The problem of measurement error in modelling the effect of compliance in a randomized trial.
21
TL;DR: 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.
read more
Abstract: 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.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Partial Least Squares
Ton J. Cleophas,Aeilko H. Zwinderman +1 more
- 01 Jan 2013
TL;DR: The authors used Principal Component Analysis (PCA) and partial least square analysis (PLSSA) to reduce large numbers of variables in clinical trials, but they did not consider the relative importance of the separate variables, their interactions and differences in units.
953
Improving on a Coin Toss To Predict Patient Adherence to Medications
TL;DR: Of all four measures, CAS (supplemented MEMS) was the most strongly predictive of viral load; the receiver-operating characteristic (ROC) curve statistic was 0.84; the improvement in the ROC statistic for CAS compared with the MEMS-only measure shows that missing electronic monitor data may be informative.
116
Weighted Blankets and Sleep in Autistic Children—A Randomized Controlled Trial
Paul Gringras,Dido Green,Barry Wright,Carla Rush,Masako Sparrowhawk,Karen Pratt,Victoria Allgar,Naomi Hooke,Danielle Collingridge Moore,Zenobia Zaiwalla,Luci Wiggs +10 more
TL;DR: The use of a weighted blanket did not help children with ASD sleep for a longer period of time, fall asleep significantly faster, or wake less often, but the weighted blanket was favored by children and parents, and blankets were well tolerated over this period.
101
Structural mean models for compliance analysis in randomized clinical trials and the impact of errors on measures of exposure.
TL;DR: This paper argues that measurement error on exposure is an important practical complication that has, however, not yet been addressed, and shows how standard additive linear structural mean models are robust against unbiased measurement error and how efficient, asymptotically unbiased inference can be drawn when the degree of measurement error bias is known.
Beyond Intention to Treat
Els Goetghebeur,Tom Loeys +1 more
TL;DR: It is a special pleasure to review approaches beyond intention to treat (ITT) for an audience of epidemiologists and draw robust conclusions based on simple contrasts between groups as randomized.
42
References
Estimating causal effects of treatments in randomized and nonrandomized studies.
TL;DR: A discussion of matching, randomization, random sampling, and other methods of controlling extraneous variation is presented in this paper, where the objective is to specify the benefits of randomization in estimating causal effects of treatments.
10.2K
Statistics and Causal Inference
TL;DR: In this article, the authors use a particular model for causal inference (Holland and Rubin 1983; Rubin 1974) to critique the discussions of other writers on causation and causal inference.