Annelie M. Landgren, MPH, PMP [Contractor], Surveillance Research Program, DCCPS, NCI
Martha S. Linet, MD, MPH, Scientist Emerita, Radiation Epidemiology Branch, DCEG, NCI
Accurate cancer incidence and cause-of-death ascertainment is critical for epidemiological research. Since there is no nationwide cancer registry in the United States, population-based cohort studies often collect these data from participant self-reported cancer and death certificates—a time consuming and costly enterprise. Data from these sources is subject to error and previous investigations to measure error and validity have been limited by linkage to a small number of cancer registries and a relatively short period of follow-up. An analysis of self-reported and death certificate data from the nationwide U.S. Radiologic Technologists (USRT) cohort study revealed a number of shortcomings from those data sources. The authors conclude that a virtual pooled cancer registry could address these gaps. Their findings and conclusions were published online in the American Journal of Epidemiology (Liu D et al https://pubmed.ncbi.nlm.nih.gov/35872590/; doi: 10.1093/aje/kwac121).
The entire USRT cohort (N=146,022) was linked with 43 population-based U.S. cancer registries for the period 1999-2012. Living members who agreed to participate answered questionnaires administered about every 10 years since the mid-1980s. Death certificate data from cohort members were obtained through 3 linkages of the cohort through 2012. The investigators found sensitivity varied notably by type of cancer. Considering the 8,858 first primary cancers identified from the registries as a gold standard, overall sensitivity was 46.5% for self-reports only (ranging from 102% for the deadliest to 72.4% for cancers with the highest survival rates) and 63.0% for combined self-reports and death certificates. Of the false-negative reports, most were due to subjects not completing questionnaires (27.8%) and to a substantially lesser extent misreporting (9.2%).
Among 90,086 respondents with no record of cancer from registries, self-report and death certificates revealed no record of cancer in 87,831 subjects (specificity of 97.5%). Remaining 2,255 subjects reported a cancer that was not identified in any registry (32.6% of these subjects resided in states/regions not covered by the participating 43 cancer registries). Among the 6,213 self-reported cancers (excluding non-melanoma skin cancer), registries identified 66.2% with the same cancer (concordance or predictive value ranging from 36.8% for melanoma to 81.7% and 77.3% for breast and prostate cancer, respectively), 5.2% with a different cancer, and 28.6% with no cancer.
The findings from this study suggest that National Cancer Institute-supported nationwide virtual pooled cancer registry will substantially improve the completeness and accuracy of cancer case ascertainment as well as the level of cancer characterization detail feasible in U.S. epidemiological observational cohort studies.
Abstract:
Follow-up of US cohort members for incident cancer is time-consuming, is costly, and often results in underascertainment when the traditional methods of self-reporting and/or medical record validation are used. We conducted one of the first large-scale investigations to assess the feasibility, methods, and benefits of linking participants in the US Radiologic Technologists (USRT) Study (n = 146,022) with the majority of US state or regional cancer registries. Follow-up of this cohort has relied primarily on questionnaires (mailed approximately every 10 years) and linkage with the National Death Index. We compared the level of agreement and completeness of questionnaire/death-certificate-based information with that of registry-based (43 registries) incident cancer follow-up in the USRT cohort. Using registry-identified first primary cancers from 1999-2012 as the gold standard, the overall sensitivity was 46.5% for self-reports only and 63.0% for both self-reports and death certificates. Among the 37.0% false-negative reports, 27.8% were due to dropout, while 9.2% were due to misreporting. The USRT cancer reporting patterns differed by cancer type. Our study indicates that linkage to state cancer registries would greatly improve completeness and accuracy of cancer follow-up in comparison with questionnaire self-reporting. These findings support ongoing development of a national US virtual pooled registry with which to streamline cohort linkages.
