libaiyi
★    

China,
2019-04-19 09:32

Posting: # 20192
Views: 407
 

 SAS code for the mixed effects model with sequence mistakenly assigned [Study As­sess­ment]

Dear All,

We know that the model statement in SAS for the mixed effects model to evaluate bioequivalence is
model logpkp =trtseqan aperiod trtan / ..., in which the planned sequence is used.

In a 2*2 BE study, if subject A with planned sequence T-R is assigned to R-T and B with planned R-T has the actual sequence of T-R, what code can we use for the model statement with these two subjects included for a sensitivity analysis?

Thanks in advance!


Edit: Category changed; see also this post #1[Helmut]
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2019-04-19 12:07

@ libaiyi
Posting: # 20193
Views: 360
 

 Failure in randomization

Hi libaiyi,

» In a 2*2 BE study, if subject A with planned sequence T-R is assigned to R-T and B with planned R-T has the actual sequence of T-R, what code can we use for the model statement with these two subjects included for a sensitivity analysis?

Is this related to this post of yours? That’s not a statistical issue but a serious violation of GCP. Of course, you could use the actual sequences in a sensitivity analysis. Even if both the per-protocol analysis and the sensitivity analysis pass, it would rise serious doubts about the procedures of the CRO. There were too many cases in the past (see this post) which would – rightly – set any assessor’s alarm bells off and likely trigger an inspection. Get prepared.

I faced not following the randomization once. The CRO ignored the randomization provided by the sponsor (well, claimed that it never was received) and used their own. How stupid is this? If the CRO really didn’t receive it in a timely manner, why not simply pick up the phone and sort it out with the sponsor? However, I got the randomization from the sponsor, study passed, statistical report (PK part) final.
I was also responsible to evaluate the safety part. When I got the CRFs from the CRO I realized that the drug administration didn’t match the randomization I had. F**k! Sensitivity analysis with the “true” randomization as an amendment to the statistical report. Study passed.
However, after some legal to and fro the CRO repeated the study at its own cost and never was contracted by the sponsor again.

Cheers,
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. ☼
Science Quotes
mittyri
★★  

Russia,
2019-04-19 16:19

@ Helmut
Posting: # 20198
Views: 310
 

 Real sensitivity

Hi Helmut,

» Sensitivity analysis with the “true” randomization as an amendment to the statistical report. Study passed.

To perform true sensitivity analysis did you use not only TRUE/FALSE randomization lists but also other possible random sequences too?;-)

Kind regards,
Mittyri
ElMaestro
★★★

Denmark,
2019-04-19 15:14

@ libaiyi
Posting: # 20194
Views: 321
 

 What is your root cause?

Hi Libaiyi,

may I ask if you did a root cause analysis and if you did, were you able to assign a root cause to this case?

The RC determines the CAPA. In the absence of a really good RC here, the CAPA could be quite impossible. If nothing is done, it could happen again, so obviously something's gotta be changed here.

if (3) 4

x=c("Foo", "Bar")
b=data.frame(x)
typeof(b[,1]) ##aha, integer?
b[,1]+1 ##then let me add 1


Best regards,
ElMaestro

“(...) targeted cancer therapies will benefit fewer than 2 percent of the cancer patients they’re aimed at. That reality is often lost on consumers, who are being fed a steady diet of winning anecdotes about miracle cures.” New York Times (ed.), June 9, 2018.
Activity
 Thread view
Bioequivalence and Bioavailability Forum |  Admin contact
19,490 posts in 4,135 threads, 1,335 registered users;
online 7 (0 registered, 7 guests [including 6 identified bots]).
Forum time (Europe/Vienna): 19:03 CEST

No rational argument will have a rational effect on a man
who does not want to adopt a rational attitude.    Karl R. Popper

The BIOEQUIVALENCE / BIOAVAILABILITY FORUM is hosted by
BEBAC Ing. Helmut Schütz
HTML5