surajpatel
☆    

India,
2021-05-27 10:14
(1037 d 07:43 ago)

Posting: # 22369
Views: 1,888
 

 Exclusion of Outlier Data [Outliers]

Hi,

We completed fasting bioequivalence study for Health Canada with 28 subjects. 4 subjects are showing outliers (more than 10 times of Cmax from other subjects). Can we exclude these subjects and submit the data to Health Canada. Will they accept after exclusion because the study fails if we include those 4 outliers data or we need to go for re-dosing those 4 subjects and submit the new data. Thanks.

Suraj
drgunasakaran1
★★  
avatar

2021-05-27 10:58
(1037 d 06:58 ago)

@ surajpatel
Posting: # 22370
Views: 1,630
 

 Exclusion of Outlier Data

Dear Mr Suraj,

❝ We completed fasting bioequivalence study for Health Canada with 28 subjects. 4 subjects are showing outliers (more than 10 times of Cmax from other subjects). Can we exclude these subjects and submit the data to Health Canada. Will they accept after exclusion because the study fails if we include those 4 outliers data or we need to go for re-dosing those 4 subjects and submit the new data.


As per Health Canada Guidance Document on Conduct and Analysis of Comparative Bioavailability Studies "No more than 5% of the subjects may be considered to be outliers, unless there are 20 or fewer subjects. In your case, more than 5% of subjects are falling under the category of outliers.
Also, Re‐testing of subjects identified as outliers is not recommended as per Health Canada"
.

Dr S Gunasakaran MBBS MD
Disclaimer: The replies/posts are my personal opinions and it does not represent my company views on the same.
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2021-05-27 12:14
(1037 d 05:43 ago)

@ surajpatel
Posting: # 22371
Views: 1,619
 

 Health Canada: read the guidance

Hi Suraj,

adding to Dr S Gunasakaran’s comments:
  1. The procedure to deal with outliers has to be stated in the protocol.
  2. Identification of outliers should be based on a test. ±3 studentized model residuals is recommended. The magnitude of the deviation of the observed values is not relevant.
  3. The procdure should be followed before the study is assessed for BE.
Seemingly you didn’t comply with all of the mentioned conditions.

Next time read the respective guidance before planning the study.

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
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,636 registered users;
98 visitors (0 registered, 98 guests [including 11 identified bots]).
Forum time: 16:57 CET (Europe/Vienna)

With four parameters I can fit an elephant,
and with five I can make him wiggle his trunk.    John von Neumann

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