THE EFFICIENCY OF THE CUSCORE TEST AS COMPARED TO THAT APPLIED TO SMR IN DETECTION OF A CARCINOGENIC EXPOSURE, pp. 153 - 160
Authors: Rina Chen
Abstract: Quite often statistical analyses are carried out in response to a public alert regarding suspected elevated risk of cancer. Usually no control group is available in these ad-hoc instances, and the analysis is based on the SMR, i.e., on the contrast between the observed and the expected number of events. The SMR (Standardized Morbidity (or Mortality) Ratio) is the ratio between the observed and the expected number of events. The expected number is evaluated according to the age and gender specific rates observed in the reference population. Both types of statistical errors are inflated in these analyses . Namely, frequent false alerts coupled with unsatisfactory power of the test are involved in these analyses. This commentary presents the relative efficiency of the CUSCORE test in alleviating the two types of statistical errors. Analyses of colon cancer deaths among asbestos workers in Israel, demonstrate the efficiency of the procedure in detecting clustering and in providing clues indicating that the significant results are not spurious. Some suggestions are presented in order to increase the efficiency in detection and interpretation of the analyses.
Open Access item.
Click below PDF icon for free download.
This is an Open Access item. Click above PDF icon for free download.