The Assessment of Non-Linear Effects in Clinical Research

Ton  J. Cleophas; Aeilko  H. Zwinderman

doi:10.4236/pp.2012.32021

Ton J. Cleophas, Aeilko H. Zwinderman

Abstract: Background: Novel models for the assessment of non-linear data are being developed for the benefit of making better predictions from the data. Objective: To review traditional and modern models. Results, and Conclusions: 1) Logit and probit transformations are often successfully used to mimic a linear model. Logistic regression, Cox regression, Poisson regression, and Markow modeling are examples of logit transformation; 2) Either the x- or y-axis or both of them can be logarithmically transformed. Also Box Cox transformation equations and ACE (alternating conditional expectations) or AVAS (additive and variance stabilization for regression) packages are simple empirical methods often successful for linearly remodeling of non-linear data; 3) Data that are sinusoidal, can, generally, be successfully modeled using polynomial regression or Fourier analysis; 4) For exponential patterns like plasma concentration time relationships exponential modeling with or without Laplace transformations is a possibility. Spline and Loess are computationally intensive modern methods, suitable for smoothing data patterns, if the data plot leaves you with no idea of the relationship between the y- and x-values. There are no statistical tests to assess the goodness of fit of these methods, but it is always better than that of traditional models.

Keywords: Non-Linear Effects; Clinical Research; Logit/Probit Transformation; Box Cox Transformation; ACE/AVAS Packages; Curvilinear Data; Spline Modeling; Loess Modeling

Cite this paper: T. J. Cleophas and A. H. Zwinderman, "The Assessment of Non-Linear Effects in Clinical Research," Pharmacology & Pharmacy, Vol. 3 No. 2, 2012, pp. 139-147. doi: 10.4236/pp.2012.32021.

References

[1] J. C. Ferguson, “Multi-Variable Curve Interpolation,” Journal of the Association for Computing Machinery, Vol. 11, No. 2, 1964, pp. 221-228. doi:10.1145/321217.321225

[2] F. Birkhof and R. De Boor, “Piecewise Polynomial Interpretation and Approximation,” Proceedings of General Motors Symposium of 1964, Elsevier Publishing Co., New York, pp. 164-190.

[3] T. J. Cleophas and A. H. Zwinderman, “Monte Carlo methods,” 4th Edition, Statistics Applied to Clinical Trials, Springer, Dordrecht, 2009, pp. 479-485.

[4] SPSS, 2012. www.spss.com

[5] Stata, 2012. www.stat.com

[6] Box-Cox Normality Plot, 2012. http://itl.nist.gov/div898/handbook/eda/section3/eda336.htm

[7] Additive Regression and Transformation Using Ace or Avas, 2011. http://pinard.progicielsbpi.ca/LibR/library/Hmisc/html/transace.html

[8] Anonymous. Free Statistics and Forecasting Software, Box-Cox Normality Plot Calculator, 2012. www.essa.net/rwasp_boxcoxnorm.wasp/

[9] T. J. Cleophas and A. H. Zwinderman, “Curvilinear Regression,” 4th Edition, Statistics Applied to Clinical Trials, Springer, Dordrecht, 2009, pp. 185-196.

[10] T. J. Cleophas and A. H. Zwinderman, “Regression Analysis with Laplace Transformations,” 4th Edition, Statistics Applied to Clinical Trials, Springer, Dordrecht, 2009, pp. 213-216.

[11] L. B. Sheiner and S. L. Beal, “Evaluation of Methods fpr Estimation of Population Pharmacokinetic Parameters,” Journal of Pharmacokinetics and Pharmacodynamics, Vol. 11. 1983, pp. 303-319.

[12] Cubic Spline for Excel, 2012. www.srs1software.com/download.htm#pline.

[13] Lambda-Calculus, 2012. http://en.wikipedia.org/wiki/lambda_calculus.

[14] R. J. Hastie and T. J. Tibshirani “Generalized Additive Models,” Chapman & Hall, London, 1990.

[15] Generalized Additive Model, 2012. http://en.wikipedia.org/wiki/generalized_additive_model.

[16] Local Regression, 2012. http://en.wikidepia.org/wiki/Local_regression.

[17] Akaike Information Criterion, 2012. http://en.wikipedia.org/wiki/Akaike_information_criterion.