Bioequivalence and Bioavailability Forum

Two regions: different populations [General Statistics]

posted by Helmut – Vienna, Austria, 2020-08-31 13:26 (1503 d 17:42 ago) – Posting: # 21900
Views: 7,597

Hi Brus,

❝ But, if α is not adjusted, since there are two reference products, even though they are from different regions, would there be no multiplicity issues?

Consider ElMaestro’s post.
The current regulatory thinking is that an α adjustment is not required since the populations of patients are different (as are the local reference products). If you are concerned about tourists, go ahead and adjust.* If you employ a lower α, any agency will accept that. However, depending on the variability expect roughly 25% larger sample sizes.
BTW, what if tourists switch from one regions’ reference to the other one’s? There is no guarantee that they are BE (see this post). Given, generally we don’t have to worry (see the references in this post).

Thinking about the US and EU: How large is the subpopulation crossing the pond (not to speak of Europeans visiting Trumpistan). How long will tourists stay on the average? How many will run out of the medication brought with them? If you are able to estimate that, possibly you discover that you have to adjust α to 0.049999999.

library(PowerTOST) # defaults: theta0 = 0.95, targetpower = 0.8 CV <- seq(0.15, 0.4, 0.05) res <- data.frame(CV = CV, alpha1 = 0.050, n1 = NA, # unadjusted alpha2 = 0.025, n2 = NA) # Bonferroni for (j in 1:nrow(res)) { res$n1[j] <- sampleN.TOST(alpha = 0.050, CV = CV[j], details = FALSE, print = FALSE)[["Sample size"]] res$n2[j] <- sampleN.TOST(alpha = 0.025, CV = CV[j], details = FALSE, print = FALSE)[["Sample size"]] } res$penalty <- sprintf("%+.2f%%", 100 * (res$n2 - res$n1) / res$n1) names(res)[2:5] <- rep(c("alpha", "n"), 2) print(res, row.names = FALSE) CV alpha n alpha n penalty 0.15 0.05 12 0.025 16 +33.33% 0.20 0.05 20 0.025 24 +20.00% 0.25 0.05 28 0.025 36 +28.57% 0.30 0.05 40 0.025 50 +25.00% 0.35 0.05 52 0.025 66 +26.92% 0.40 0.05 66 0.025 82 +24.24%

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

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

Complete thread:

Decision of confidence interval of three period BE study libaiyi 2018-06-26 12:00 [General Statistics]
- Two tests and one reference Helmut 2018-06-26 13:13
  - Two tests and one reference Brus 2020-08-31 09:58
    - Two tests and one reference ElMaestro 2020-08-31 10:23
    - Two regions: No α adjustment Helmut 2020-08-31 10:26
      - Two regions: No α adjustment Brus 2020-08-31 11:09
        
        Two regions: different populationsHelmut 2020-08-31 11:26
- Decision of confidence interval of three period BE study ElMaestro 2018-06-26 22:20
  - Any of two tests and one reference Helmut 2018-06-27 00:33
    - Any of two tests and one reference libaiyi 2018-06-27 08:25
    - Hey wait a moment.... ElMaestro 2018-06-27 08:29
      - Hey wait a moment.... Helmut 2018-06-27 11:45
        
        Pilot or pivotal ? Ohlbe 2018-06-27 12:14
        
        Pilot or pivotal ? ElMaestro 2018-06-27 13:01
        
        Pilot or pivotal ? Ohlbe 2018-06-27 14:13
        
        Pilot or pivotal ? ElMaestro 2018-06-27 14:48
        
        Pilot or pivotal ? Ohlbe 2018-06-27 15:43
        
        Pilot or pivotal ? ElMaestro 2018-06-27 22:13
        
        Pilot or pivotal ? jag009 2018-06-29 20:53
        
        Pilot or pivotal ? ElMaestro 2018-06-29 22:11
        
        Pilot or pivotal ? jag009 2018-07-03 17:38
        
        Hey, hey wait another moment.... d_labes 2018-06-28 19:48
        
        Hey, hey wait another moment.... Helmut 2018-07-03 17:22
        
        Old belives die hard d_labes 2018-07-04 13:28

Two regions: different populations [General Sta­tis­tics]

Complete thread:

Two regions: different populations [General Statistics]