertain all deaths. Since 1979, all state vital statistics offices have reported their death certificate information to the National Death Index (NDI). OBJECTIVE: This study was designed to measure the impact of linkage with the NDI on population-based relative and cancer cause-specific survival rates in the US. METHODS: Central cancer registry records for patients diagnosed 1993-1995 from California, Colorado, and Idaho were linked with death certificate information (deaths 1993-2004) from their individual state vital statistics offices and with the NDI. Two databases were created: one contained incident records with deceased patients linked only to state death records and the second database contained incident records with deceased patients linked to both state death records and the NDI. Survival estimates and 95% confidence intervals from each database were compared by state and primary site category. RESULTS: At 60 months follow-up, 42.1-48.1% of incident records linked with state death records and an additional 0.7-3.4% of records linked with the NDI. Survival point estimates from the analysis without NDI were not contained within the corresponding 95% CIs from the NDI augmented analysis for all sites combined and colorectal, pancreas, lung and bronchus, breast, prostate, non-Hodgkin lymphoma, and Kaposi sarcoma cases in all 3 states using relative survival methods. Additional combinations of state and primary site had significant survival estimate differences, which differed by method (relative versus cause-specific survival). CONCLUSION: To ensure accurate population-based cancer survival rates, linkage with the National Death Index to ascertain out of state and late registered deaths is a necessary process for US central cancer registries.

Regional, provider, and economic factors associated with the choice of active surveillance in the treatment of men with localized prostate cancer.

Data on initial treatment of 8232 cases of localized prostate cancer diagnosed in 2004 were obtained by medical record abstraction (including hospital and outpatient locations) from seven state cancer registries participating in the Centers for Disease Control and Prevention's Breast and Prostate Cancer Data Quality and Patterns of Care Study. Distinction was made between men receiving no therapy with no monitoring plan (no therapy/no plan [NT/NP]) and those receiving active surveillance (AS). Overall, 8.6% received NT/NP and 4.7% received AS. Older age at diagnosis, lower clinical risk group, and certain registry locations were significant predictors of use of both AS and NT/NP. AS was also related to having more severe comorbidities, whereas nonwhite race was predicted receiving NT/NP. Men receiving AS lived in areas with a higher number of urologists per 100 000 men than those receiving NT/NP. In summary, physician and clinical factors were stronger predictors of AS, whereas demographic and regional factors were related to receiving NT/NP. Physicians appear reluctant to recommend AS for younger patients with no comorbidities.

Impact of using multiple causes of death codes to compute site-specific, death certificate-based cancer mortality statistics in the United States.

BACKGROUND: Cancer mortality statistics, an important indicator for monitoring cancer burden, are traditionally restricted to instances when cancer is determined to be the underlying cause of death (UCD) based on information recorded on standard certificates of death. This study's objective was to determine the impact of using multiple causes of death codes to compute site-specific cancer mortality statistics. METHODS: The state cancer registries of California, Colorado and Idaho provided linked cancer registry and death certificate data for individuals who died between 2002 and 2004, had at least one cancer listed on their death certificate and were diagnosed with cancer between 1993 and 2004. These linked data were used to calculate the site-specific proportion of cancers not selected as the UCD (non-UCD) among all cancer-related deaths (any mention on the death certificate). In addition, the retrospective concordance between the death certificate and the population-based cancer registry, measured as confirmations rates, was calculated for deaths with cancer as the UCD, as a non-UCD, and for any mention. RESULTS: Overall, non-UCD deaths comprised 9.5 percent of total deaths; 11 of the 79 cancer sites had proportions greater than 3 standard deviations from 9.5 percent. The confirmation rates for UCD and for any mention did not differ significantly for any of the cancer sites. CONCLUSION AND IMPACT: The site-specific variation in proportions and rates suggests that for a few cancer sites, death rates might be computed for both UCD and any mention of the cancer site on the death certificate. Nevertheless, this study provides evidence that, in general, restricting to UCD deaths will not under report cancer mortality statistics.

Variability in cancer death certificate accuracy by characteristics of death certifiers.

Death certificates are the source for mortality statistics and are used to set public health goals. Accurate death certificates are vital in tracking outcomes of cancer. Deaths may be certified by physicians or other medical professionals, coroners, or medical examiners. Idaho is one of 3 states that participated in a Centers for Disease Control and Prevention-funded study to assess the concordance between cancer-specific causes of death and primary cancer site among linked cancer registry/death certificate data. We investigated variability in the accuracy of cancer death certificates by characteristics of death certifiers, including certifier type (physician vs coroner), physician specialty, years of experience as death certifier, and number of deaths certified. This study showed significant differences by certifier type/physician specialty in the accuracy of cancer mortality measured by death certificates. Nonphysician coroners had lower accuracy rates compared with physicians. Although nonphysician coroners certified less than 5% of cancer deaths in Idaho, they were significantly less likely to match the primary site from the cancer registry. Results from this study may be useful in the future training of death certifiers to improve the accuracy of death certificates and cancer mortality statistics.

Quality of cancer registry data: findings from CDC-NPCR's Breast and Prostate Cancer Data Quality and Patterns of Care Study.

BACKGROUND: The Breast and Prostate Cancer Data Quality and Patterns of Care (POC-BP) Study enabled a reabstraction study of the quality of population-based, central cancer registry data on the characteristics and initial treatment of breast cancer in females and prostate cancer in the United States. METHODS: Stratified random samples of 9,103 female breast cancers and 8,995 prostate cancers were available for the analysis, using the independently reabstracted data as the gold standard to compute measurements of agreement. RESULTS: A slight majority (53% [8/15]) of the cancer site and treatment combinations showed kappa statistics > or = 0.60 and percent agreements, sensitivities, and predictive values positive > or = 80%: surgery and radiation for the 2 cancers, radiation completed and chemotherapy for breast cancer, and radiation modality and hormone therapy for prostate cancer. The qualities of the Collaborative Stage (CS) site-specific factors and derived variables for the 2 cancers were inconsistent, which confirmed the need to evaluate the recently-implemented CS algorithm. CONCLUSION: The data quality analysis from POC-BP underscores the importance of examining the quality of specific data variables by cancer site, thereby highlighting those variables for which data collection procedures could be improved.

Effect of misclassified underlying cause of death on survival estimates of colon and rectal cancer.

Inaccurate coding of patients' Underlying Cause of Death (UCOD) has constrained cause-specific survival estimates for colon and rectal cancers. Using California data from the Accuracy of Cancer Mortality study, we compared the cancer site data from the California Cancer Registry (CCR) with UCODs reported on death certificates and reclassified the UCODs based on cancer registry data when they disagreed. We then calculated 1-, 3-, 5-, and 10-year cause-specific survival for colon and rectal cancers separately, before and after the reclassification. Records from 26 312 colon and 10 687 rectal cancer patients were examined. UCOD records disagreed with CCR records for 700 (6%) of 11 404 colon cancer deaths and with 1958 (39%) of 5011 rectal cancer deaths, and 82% of the misclassified rectal cancer deaths were coded as colon cancer deaths in the UCOD. Reclassification decreased cause-specific survival for both colon and rectal cancers, but the impact was more pronounced for rectal cancer (eg, 5-year cause-specific survival of colon cancer decreased by 2.8% and of rectal cancer decreased by 20.0% relative to previous estimates; absolute rates changed from 65.4% to 63.6%, and 81.2% to 64.9%, respectively, after reclassification). Interchangeable use of the terms colon cancer and colorectal cancer is likely to be one of the reasons for UCOD misclassification. Educational measures could improve the accuracy of UCOD for colon and rectal cancer deaths.

The accuracy of cancer mortality statistics based on death certificates in the United States.

BACKGROUND: One measure of the accuracy of cancer mortality statistics is the concordance between cancer defined as the underlying cause of death from death certificates and cancer diagnoses recorded in central, population-based cancer registries. Previous studies of such concordance are outdated. OBJECTIVE: To characterize the accuracy of cancer mortality statistics from the concordance between cancer cause of death and primary cancer site at diagnosis. DESIGN: Central cancer registry records from California, Colorado, and Idaho in the U.S. were linked with state vital statistics data and evaluated by demographic and tumor information across 79 site categories. A retrospective arm (confirmation rate per 100 deaths) compared death certificate data from 2002 to 2004 with cancer registry diagnoses from 1993 to 2004, while a prospective arm (detection rate per 100 deaths) compared cancer registry diagnoses from 1993 to 1995 with death certificate data from 1993 to 2004 by International Statistical Classification of Diseases and Related Health Problems (ICD) version used to code deaths. RESULTS: With n=265,863 deaths where cancer was recorded as the underlying cause based on the death certificate, the overall confirmation rate for ICD-10 was 82.8% (95% confidence interval [CI], 82.6-83.0%), the overall detection rate for ICD-10 was 81.0% (95% CI, 80.4-81.6%), and the overall detection rate for ICD-9 was 85.0% (95% CI, 84.8-85.2%). These rates varied across primary sites, where some rates were <50%, some were 95% or greater, and notable differences between confirmation and detection rates were observed. CONCLUSIONS: Important unique information on the quality of cancer mortality data obtained from death certificates is provided. In addition, information is provided for future studies of the concordance of primary cancer site between population-based cancer registry data and data from death certificates, particularly underlying causes of death coded in ICD-10.

Using a statistical process control chart during the quality assessment of cancer registry data.

Statistical process control (SPC) charts may be used to detect acute variations in the data while simultaneously evaluating unforeseen aberrations that may warrant further investigation by the data user. Using cancer stage data captured by the Summary Stage 2000 (SS2000) variable, we sought to present a brief report highlighting the utility of the SPC chart during the quality assessment of cancer registry data. Using a county-level caseload for the diagnosis period of 2001-2004 (n=25,648), we found the overall variation of the SS2000 variable to be in control during diagnosis years of 2001 and 2002, exceeded the lower control limit (LCL) in 2003, and exceeded the upper control limit (UCL) in 2004; in situ/localized stages were in control throughout the diagnosis period, regional stage exceeded UCL in 2004, and distant stage exceeded the LCL in 2001 and the UCL in 2004. Our application of the SPC chart with cancer registry data illustrates that the SPC chart may serve as a readily available and timely tool for identifying areas of concern during the data collection and quality assessment of central cancer registry data.

Surveillance of screening-detected cancers (colon and rectum, breast, and cervix) - United States, 2004-2006.

PROBLEM/CONDITION: Population-based screening is conducted to detect diseases or other conditions in persons before symptoms appear; effective screening leads to early detection and treatment, thereby reducing disease-associated morbidity and mortality. Based on systematic reviews of the evidence of the benefits and harms and assessments of the net benefit of screening, the U.S. Preventive Services Task Force (USPSTF) recommends population-based screening for colon and rectum cancer, female breast cancer, and uterine cervix cancer. Few publications have used national data to examine the stage at diagnosis of these screening-amenable cancers. REPORTING PERIOD COVERED: 2004-2006. DESCRIPTION OF SYSTEMS: Data were obtained from cancer registries affiliated with CDC's National Program of Cancer Registries (NPCR) and the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) program. Combined data from the NPCR and SEER programs provide the best source of information on national population-based cancer incidence. Data on cancer screening were obtained from the Behavioral Risk Factor Surveillance System. This report provides stage-specific cancer incidence rates and screening prevalence by demographic characteristics and U.S. state. RESULTS: Approximately half of colorectal and cervical cancer cases and one third of breast cancer cases were diagnosed at a late stage of disease. Incidence rates of late-stage cancer differed by age, race/ethnicity, and state. Incidence rates of late-stage colorectal cancer increased with age and were highest among black men and women. Incidence rates of late-stage breast cancer were highest among women aged 60-79 years and black women. Incidence rates of late-stage cervical cancer were highest among women aged 50-79 years and Hispanic women. The percentage of persons who received recommended screening differed by age, race/ethnicity, and state. INTERPRETATION: Differences in late-stage cancer incidence rates might be explained partially by differences in screening use. PUBLIC HEALTH ACTION: The findings in this report emphasize the need for ongoing population-based surveillance and reporting to monitor late-stage cancer incidence trends. Screening can identify colorectal, cervical, and breast cancers in earlier and more treatable stages of disease. Multiple factors, including individual characteristics and health behaviors as well as provider and clinical systems factors, might account for why certain populations are underscreened. Cancer control planners, including comprehensive cancer-control programs, can use late-stage cancer incidence and screening prevalence data to identify populations that would benefit from interventions to increase screening utilization and to monitor performance of early detection programs.

Rapid-learning system for cancer care.

Compelling public interest is propelling national efforts to advance the evidence base for cancer treatment and control measures and to transform the way in which evidence is aggregated and applied. Substantial investments in health information technology, comparative effectiveness research, health care quality and value, and personalized medicine support these efforts and have resulted in considerable progress to date. An emerging initiative, and one that integrates these converging approaches to improving health care, is "rapid-learning health care." In this framework, routinely collected real-time clinical data drive the process of scientific discovery, which becomes a natural outgrowth of patient care. To better understand the state of the rapid-learning health care model and its potential implications for oncology, the National Cancer Policy Forum of the Institute of Medicine held a workshop entitled "A Foundation for Evidence-Driven Practice: A Rapid-Learning System for Cancer Care" in October 2009. Participants examined the elements of a rapid-learning system for cancer, including registries and databases, emerging information technology, patient-centered and -driven clinical decision support, patient engagement, culture change, clinical practice guidelines, point-of-care needs in clinical oncology, and federal policy issues and implications. This Special Article reviews the activities of the workshop and sets the stage to move from vision to action.

Factors associated with initial treatment and survival for clinically localized prostate cancer: results from the CDC-NPCR Patterns of Care Study (PoC1).

BACKGROUND: Despite the large number of men diagnosed with localized prostate cancer, there is as yet no consensus concerning appropriate treatment. The purpose of this study was to describe the initial treatment patterns for localized prostate cancer in a population-based sample and to determine the clinical and patient characteristics associated with initial treatment and overall survival. METHODS: The analysis included 3,300 patients from seven states, diagnosed with clinically localized prostate cancer in 1997. We examined the association of sociodemographic and clinical characteristics with four treatment options: radical prostatectomy, radiation therapy, hormone therapy, and watchful waiting. Diagnostic and treatment information was abstracted from medical records. Socioeconomic measures were derived from the 2000 Census based on the patient's residence at time of diagnosis. Vital status through December 31, 2002, was obtained from medical records and linkages to state vital statistics files and the National Death Index. Multiple logistic regression analysis and Cox proportional hazards models identified factors associated with initial treatment and overall survival, respectively. RESULTS: Patients with clinically localized prostate cancer received the following treatments: radical prostatectomy (39.7%), radiation therapy (31.4%), hormone therapy (10.3%), or watchful waiting (18.6%). After multivariable adjustment, the following variables were associated with conservative treatment (hormone therapy or watchful waiting): older age, black race, being unmarried, having public insurance, having non-screen detected cancer, having normal digital rectal exam results, PSA values above 20, low Gleason score (2-4), comorbidity, and state of residence. Among patients receiving definitive treatment (radical prostatectomy or radiation therapy), older age, being unmarried, PSA values above 10, unknown Gleason score, state of residence, as well as black race in patients under 60 years of age, were associated with receipt of radiation therapy. Overall survival was related to younger age, being married, Gleason score under 8, radical prostatectomy, and state of residence. Comorbidity was only associated with risk of death within the first three years of diagnosis. CONCLUSIONS: In the absence of clear-cut evidence favoring one treatment modality over another, it is important to understand the factors that inform treatment selection. Since state of residence was a significant predictor of both treatment as well as overall survival, true regional differences probably exist in how physicians and patients select treatment options. Factors affecting treatment choice and treatment effectiveness need to be further explored in future population-based studies.

Incidence of testicular cancer in the United States, 1999-2004.

Testicular cancer is rare but primarily affects young men. To characterize the current incidence of testicular cancer in the United States, U.S. Cancer Statistics data from 1999 through 2004 were examined. Age-adjusted (2000 U.S. standard) incidence rates were calculated for seminoma and nonseminoma testicular germ cell tumors (TGCTs). Hispanic men had the largest increase in incidence rates for nonseminomas, followed by non-Hispanic White men (annual percentage change of 3.2% and 1.9%, respectively, p < .05). Nonseminomas peaked at a younger age for Hispanic, American Indian/Alaska Native (AIAN), and Asian/Pacific Islander (API) men. Whereas 9.6% of TGCTs were diagnosed at a distant stage in non-Hispanic White men, more Hispanic (16.1%), Black (13.8%), AIAN (16.8%), and API (14.9%) men with TGCTs were diagnosed with distant stage. Monitoring incidence rates for rare cancers by race/ethnicity has improved with national population-based cancer registry coverage. Disparities in diagnosis stage have implications for effective treatment of TGCTs.

Burden of invasive squamous cell carcinoma of the penis in the United States, 1998-2003.

BACKGROUND: Invasive squamous cell carcinoma (SCC) of the penis is rare in the United States. Although human papillomavirus (HPV) infection is an established etiologic agent in at least 40% of penile SCCs, relatively little is known about the epidemiology of this malignancy. METHODS: Population-based data from the National Cancer Institute's Surveillance, Epidemiology and End Results (SEER) program, the Centers for Disease Control and Prevention's National Program for Cancer Registries, and the National Center for Health Statistics were used to examine invasive penile SCC incidence and mortality in the United States. SEER data were used to examine treatment of penile SCC. RESULTS: From 1998 to 2003, 4967 men were diagnosed with histologically confirmed invasive penile SCC in the United States, representing less than 1% of new cancers in men. The annual, average age-adjusted incidence rate was 0.81 cases per 100,000 men, and rates increased steadily with age. Overall, penile SCC incidence was comparable in whites and blacks, but approximately 2-fold lower in Asians/Pacific Islanders. Rates among Hispanics were 72% higher compared with non-Hispanics. Blacks compared with whites and Asians/Pacific Islanders and Hispanics compared with non-Hispanics were diagnosed at significantly younger ages. Higher rates of mortality were also observed among blacks compared with whites and Hispanics compared with non-Hispanics. Penile SCC incidence and mortality were elevated in Southern states and in regions of low socioeconomic status (SES). Some histologic and anatomic site differences were observed by race and ethnicity. Treatment of penile SCC varied with age, stage, and other tumor characteristics. CONCLUSIONS: There are considerable disparities in invasive penile cancer incidence and mortality in the United States. Key risk factors for excess incidence include Hispanic ethnicity and residence in the South and in low SES regions. Risks for excess mortality include these factors in addition to black race. Decreases in penile cancer incidence and mortality in the United States may be realized in the future as the indirect result of prophylactic HPV vaccination of females. Further research is needed to better understand the epidemiology of penile cancer including the role of HPV.

Prostate cancer incidence among American Indian and Alaska Native men, US, 1999-2004.

BACKGROUND: American Indian and Alaska Native (AI/AN) men experience lower incidence of prostate cancer than other race/ethnic populations in the US, but racial misclassification of AI/AN men threatens the validity of these estimates. To the authors' knowledge, little is known concerning prostate-specific antigen (PSA) testing in AI/AN men. METHODS: The authors linked cancer registry data with Indian Health Service enrollment records to improve race classification. Analyses comparing cancer incidence rates and stage at diagnosis for AI/AN and non-Hispanic white (NHW) men for 6 geographic regions focused on counties known to have less race misclassification. The authors also used Behavioral Risk Factors Surveillance System data to characterize PSA testing in AI/AN men. RESULTS: Prostate cancer incidence rates were generally lower in AI/AN than in NHW men for all regions combined (rate ratio of 0.68). However, regional variation was noted among AI/AN men, with incidence rates (per 100,000 population) ranging from 65.7 in the Southwest to 174.5 on the Northern Plains. The rate of distant stage disease was somewhat higher among AI/AN (7.8) than NHW (6.2) men. Nationally, AI/AN men were less likely than NHW men to have undergone recent PSA testing (48.4% vs 58.0%), with prominent regional variation in screening rates noted. CONCLUSIONS: Prostate cancer incidence rates and the proportion of men with recent PSA testing were lower for AI/AN men than for NHW men. However, incident rates and rate of distant stage varied by region more for AI/AN than for NHW. Further research is needed among AI/AN men to evaluate strategies for better understanding the causes of the regional variation in prostate cancer incidence.

Case completeness and data accuracy in the Centers for Disease Control and Prevention's National Program of Cancer Registries.

BACKGROUND: Issues of case completeness (CC) and data quality within the National Program of Cancer Registries (NPCR)-Cancer Surveillance System (NPCR-CSS) are assessed in part by the NPCR Technical Assistance and Audit Program (NPCR-TAA). In addition, the NPCR Annual Program Evaluation Instrument (NPCR-APEI) provides information about NPCR-supported central cancer registries (CCRs). The current report includes a unique, national-level analysis of NPCR-TAA results linked with NPCR-APEI data and other covariates. METHODS: NPCR-TAA results for 34 CCRs were aggregated across diagnosis years 1998 to 2001 for analysis of average CC rates and site-specific data accuracy (DA) rates by covariates obtained from the NPCR-APEI, United States Cancer Statistics (USCS) publications, and the North American Association of Central Cancer Registries (NAACCR) Web site. Site-specific DA rates were calculated for the 13 data elements examined in the audit program. Small-sample Student t tests were used to determine statistically significant differences in covariates (alpha = .05). RESULTS: Overall, the average CC and DA rates were 96.4% and 95%, respectively. Both site- and data element-specific DA issues were highlighted. Higher CC and DA rates were observed for CCRs that were staffed with more certified tumor registrars, had supplementary sources reporting, and met USCS publication standards and/or achieved NAACCR certification. CONCLUSIONS: Study findings underscored the importance of CCRs having adequate, well-trained staff, procuring supplemental reporting sources, and attaining compliance with national data standards. The study results also demonstrated the overall high completeness and quality of NPCR-CSS data and provided guidance to users of the data.

Obesity is negatively associated with prostate-specific antigen in U.S. men, 2001-2004.

BACKGROUND: Recent studies have shown a negative association between body mass index (BMI) and prostate-specific antigen (PSA), a commonly used serum marker for the detection and diagnosis of prostate cancer. We have examined the association between several anthropometric measures and PSA in a nationally representative sample of men. METHODS: We analyzed data from the 2001-2004 National Health and Nutrition Examination Survey. Participants in this study were men ages >or=40 years without previously diagnosed prostate cancer who had PSA measured. Height, weight, waist circumference, BMI, triceps skinfold, subscapular skinfold, and calculated total body water were examined categorically by quintiles using multiple linear regression models. All tests of significance were two sided. RESULTS: Among white men, we report a trend for decreasing PSA with increasing weight, BMI, waist circumference, triceps skinfold thickness, and calculated total body water. Among Mexican American men, we found a trend for decreasing PSA with increasing BMI, and among black men we found a trend for decreasing PSA with increasing triceps thickness. None of the interaction terms between race/ethnicity and any of the anthropometric measures were statistically significant. Controlling for age and race/ethnicity in the multiple linear regression model, we found moderate declines in PSA with a 1 SD increase in BMI [5.9% decrease (95% confidence interval, -9.0% to -2.8%) in geometric mean PSA per 5.2-unit increase], weight [5.9% decline (-8.8% to -2.8%) per 17.7-kg increase], waist circumference [6.6% decline (-9.4% to -3.6%) per 13.4-cm increase], triceps skinfold [5.4% decline (-8.9% to -1.8%) per 6.4-mm increase], and calculated total body water [5.7% decline (-8.9% to -2.4%) per 6.5-liter increase]. CONCLUSION: Our population-based, nationally representative results expand the validity of previous studies on obesity and PSA. Higher weight, BMI, waist circumference, triceps skinfold, and total body water are associated with moderately lower PSA values. A prospective study is needed to verify whether this association affects the accuracy of the PSA test in obese men.