Letters Section Editor: Robert M. Golub,
MD, Senior Editor.
In Reply: We agree with Drs Zahm and Fraumeni
that the meta-analysis process is limited by the information provided in the
individual studies. If data are available in only 1 or 2 studies, it may be
difficult to include them in a meta-analysis. This is why we did not claim
that our meta-analysis was conclusive about the potential carcinogenic effect
of hair dyes and called for further studies.
Zahm and Fraumeni mention 5 individual ORs for dark colored dyes, but
they provide the corresponding 95% CI only for Zhang et al (OR, 2.1; 95% CI,
1.0-4.0).1 The 95% CI corresponding to the
OR of 2.1 from the study by Altekruse et al2 is
0.7 to 6.7 and is based on only 3 exposed cases. The OR for black hair coloring
in the study by Zahm et al3 is 4.1, but its
95% CI is 0.9 to 18.8 (representing 4 cases), and the OR for red hair dyes
is 3.0, with a 95% CI of 0.5 to 16.8 (representing 3 cases). The imprecision
of these estimates does not favor any firm conclusion about a causal effect
of hair dyes on cancer.
Takkouche B, Montes-Martínez A, Etminan M. Hair Dye Use and Non-Hodgkin Lymphoma—Reply. JAMA. 2005;294(10):1205. doi:10.1001/jama.294.10.1205-b