JNCI Journal of the National Cancer Institute 2006 98(11):796; doi:10.1093/jnci/djj238
© The Author 2006. Published by Oxford University Press.
CORRESPONDENCE |
Erratum
"Meat and Fat Intake as Risk Factors for Pancreatic Cancer: The Multiethnic Cohort Study" by Nöthlings et al. [J Natl Cancer Inst 2005;97:1458–65 (Issue 19)]. Because of an oversight in how the analysis dataset for the Anderson Gill Cox regression model was set up, some cases were counted as multiple events. As a result, incorrect relative risks, 95% confidence intervals, and P values were calculated. On page 1458, abstract, first sentence of Results section, "68% increased risk" should read "66% increased risk" and "relative risk = 1.68, 95% confidence interval = 1.35 to 2.07, Ptrend<.01" should read "relative risk = 1.66, 95% confidence interval = 1.20 to 2.30, Ptrend = .07." Thus, whereas the relative risk changed minimally, the Ptrend lost statistical significance. In line 23 of the abstract, "50% increases in risk" should read "48% and 42% increases in risk, respectively." In line 25 of the abstract, "(both Ptrend<.01)" should read "(Ptrend<.01 and .03, respectively)." On page 1461, first column, first full paragraph, line 10, "almost 70% higher" should read "over 65% higher." On page 1462, second column, the second sentence of the first full paragraph should read "There was no statistically significant association between nitrosamine intake and pancreatic cancer risk (relative risk = 1.22, 95% confidence interval = 0.89 to 1.69 for the fifth versus the first quintile, Ptrend = .20)." In addition, incorrect data were presented in Table 2 "Relative risks (with 95% confidence intervals) of exocrine pancreatic cancer across quintiles of daily intake of meat, dairy products, and eggs in the Multiethnic Cohort Study, n = 190 545" on page 1461 and Table 3 "Relative risks (with 95% confidence intervals) of exocrine pancreatic cancer across quintiles of daily intake of total fat, saturated fat, and cholesterol intake in the Multiethnic Cohort Study, n = 190 545" on page 1462. The corrected tables are available in the electronic version of the Journal (http://jncicancerspectrum.oxfordjournals.org/jnci/content/vol98/issue11/). Although a few relative risks and P values for trend have lost statistical significance in the corrected analysis, the error does not alter the authors' overall conclusions. The authors regret the error.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||