A Rank-Based Test for Comparison of Multidimensional Outcomes

For comparison of multiple outcomes commonly encountered in biomedical research, Huang et al. (2005) improved O’Brien’s (1984) rank-sum tests through the replacement of the ad hoc variance by the asymptotic variance of the test statistics. The improved tests control the type I error rate at the desired level and gain power when the differences between the two comparison groups in each outcome variable lie in the same direction; however, they may lose power when the differences are in different directions (e.g., some are positive and some are negative). These tests and the popular Bonferroni correction failed to show important significant differences when applied to compare heart rates from a clinical trial to evaluate the effect of a procedure to remove the cardioprotective solution HTK. We propose an alternative test statistic, taking the maximum of the individual rank-sum statistics, which controls the type I error rate and maintains satisfactory power regardless of the direction of the differences. Simulation studies show the proposed test to be of higher power than other tests in a certain alternative parameter space of interest. Furthermore, when used to analyze the heart rate data, the proposed test yields more satisfactory results.

[1]  Edgar Brunner,et al.  Nonparametric analysis of longitudinal data in factorial experiments , 2012 .

[2]  George P Chrousos,et al.  Elevated levels of growth‐related hormones in autism and autism spectrum disorder , 2007, Clinical endocrinology.

[3]  Subir Ghosh,et al.  Nonparametric Analysis of Longitudinal Data in Factorial Experiments , 2003, Technometrics.

[4]  I. Olkin,et al.  A generalized bivariate exponential distribution , 1967 .

[5]  S. Janson,et al.  Clinical differences among women with and without self-reported perimenstrual asthma. , 1998, Annals of allergy, asthma & immunology : official publication of the American College of Allergy, Asthma, & Immunology.

[6]  M. Puri,et al.  The multivariate nonparametric Behrens–Fisher problem , 2002 .

[7]  J. Schmee An Introduction to Multivariate Statistical Analysis , 1986 .

[8]  D. Li,et al.  Randomized controlled trial of interferon-beta-1a in secondary progressive MS: MRI results , 2001 .

[9]  Stuart Lipsitz,et al.  Adjusting O'Brien's Test to Control Type I Error for the Generalized Nonparametric Behrens–Fisher Problem , 2005, Biometrics.

[10]  P. O'Brien Procedures for comparing samples with multiple endpoints. , 1984, Biometrics.

[11]  J. Shapiro,et al.  Finasteride in the treatment of men with androgenetic alopecia. Finasteride Male Pattern Hair Loss Study Group. , 1998, Journal of the American Academy of Dermatology.

[12]  B. Turnbull,et al.  Group Sequential Methods with Applications to Clinical Trials , 1999 .

[13]  N L Geller,et al.  The analysis of multiple endpoints in clinical trials. , 1987, Biometrics.

[14]  S. Pillemer,et al.  Global statistical tests for comparing multiple outcomes in rheumatoid arthritis trials. MIRA Trial Group. , 1999, Arthritis and rheumatism.