Permutation Tests for Matched Pairs with Adjustments for Covariates

SUMMARY In observational studies, treated and control subjects are often matched on the basis of observed covariates to permit comparisons of subjects who appeared similar before treatment. In many studies, however, it is not possible to find an exact match for every treated subject, so that within matched pairs treated and control subjects may still differ with respect to some observed covariates. This paper discusses a simple general method for obtaining permutation inferences that adjust for such observed differences within pairs. The method generalizes both the Wilcoxon signed rank test for continuous responses and the McNemar-Cox test for binary responses, as well as many other permutation tests for matched paris.

[1]  P. Rosenbaum.  The Role of a Second Control Group in an Observational Study , 1987 .

[2]  D. Rubin,et al.  Constructing a Control Group Using Multivariate Matched Sampling Methods That Incorporate the Propensity Score , 1985 .

[3]  P. Rosenbaum.  Conditional Permutation Tests and the Propensity Score in Observational Studies , 1984 .

[4]  P. Rosenbaum.  From Association to Causation in Observational Studies: The Role of Tests of Strongly Ignorable Treatment Assignment , 1984 .

[5]  D. Rubin,et al.  The central role of the propensity score in observational studies for causal effects , 1983 .

[6]  A. M. Walker,et al.  Efficient Assessment of Confounder Effects in Matched Follow-up Studies , 1982 .

[7]  N. Breslow,et al.  Statistical methods in cancer research. Vol. 1. The analysis of case-control studies. , 1981 .

[8]  K. Rothman,et al.  VASECTOMY AND NON-FATAL MYOCARDIAL INFARCTION , 1981, The Lancet.

[9]  T. Holford.  The analysis of pair-matched case-control studies, a multivariate approach. , 1978, Biometrics.

[10]  J. Kelsey,et al.  Multivariate analysis for matched case-control studies. , 1978, American journal of epidemiology.

[11]  R. Prentice.  Use of the logistic model in retrospective studies. , 1976, Biometrics.

[12]  D. Rubin.  Assignment to Treatment Group on the Basis of a Covariate , 1976 .

[13]  D. Rubin.  Estimating causal effects of treatments in randomized and nonrandomized studies. , 1974 .

[14]  D. Rubin.  Matched Sampling for Causal Effects: The Use of Matched Sampling and Regression Adjustment to Remove Bias in Observational Studies , 1973 .

[15]  W. G. Cochran.  The Planning of Observational Studies of Human Populations , 1965 .

[16]  D. Cox.  Two further applications of a model for binary regression , 1958 .

[17]  D. Cox.  THE INTERPRETATION OF THE EFFECTS OF NON-ADDITIVITY IN THE LATIN SQUARE , 1958 .

[18]  M. B. Wilk,et al.  THE RANDOMZATION ANALYSIS OF A GENERALIZED RANDOMIZED BLOCK DESIGN , 1955 .

[19]  Charles Stein,et al.  On the Theory of Some Non-Parametric Hypotheses , 1949 .

[20]  Q. Mcnemar.  Note on the sampling error of the difference between correlated proportions or percentages , 1947, Psychometrika.

[21]  F. Wilcoxon.  Individual Comparisons by Ranking Methods , 1945 .

[22]  B. L. Welch.  ON THE z-TEST IN RANDOMIZED BLOCKS AND LATIN SQUARES , 1937 .

[23]  D. Rubin,et al.  Assessing Sensitivity to an Unobserved Binary Covariate in an Observational Study with Binary Outcome , 1983 .

[24]  N. Breslow,et al.  Statistical methods in cancer research: volume 1- The analysis of case-control studies , 1980 .