Premium
The CUSCORE test and the q ‐interval in cluster analyses of colon cancer and of lymphoma among asbestos workers
Author(s) -
Chen R.,
Froom P.
Publication year - 2003
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.1534
Subject(s) - asbestos , statistic , medicine , cluster (spacecraft) , cluster analysis , lymphoma , statistics , colorectal cancer , cancer , confidence interval , test (biology) , mathematics , biology , computer science , paleontology , materials science , metallurgy , programming language
Exposure to a carcinogen is likely to be reflected by temporal clustering of cancer cases among the community members. Detection of such clustering is a complicated task in general and more so in a small workplace community with low turnover rate. In this study we applied the CUSCORE test and the q‐interval statistic to lymphoma and colon cancer death data among 2200 asbestos workers. The CUSCORE test was applied in order to identify clusters, and the q‐interval was used to provide clues related to the cause of the cluster. We also evaluated and tested the data according to the standardized mortality ratio (SMR). Significant increases in both colon cancer and lymphoma among asbestos workers in Israel (1987–1997) were demonstrated by the CUSCORE test but not so by the SMR. The fact that an apparent clustering was observed from the pattern of the q‐intervals leads to the conclusion that the detected clusters are real. Copyright 2003 John Wiley & Sons, Ltd.