Dear all,
We like to announce and share the new release of
bear v2.2.0. The most important new function in this release is the data analysis for
replicate crossover BE study (such as 2-tr x 2-seq x 3 periods, 2x2x4... up to 2x2x6) using
lme (same as SAS PROC MIXED) from R package
nlme. Now we are working on parallel BE study (2x2x1). Please use "Packages" -> "Update packages" from
R console to install the new version if you have installed previous version. You can browse this page for
how-to-install R packages. Other new functions added in bear v2.2.0 are listed as follows:
- add point estimate along with 90% CI in output file called Statistical_summaries.txt (suggested by DLabes).
- add sample size estimation for replicate BE study, and output re-arrangement.
- add Hotelling T2 test and boxplots for outlier detection
- add Quantiles for intrasubject and intersubject (with boxplot)
- add replicated study for 2*2*3, 2*2*4, 2*2*5, and 2*2*6 (using lme to analyze replicated BE study)
- add sample size estimation for replicated study (using 2*2*2 sample size estimation extended to 2*2*n sample size of replicate crossover design)
The following are some output examples obtained from bear v2.2.0.
<<Sample Size Estimation>>
Upper acceptance limit = 125(%)
Lower acceptance limit = 80(%)
Expected ratio T/R = 95.00(%)
Target power = 80.00(%)
Intra-subject CV = 20.0(%)
study 2x2x2 2x2x3 2x2x4
design crossover replicated replicated
------- ----------- ------------ ----------
N 20 16 10
Estimated power= 83.46802(%)
Hotelling T
2 test
(...) Hotelling T^2
--------------------------------------------------------------------------
subj HT_Cmax HT_lnCmax HT_AUC0t HT_lnAUC0t HT_AUC0INF HT_lnAUC0INF
1 1 0.88638 0.85650 0.97444 0.90466 0.82225 0.76937
2 2 1.37871 1.25816 0.46862 0.48286 0.54162 0.56592
3 3 10.76389 8.19027 2.55956 2.47460 2.15554 2.13130
4 4 0.59829 0.53345 1.36768 1.19256 1.32453 1.15615
5 5 1.82827 1.97183 1.27839 1.29252 1.17989 1.20035
6 6 1.59714 1.41680 1.45271 1.28884 1.27021 1.13141
7 7 0.01658 0.04381 0.71059 0.65357 0.68229 0.63622
8 8 4.67137 3.47390 0.06496 0.13465 0.00974 0.04163
9 9 0.58431 0.55860 0.90283 0.87754 0.91402 0.88872
10 10 0.99704 0.92403 1.53365 1.57887 1.21948 1.24508
11 11 5.41157 7.05871 5.44089 6.42777 5.76052 6.93837
12 12 5.58454 7.54222 4.16338 4.89954 4.29064 5.12118
13 13 2.26730 2.47522 4.16519 4.53151 3.94232 4.17831
14 14 0.24448 0.31984 11.39161 9.03305 15.08591 11.63398
-------------------------------------------------
HT_Cmax: Hotelling T^2 for Cmax
HT_lnCmax: Hotelling T^2 for lnCmax
HT_AUC0t: Hotelling T^2 for AUC0t
HT_lnAUC0t: Hotelling T^2 for lnAUC0t
HT_AUC0INF: Hotelling T^2 for AUC0INF
HT_lnAUC0INF: Hotelling T^2 for lnAUC0INF(...)
Display the point estimate:
(...) BE Summary Report - replicate BE study
--------------------------------------------------------------------------
Dependent Variable: lnCmax
--------------------------------------------------------------------------
n1(seq 1)= 7
n2(seq 2)= 7
N(n1+n2) = 14
MEAN-ref = 7.363255
MEAN-test = 7.409809
Estimate(test-ref) = 0.04655381
**************** Classical (Shortest) 90% C.I. for lnCmax *****************
CI90_lower Point_estimated CI90_upper
1 100.312 104.765 109.416
--------------------------------------------------------------------------
And finally, two boxplots.
Hope bear can be useful for you.
All the best,
Hsin-ya Lee, Yung-jin Lee
College of Pharmacy,
Kaohsiung Medical University,
Kaohsiung, Taiwan