The problem of measurement error in modelling the effect of compliance in a randomized trial.

This paper explores the implications of measurement error in the analysis of compliance-response relationships in data from randomized trials. Given that compliance measures are rarely, if ever, error-free indicators of exposure it is argued that both the designs for the collection of compliance data and the statistical models for their resulting analysis should be changed to take the possibility of measurement error into account. An analysis which ignores measurement error in the compliance measurements will provide biased estimates of compliance-response relationships. Provided that one has two or more indicators of compliance for each subject, more appropriate models can be fitted using covariance structure modelling software. If one wishes to explore interactions from repeated measures data on both compliance and response then it is also important that one recognizes that the response measures are also error-prone and that they too are dealt with appropriately.

[1]  E J Goetghebeur,et al.  Analysing non-compliance in clinical trials: ethical imperative or mission impossible? , 1996, Statistics in medicine.

[2]  B Vrijens,et al.  Comparing compliance patterns between randomized treatments. , 1997, Controlled clinical trials.

[3]  K. Jöreskog.  Simultaneous factor analysis in several populations , 1971 .

[4]  S. Zeger,et al.  On estimating efficacy from clinical trials. , 1991, Statistics in medicine.

[5]  Graham Dunn,et al.  Review papers : Design and analysis of reliability studies , 1992 .

[6]  Joshua D. Angrist,et al.  Identification of Causal Effects Using Instrumental Variables , 1993 .

[7]  B. Efron,et al.  Compliance as an Explanatory Variable in Clinical Trials , 1991 .

[8]  D. Tritchler.  Explanatory analyses of randomized studies. , 1996, Biometrics.

[9]  J Urquhart,et al.  Patient non-compliance with drug regimens: measurement, clinical correlates, economic impact. , 1996, European heart journal.

[10]  S. Zeger,et al.  Comment: Dose-Response Estimands , 1991 .

[11]  P. Holland.  Statistics and Causal Inference , 1985 .

[12]  Kenneth A. Bollen,et al.  Structural Equations with Latent Variables , 1989 .

[13]  P. Allison.  Estimation of Linear Models with Incomplete Data , 1987 .

[14]  B. Everitt,et al.  Modelling Covariances and Latent Variables Using EQS , 1993 .

[15]  J. Robins,et al.  Correcting for non-compliance in randomized trials using rank preserving structural failure time models , 1991 .

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

[17]  Geert Molenberghs,et al.  Causal Inference in a Placebo-Controlled Clinical Trial with Binary Outcome and Ordered Compliance , 1996 .

[18]  J Urquhart,et al.  Contending paradigms for the interpretation of data on patient compliance with therapeutic drug regimens. , 1998, Statistics in medicine.

[19]  T. Permutt,et al.  Simultaneous-equation estimation in a clinical trial of the effect of smoking on birth weight. , 1989, Biometrics.

[20]  Barnett Vd.  Simultaneous pairwise linear structural relationships. , 1969 .

[21]  S. Pocock,et al.  The hope and the hazards of using compliance data in randomized controlled trials. , 1998, Statistics in medicine.

[22]  Els Goetghebeur,et al.  The Effect of Treatment Compliance in a Placebo‐controlled Trial: Regression with Unpaired Data , 1997 .

[23]  J. Robins.  Correcting for non-compliance in randomized trials using structural nested mean models , 1994 .

[24]  J. Rochon.  Supplementing the Intent-to-Treat Analysis: Accounting for Covariates Observed Postrandomization in Clinical Trials , 1995 .

[25]  G. Dunn.  Comparative calibration without a gold standard by Ying Lu, Keying Ye, Ashwini K. Mathur, Siu Hui, Thomas P. Fuerst and Harry K. Genant, Statistics in Medicine, 16, 1889–1905 (1997) , 1998 .

[26]  Bengt Muthén,et al.  On structural equation modeling with data that are not missing completely at random , 1987 .

[27]  C E Werts,et al.  Confirmatory Factor Analysis Applications: Missing Data Problems And Comparison Of Path Models Between Populations. , 1979, Multivariate behavioral research.

[28]  J. Angrist,et al.  Two-Stage Least Squares Estimation of Average Causal Effects in Models with Variable Treatment Intensity , 1995 .

[29]  G Lewis,et al.  Measuring psychiatric disorder in the community: a standardized assessment for use by lay interviewers , 1992, Psychological Medicine.

[30]  Donald B. Rubin,et al.  Comment: Dose-Response Estimands , 1991 .

[31]  P. Holland.  CAUSAL INFERENCE, PATH ANALYSIS AND RECURSIVE STRUCTURAL EQUATIONS MODELS , 1988 .