Journal of biopharmaceutical statistics
-
Comparative Study
A comparison of the random-effects pattern mixture model with last-observation-carried-forward (LOCF) analysis in longitudinal clinical trials with dropouts.
The last-observation-carried-forward imputation method is commonly used for imputting data missing due to dropouts in longitudinal clinical trials. The method assumes that outcome remains constant at the last observed value after dropout, which is unlikely in many clinical trials. Recently, random-effects regression models have become popular for analysis of longitudinal clinical trial data with dropouts. ⋯ First, subjects are divided into groups depending on their missing-data patterns, and then model parameters are estimated for each pattern. Finally, overall estimates are obtained by averaging over the missing-data patterns and corresponding standard errors are obtained using the delta method. A typical longitudinal clinical trial data set is used to illustrate and compare the above methods of data analyses in the presence of missing data due to dropouts.
-
We present a design for adaptive survival trials, where the probability of randomization to one of two treatments is skewed away from 0.5 according to the current value of the logrank statistic. A formula mapping the logrank statistic onto [0,1] is given, which is then used to bias a coin used for randomization. ⋯ Power is not adversely affected by the resulting unequal allocation. The usual test statistic appears to be standard normal under the proposed allocation scheme.
-
Comparative Study
Statistical comparison between dissolution profiles of drug products.
The problem for assessment of similarity between dissolution profiles of two drug products is considered. The commonly used classical methods including model-dependent and model-independent approaches are reviewed. Most methods encountered the difficulties of no well-defined equivalence limits and the correlation between consecutive time points. ⋯ Accordingly, we proposed a time series approach, which accounts for correlation between dissolution results at different time points. The proposed model is shown to be useful in assessment of similarity between dissolution profiles of two drug products. An example concerning dissolution testing of two lots of a drug product is used to illustrate the proposed equivalence limits and statistical methodology.
-
We investigated the hypothesis that distributions of continuous pharmacokinetic variables are positively skewed in nature and that logarithmic transformation of these variables restores normality. The distributions of common continuous noncompartmental pharmacokinetic variables were investigated for four different Glaxo Wellcome compounds, administered by three different routes of administration: ranitidine (po), sumatriptan (sc), ondansetron (iv), and bismuth, from ranitidine bismuth citrate (po). The distributions of all the investigated noncompartmental pharmacokinetic variables were adequately described by a log-normal distribution, whereas statistically significant departures from normality occurred in the majority of cases. Thus, unless there is strong and consistent evidence for a departure from log-normality, the parametric statistical analysis of common noncompartmental pharmacokinetic variables should be carried out after a priori log transformation.