Danny V Colombara1, James P Hughes2, Andrea N Burnett-Hartman3, Stephen E Hawes4, Denise A Galloway5, Stephen M Schwartz3, Roberd M Bostick6, John D Potter7, Lisa E Manhart8. 1. Public Health Sciences Division, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. N., P.O. Box 19024, Seattle, WA 98109-1024, USA; Department of Epidemiology, School of Public Health, University of Washington, F-263 Health Sciences Building, Box 357236, Seattle, WA 98195-7236, USA. Electronic address: dvc2@uw.edu. 2. Public Health Sciences Division, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. N., P.O. Box 19024, Seattle, WA 98109-1024, USA; Department of Biostatistics, School of Public Health, University of Washington, F-600 Health Sciences Building, Box 357232, Seattle, WA 98195-7232, USA. 3. Public Health Sciences Division, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. N., P.O. Box 19024, Seattle, WA 98109-1024, USA; Department of Epidemiology, School of Public Health, University of Washington, F-263 Health Sciences Building, Box 357236, Seattle, WA 98195-7236, USA. 4. Department of Epidemiology, School of Public Health, University of Washington, F-263 Health Sciences Building, Box 357236, Seattle, WA 98195-7236, USA. 5. Human Biology Division, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. N., P.O. Box 19024, Seattle, WA 98109-1024, USA; Department of Microbiology, School of Medicine, University of Washington, Box 357735, Seattle, WA 98195-7735, USA. 6. Department of Epidemiology, Rollins School of Public Health, Emory University, 1518 Clifton Road, NE, Atlanta, GA 30322, USA; Winship Cancer Institute, Emory University, 1365-C Clifton Road NE, Atlanta, GA 30322, USA. 7. Public Health Sciences Division, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. N., P.O. Box 19024, Seattle, WA 98109-1024, USA; Department of Epidemiology, School of Public Health, University of Washington, F-263 Health Sciences Building, Box 357236, Seattle, WA 98195-7236, USA; Centre for Public Health Research, Massey University, PO Box 756, Wellington 6140, New Zealand. 8. Department of Epidemiology, School of Public Health, University of Washington, F-263 Health Sciences Building, Box 357236, Seattle, WA 98195-7236, USA; University of Washington Center for AIDS and STD, 325 9th Ave, Campus Box 359931, Seattle, WA 98104, USA.
Abstract
BACKGROUND: Liquid bead microarray antibody (LBMA) assays are used to assess pathogen-cancer associations. However, studies analyze LBMA data differently, limiting comparability. METHODS: We generated 10,000 Monte Carlo-type simulations of log-normal antibody distributions (exposure) with 200 cases and 200 controls (outcome). We estimated type I error rates, statistical power, and bias associated with t-tests, logistic regression with a linear exposure and with the exposure dichotomized at 200 units, 400 units, the mean among controls plus two standard deviations, and the value corresponding to the optimal sensitivity and specificity. We also applied these models, and data visualizations (kernel density plots, receiver operating characteristic (ROC) curves, predicted probability plots, and Q-Q plots), to two empirical datasets to assess the consistency of the exposure-outcome relationship. RESULTS: All strategies had acceptable type I error rates (0.03 ≤ P ≤ 0.048), except for the dichotomization according to optimal sensitivity and specificity, which had a type I error rate of 0.27. Among the remaining methods, logistic regression with a linear predictor (Power=1.00) and t-tests (Power=1.00) had the highest power to detect a mean difference of 1.0 MFI (median fluorescence intensity) on the log scale and were unbiased. Dichotomization methods upwardly biased the risk estimates. CONCLUSION: These results indicate that logistic regression with linear predictors and unpaired t-tests are superior to logistic regression with dichotomized predictors for assessing disease associations with LBMA data. Logistic regression with continuous linear predictors and t-tests are preferable to commonly used LBMA dichotomization methods.
BACKGROUND: Liquid bead microarray antibody (LBMA) assays are used to assess pathogen-cancer associations. However, studies analyze LBMA data differently, limiting comparability. METHODS: We generated 10,000 Monte Carlo-type simulations of log-normal antibody distributions (exposure) with 200 cases and 200 controls (outcome). We estimated type I error rates, statistical power, and bias associated with t-tests, logistic regression with a linear exposure and with the exposure dichotomized at 200 units, 400 units, the mean among controls plus two standard deviations, and the value corresponding to the optimal sensitivity and specificity. We also applied these models, and data visualizations (kernel density plots, receiver operating characteristic (ROC) curves, predicted probability plots, and Q-Q plots), to two empirical datasets to assess the consistency of the exposure-outcome relationship. RESULTS: All strategies had acceptable type I error rates (0.03 ≤ P ≤ 0.048), except for the dichotomization according to optimal sensitivity and specificity, which had a type I error rate of 0.27. Among the remaining methods, logistic regression with a linear predictor (Power=1.00) and t-tests (Power=1.00) had the highest power to detect a mean difference of 1.0 MFI (median fluorescence intensity) on the log scale and were unbiased. Dichotomization methods upwardly biased the risk estimates. CONCLUSION: These results indicate that logistic regression with linear predictors and unpaired t-tests are superior to logistic regression with dichotomized predictors for assessing disease associations with LBMA data. Logistic regression with continuous linear predictors and t-tests are preferable to commonly used LBMA dichotomization methods.
Authors: M L Gillison; W M Koch; R B Capone; M Spafford; W H Westra; L Wu; M L Zahurak; R W Daniel; M Viglione; D E Symer; K V Shah; D Sidransky Journal: J Natl Cancer Inst Date: 2000-05-03 Impact factor: 13.506
Authors: Nubia Muñoz; F Xavier Bosch; Xavier Castellsagué; Mireia Díaz; Silvia de Sanjose; Doudja Hammouda; Keerti V Shah; Chris J L M Meijer Journal: Int J Cancer Date: 2004-08-20 Impact factor: 7.396
Authors: Ali Rowhani-Rahbar; Joseph J Carter; Stephen E Hawes; James P Hughes; Noel S Weiss; Denise A Galloway; Laura A Koutsky Journal: J Infect Dis Date: 2009-11-01 Impact factor: 5.226
Authors: Tim Waterboer; Rachel Neale; Kristina M Michael; Peter Sehr; Maurits N C de Koning; Sönke J Weißenborn; Francesca Sampogna; Damiano Abeni; Adele C Green; Jan Nico Bouwes Bavinck; Michael Pawlita Journal: J Gen Virol Date: 2009-04-22 Impact factor: 3.891
Authors: Joseph J Carter; Kelly G Paulson; Greg C Wipf; Danielle Miranda; Margaret M Madeleine; Lisa G Johnson; Bianca D Lemos; Sherry Lee; Ashley H Warcola; Jayasri G Iyer; Paul Nghiem; Denise A Galloway Journal: J Natl Cancer Inst Date: 2009-09-23 Impact factor: 13.506
Authors: Kristina M Michael; Tim Waterboer; Peter Sehr; Annette Rother; Ulrich Reidel; Heiner Boeing; Ignacio G Bravo; Jörg Schlehofer; Barbara C Gärtner; Michael Pawlita Journal: PLoS Pathog Date: 2008-06-20 Impact factor: 6.823