Analysis Of Cross-over Trials Research Articles

Evaluation of a flexible piecewise linear mixed-effects model in the analysis of randomized cross-over trials.

Cross-over designs are commonly used in randomized clinical trials to estimate efficacy of a new treatment. They have received a lot of attention, particularly in connection with regulatory requirements for new drugs. The main advantage of using cross-over designs over conventional parallel designs is increased precision, thanks to within-subject comparisons. In the statistical literature, more recent developments are discussed in the analysis of cross-over trials, in particular regarding repeated measures. A piecewise linear model within the framework of mixed effects has been proposed in the analysis of cross-over trials. In this article, we report on a simulation study comparing performance of a piecewise linear mixed-effects (PLME) model against two commonly cited models-Grizzle's mixed-effects (GME) and Jones & Kenward's mixed-effects (JKME) models-used in the analysis of cross-over trials. Our simulation study tried to mirror real-life situation by deriving true underlying parameters from empirical data. The findings from real-life data confirmed the original hypothesis that high-dose iodine salt have significantly lowering effect on diastolic blood pressure (DBP). We further sought to evaluate the performance of PLME model against GME and JKME models, within univariate modeling framework through a simulation study mimicking a 2 × 2 cross-over design. The fixed-effects, random-effects and residual error parameters used in the simulation process were estimated from DBP data, using a PLME model. The initial results with full specification of random intercept and slope(s), showed that the univariate PLME model performed better than the GME and JKME models in estimation of variance-covariance matrix (G) governing the random effects, allowing satisfactory model convergence during estimation. When a hierarchical view-point is adopted, in the sense that outcomes are specified conditionally upon random effects, the variance-covariance matrix of the random effects must be positive-definite. The PLME model is preferred especially in modeling an increased number of random effects, compared to the GME and JKME models that work equally well with random intercepts only. In some cases, additional random effects could explain much variability in the data, thus improving precision in estimation of the estimands (effect size) parameters.

A note on misspecification in general linear models with correlated errors for the analysis of crossover clinical trials.

Among various approaches to the repeated measures analysis in crossover clinical trials, the general linear models (GLMs) with correlated errors attract substantial attention due to their simplicity in model specification, implementation, and interpretation. The goal of this research note is to conduct simulation studies to numerically investigate the impact of model misspecification in the GLMs with correlated errors in the analysis of crossover trials. A series of synthetic two-treatment and three-treatment crossover trials were designed, and simulation studies were conducted to assess how treatment effect estimation, type I error rates, and power can be affected by misspecified period effects, carryover effects, and variance-covariance structures in the GLMs. Numerical studies confirm that (i) the GLMs with terms for both carryover and period effects and with an unstructured variance-covariance matrix can provide unbiased treatment effect estimates and control of Type I error rates and that (ii) misspecification in either period effects, carryover effects, or covariance structures in the GLMs can induce inflated type I error, declined power, or biased treatment effect estimates. Although methodologic contribution of this research note is minimal, we provide practical recommendations and advice to pharmaceutical sponsors and other investigational drugs and device applicants in designing and analyzing crossover trials using the GLMs with correlated errors.

Incorporating baseline measurements into the analysis of crossover trials with time‐to‐event endpoints

Two-period two-treatment (2×2) crossover designs are commonly used in clinical trials. For continuous endpoints, it has been shown that baseline (pretreatment) measurements collected before the start of each treatment period can be useful in improving the power of the analysis. Methods to achieve a corresponding gain for censored time-to-event endpoints have not been adequately studied. We propose a method in which censored values are treated as missing data and multiply imputed using prespecified parametric event time models. The event times in each imputed data set are then log-transformed and analyzed using a linear model suitable for a 2×2 crossover design with continuous endpoints, with the difference in period-specific baselines included as a covariate. Results obtained from the imputed data sets are synthesized for point and confidence interval estimation of the treatment ratio of geometric mean event times using model averaging in conjunction with Rubin's combination rule. We use simulations to illustrate the favorable operating characteristics of our method relative to two other methods for crossover trials with censored time-to-event data, ie, a hierarchical rank test that ignores the baselines and a stratified Cox model that uses each study subject as a stratum and includes period-specific baselines as a covariate. Application to a real data example is provided.

Handling of baseline measurements in the analysis of crossover trials

Different analytic approaches for modeling baseline data in crossover trials were compared based on the efficiency in estimating treatment effects. Jointly modeling baseline and post-baseline data is recommended to best utilize baseline data. It results in the most significant gain in efficiency when data are strongly correlated within the same period but weakly correlated between different periods. Its performance remains comparable to the best of various other modeling methods under small within period correlation or large between period correlation. We also examined the use of baseline data in modeling carryover effect. We noted that to model carryover effect in crossover trial generally would lead to a much less efficient estimator and much more sensitive inference.

Missing data in clinical trials: a data interpretation problem with statistical solutions?

Mike Kenward speaks to Laura Harvey, Assistant Commissioning Editor. Mike Kenward has been GlaxoSmithKline Professor of Biostatistics at the London School of Hygiene since 1999, with former positions at the Universities of Kent and Reading in the UK and research institutes in the UK, Iceland and Finland. His main research interests are in the analysis of longitudinal data, crossover trials, small sample inference in restricted maximum likelihood and the problem of missing data. He has coauthored three text books including ‘The Design and Analysis of Cross-Over Trials’ with Byron Jones and ‘Missing Data in Clinical Studies’ with Geert Molenberghs. He was formally a coeditor of Biometrics, and is currently an associate editor of Biostatistics. Over the last 25 years he has acted as a consultant in biostatistics, largely for the pharmaceutical industry, and he has given over 100 short courses worldwide on various topics in biostatistics, and has appeared as an expert witness in the US Federal District Court....

The Clinical Development of Percutaneous Heart Valve Technology: A Position Statement of The Society of Thoracic Surgeons (STS), the American Association for Thoracic Surgery (AATS), and the Society for Cardiovascular Angiography and Interventions (SCAI)

The Clinical Development of Percutaneous Heart Valve Technology: A Position Statement of The Society of Thoracic Surgeons (STS), the American Association for Thoracic Surgery (AATS), and the Society for Cardiovascular Angiography and Interventions (SCAI)

Crossover studies with continuous variables: power analysis.

The crossover design is a sensitive means of determining the efficacy of new drugs because it eliminates between-subject variability. However, if the response in the first period carries on into the second (carryover effect) or if time factors cannot be kept constant in a lengthy crossover (time effects), its statistical power may be jeopardized. This may be equally true if a negative correlation exists between treatment responses. We recently demonstrated that the crossover design with binary variables is a powerful method even if correlation between treatment responses is negative. Power analysis of crossover trials with continuous variables has not been explicitly studied. Using the Scheffé model for the assessment of treatment effect, carryover effect, and time effect, we drew power curves of hypothesized crossover studies with different levels of correlation between drug response. We demonstrate that the sensitivity of testing is largely dependent on the levels of correlation between drug response. When positive, we have a great deal of sensitivity with which to test treatment effect and little sensitivity to test carryover or time effect. When negative, the opposite is observed. The correlation level in a crossover comparison is a major determinant of the sensitivity of testing. Treatments from one class with one mode of action frequently have a positive correlation and should be particularly considered for crossover comparisons. With treatment comparisons of totally different classes of drugs/modes of action, the opposite is true. It is hoped that this work affects the design of future crossover trials.

Cross-over trials in drug development: theory and practice

Abstract It is maintained that much research into the design and analysis of cross-over trials has been of little practical relevance to drug development. The point is illustrated using three topics: the AB/BA design, bioequivalence and multi-period designs in two treatments. It is suggested that statisticians should pay more attention to the work of fellow scientists, in particular, in the field of pharmacokinetics, and also that the philosophical–inferential base employed in examining cross-over trials has often been too narrow.

35 How to perform the two-stage analysis of cross-over trials if you can't be persuaded not to

35 How to perform the two-stage analysis of cross-over trials if you can't be persuaded not to

Discussion of the Papers by Atkinson, and Bates et al.

Discussion of the Papers by Atkinson, and Bates <i>et al.</i>

Serial correlation in the design and analysis of crossover trials

Serial correlation in the design and analysis of crossover trials

Design and Analysis of Cross-Over Trials.

Design and Analysis of Cross-Over Trials.

Design and analysis of cross‐over trials. Byron Jones Michael G. Kenward, Chapman and Hall, 1989. No. of pages: 340. Price: £D27.50

Design and analysis of cross‐over trials. Byron Jones Michael G. Kenward, Chapman and Hall, 1989. No. of pages: 340. Price: £D27.50

Design and analysis of cross-over trials, by Byron Jones and Michael G. Kenward. Pp 340. £27·50. 1989. ISBN 0-412-30000-1 (Chapman and Hall)

Design and analysis of cross-over trials, by Byron Jones and Michael G. Kenward. Pp 340. £27·50. 1989. ISBN 0-412-30000-1 (Chapman and Hall)

Some Remarks on the Analysis of Cross-Over Trials with a Binary Response

SUMMARY The possible sources of information on the treatment effect in a cross-over trial with a binary response are examined. In the context of Gart's (1969) logistic regression model for matched pairs, proposals to improve upon Gart's conditional analysis are examined. The use of a general logistic regression model is considered.