Old references… [Power / Sample Size]

posted by Helmut Homepage – Vienna, Austria, 2021-09-14 19:46 (509 d 10:03 ago) – Posting: # 22571
Views: 950

Hi Selwa,

❝ […] i will be very grateful if you forward me the scan of the article and others if you have any.


Check your inbox. ;-)

But again: Don’t rely on the results. An [image]-script at the end. Here the 95% confidence interval of the two studies of 500 mg IR metformin reported by Yuen et al.:

study  n CV.AUC  CI.CV.AUC CV.Cmax CI.CV.Cmax
    1 24  15.1% 11.7–21.5%   16.3% 12.6–23.2%
    2 12   9.5%  6.6–16.8%   12.1%  8.4–21.4%

Be prepared for the worst.
What shall we think about the three studies of 300 mg IR ranitidine in twelve subjects? The CV of AUC0–∞ in the third was twice the one of the second, whereas the CVs of Cmax were similar. On the other hand, the CV of Cmax in the first was twice the one of the third. That’s not helpful.

If you don’t find references for your drug with more consistent results, I recommend a pilot study or – possibly better – a two-stage design.


library(PowerTOST)
studies <- 2L
n       <- c(24L, 12L)
CV.AUC  <- c(15.1, 9.5)
CV.Cmax <- c(16.3, 12.1)
res <- data.frame(study = 1:studies, n = n,
                  CV.AUC = CV.AUC, CI.CV.AUC = NA_real_,
                  CV.Cmax = CV.Cmax, CI.CV.Cmax = NA_real_)
for (j in 1:nrow(res)) {
  tmp <- 100 * CVCL(CV = res$CV.AUC[j] / 100,
                    df = res$n[j] - 2,
                    side = "2-sided")
  res$CI.CV.AUC[j] <- paste0(sprintf("%.1f", tmp[1]), "\u2013",
                             sprintf("%.1f%%", tmp[2]))
  tmp <- 100 * CVCL(CV = res$CV.Cmax[j] / 100,
                    df = res$n[j] - 2,
                    side = "2-sided")
  res$CI.CV.Cmax[j] <- paste0(sprintf("%.1f", tmp[1]), "\u2013",
                             sprintf("%.1f%%", tmp[2]))
}
res$CV.AUC  <- sprintf("%4.1f%%", res$CV.AUC)
res$CV.Cmax <- sprintf("%4.1f%%", res$CV.Cmax)
print(res, row.names = FALSE)


Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Complete thread:

UA Flag
Activity
 Admin contact
22,479 posts in 4,709 threads, 1,603 registered users;
16 visitors (0 registered, 16 guests [including 10 identified bots]).
Forum time: 04:50 CET (Europe/Vienna)

The rise of biometry in this 20th century,
like that of geometry in the 3rd century before Christ,
seems to mark out one of the great ages or critical periods
in the advance of the human understanding.    R.A. Fisher

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5