Empirical Comparison of Control Sampling Schemes in a Nested Case-Control Cancer Mortality Study

Background This study was designed to compare effect estimates following different schemes for selecting controls in a nested case-control study investigating lung cancer and multiple myeloma mortality separately among workers in poultry slaughtering/processing plants. These workers have a high exposure to oncogenic viruses compared to the general population.

Method Data from the ongoing Cancer Risk in Workers Exposed to Oncogenic Viruses (CRIWETOV) project for members in a local Union Pension Fund belonging to the United Food &Commercial Workers (UFCW) international union was utilized. The workers were followed–up for mortality from January 1, 1972 to December 31, 2003. This cohort was comprised of workers in poultry slaughtering/processing plants and non-poultry workers. The sampling schemes for selecting controls in this nested case-control study were the cumulative survival, cumulative incidence, case-cohort, and incidence density sampling schemes and were analyzed using logistic regression models.

Result/conclusion The entire cohort and subgroups of poultry and non-poultry workers separately had higher risks of mortality from both malignant diseases compared to the United States' general population, but slightly lower risks among poultry compared to non-poultry workers. Effect estimates were similar for nested case-control analyses that applied the cumulative survival, cumulative incidence and case-cohort sampling schemes in selecting controls. However, the incidence density sampling scheme led to markedly different results. Although results were similar for some schemes, the data still needs to meet different specific underlying assumptions for the application to be valid. For incidence density sampling, a possibly different analytical approach from logistic regression may be required.