Publications

• Computing Exact Power for Multivariate Repeated Measurements Design
  ABSTRACT
  Repeated measurements often arise in medical and laboratory research when investigators are constrained by costs or time to give all study treatments to each experimental unit/participant one at a time. In this paper, we derive a generalized method for computing the exact power for a repeated measurements design given the simple case of
  no learning effect.
  
  
• Computing Exact Power and Sample Size for Hotelling’s T2-test and Related Multivariate Procedures
  ABSTRACT
  Multiple comparisons often arise in medical and epidemiologic research and several methods exist for adjusting statistical significance levels. In this paper, we discuss the multivariate method attributed to Harold Hotelling for handling multiplicity in the pvariate two-sample case. A generalized method is derived for computing the exact power for this test. The technique may be easily adapted to other multivariate procedures that utilize the T2-statistic.
  
  
• Efird, J.T, A method for indirectly estimating gene-environment effect modification and power fiven only genotype frequency and odds ratio of environmental exposure Eur J Epidemiol. 2005; 20(5): 389-93.
  
  
• Efird JT, Holly EA, Cordier S, Mueller BA, Lubin F, Filippini G, Peris-Bonet R, McCredie M, Arslan A, Bracci P, Preston-Martin S. Beauty product-related exposures and childhood brain tumors in seven countries: results from the SEARCH International Brain Tumor Study J Neurooncol. 2005 Apr;72(2):133-47.
  
  
• Efird JT, Friedman GD, Sidney S, Klatsky A, Habel LA, Udaltsova NV, Van den Eeden S, Nelson LM. The risk for malignant primary adult-onset glioma in a large, multiethnic, managed-care cohort: cigarette smoking and other lifestyle behaviors J Neurooncol.2004 May;68(1):57-69.
  
  
• Computing Differential Sample Size for Case-control Studies ofGene-environment Interaction
  ABSTRACT
  The incidence rates for diseases such as cancer, cardiovascular disease, and diabetes are known to differ by ethnic/racial group. Yet, neither genetic nor environmental factors fully explain the observed differences. Failure to account for genetic expression in the absence or presence of an environmental factor and vice-versa, may lead to erroneous conclusions regarding the importance of these factor and vice-versa, may lead to erroneous conclusions regarding the importance of these factors in disease etiology. In this paper, the authors present a novel method for computing sample size for case-control studies involving the interaction of genetic and environmental factors. The method is based on an indirect estimate of the odds ration for gene-environment interaction given only the odds ration for environmental exposure and population genotype frequency. A table is presented providing sample sizes required for detecting a minimum odds ration for gene-environment interaction given varying genotype frequencies and environmental exposure odds ration values. Sample size is shown to increase proportionately with genotype frequency for a given environment exposure odds ratio.
  
  
• Yeo KK, Wijetunga M, Ito H, Efird JT, Tay K, Seto TB, Alimineti K, Kimata C, Schatz IJ, The association of methamphetamine use and cardiomyopathy in young patients. Am J Med. 2007 Feb;120(2):165-71.
  
  
• Steven T.S. Lim, Klaus Dragull, Chang-Shih, Harry C Bittenbender, Jimmy T. Efird, and Pratibha V. Nerurkar, Effects of Kava Alkaloid, Pipermethystine, and Kavalactones on Oxidative Stress and Cytochrome P450 in F-344 Rats

651 Ilalo St. Biosciences Bldg. 320B
Honolulu HI 96813
Voice: (808)692-1648
Fax: (808)692-1979