Effect measures in non‐parametric regression with interactions between continuous exposures

In many biomedical studies, interest is often attached to calculating effect measures in the presence of interactions between two continuous exposures. Traditional approaches based on parametric regression are limited by the degree of arbitrariness involved in transforming these exposures into categorical variables or imposing a parametric form on the regression function. In this paper, we present: (a) a flexible non‐parametric method for estimating effect measures through generalized additive models including interactions; and (b) bootstrap techniques for (i) testing the significance of interaction terms, and (ii) constructing confidence intervals for effect measures. The validity of our methodology is supported by simulations, and illustrated using data from a study of possible risk factors for post‐operative infection. This application revealed a hitherto unreported effect: for patients with high plasma glucose levels, increased risk is associated, not only with low, but also with high percentages of lymphocytes. Copyright © 2005 John Wiley & Sons, Ltd.

[1]  J. Friedman,et al.  Estimating Optimal Transformations for Multiple Regression and Correlation. , 1985 .

[2]  R. Tibshirani,et al.  Generalized additive models for medical research , 1986, Statistical methods in medical research.

[3]  P. McCullagh,et al.  Generalized Linear Models , 1992 .

[4]  Zehua Chen.  Fitting Multivariate Regression Functions by Interaction Spline Models , 1993 .

[5]  M. Wand,et al.  Multivariate Locally Weighted Least Squares Regression , 1994 .

[6]  Jianqing Fan,et al.  Fast Implementations of Nonparametric Curve Estimators , 1994 .

[7]  M. Wand.  Fast Computation of Multivariate Kernel Estimators , 1994 .

[8]  G. Wahba,et al.  Smoothing spline ANOVA for exponential families, with application to the Wisconsin Epidemiological Study of Diabetic Retinopathy : the 1994 Neyman Memorial Lecture , 1995 .

[9]  Lue Ping Zhao,et al.  Estimating Relative Risk Functions in Case-Control Studies Using a Nonparametric Logistic Regression , 1996 .

[10]  G Wahba,et al.  Using smoothing spline anova to examine the relation of risk factors to the incidence and progression of diabetic retinopathy. , 1997, Statistics in medicine.

[11]  David Ruppert,et al.  Fitting a Bivariate Additive Model by Local Polynomial Regression , 1997 .

[12]  A. Figueiras,et al.  Regression models: calculating the confidence interval of effects in the presence of interactions. , 1998, Statistics in medicine.

[13]  Jean D. Opsomer,et al.  Asymptotic Properties of Backfitting Estimators , 2000 .

[14]  M. Wand,et al.  Simple Incorporation of Interactions into Additive Models , 2001, Biometrics.

[15]  A. Figueiras,et al.  Application of nonparametric models for calculating odds ratios and their confidence intervals for continuous exposures. , 2001, American journal of epidemiology.

[16]  Dag Tjøstheim,et al.  NONPARAMETRIC ESTIMATION AND TESTING OF INTERACTION IN ADDITIVE MODELS , 2002, Econometric Theory.

[17]  Göran Kauermann,et al.  Local Likelihood Estimation in Generalized Additive Models , 2003 .

[18]  R. Burnett,et al.  The Effect of Concurvity in Generalized Additive Models Linking Mortality to Ambient Particulate Matter , 2003, Epidemiology.

[19]  Wolfgang Härdle,et al.  BOOTSTRAP INFERENCE IN SEMIPARAMETRIC GENERALIZED ADDITIVE MODELS , 2004, Econometric Theory.