Bioequivalence and Bioavailability Forum • Outlier tests: forget it!

Outlier tests: forget it! [Bioanalytics]

posted by Helmut – Vienna, Austria, 2010-01-19 18:36 (6004 d 14:07 ago) – Posting: # 4618
Views: 9,713

Dear Yuvraj!

Again: there’s no need for a formal outlier test. Reading the guidance you would have noticed that a QCs sample may be excluded, if:

Inaccuracy > ±15%
Precison (CV) > ±15%
not more than ⅔ of QCs
not more than 50% at the same level

±15% are valid for simple methods; wider limits are possible for ligand binding assays. Wider limits are also valid at the LLOQ.
Consider a simple dataset (n=3; that’s the minimum to do statistics I would say…):
QC1 80% (suspected outlier; accuracy <85%)

QC2  95%

QC3 105%

x    98.3%

(acceptable: within 100±15%)

SD   12.6%

CV   13.5%

(acceptable: <15%)
Grubbs-test: 1.060<1.155, p≥0.05
Dixon-test: 0.6<0.941, p≥0.05
Henning-test: 13.33<37.75, no outlier
Robust: no outlier

Now what? You may exclude the value according to the guidance, but none of the tests would support that. Not even a “zero” result for QC1 would give you a significant result. BTW, you may exclude values, but you don’t have to. In the example most people wouldn’t do so, because both accuracy and precision are within limits.
Forget it.

—
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:

outliers test for QC sample mathews 2007-10-11 07:58
- outliers test - but not for QCs! Helmut 2007-10-11 15:31
  - Outlier test - but not for QC's yuvrajkatkar 2010-01-19 08:49
    - Outlier test - but not for QC's H_Rotter 2010-01-19 12:55
    - Outlier tests: forget it!Helmut 2010-01-19 17:36
      - Outlier tests: forget it! Ohlbe 2010-01-19 19:00
        
        Outlier tests: forget it! Helmut 2010-01-19 19:16
        
        Outlier tests: forget it! Ohlbe 2010-01-19 21:00
        
        QCs vs. calibrators Helmut 2010-01-19 21:49
        
        QCs vs. calibrators Ohlbe 2010-01-20 19:01
        
        QCs vs. calibrators Helmut 2010-01-20 19:26
        
        QCs vs. calibrators Ohlbe 2010-01-20 19:52
        
        Two thirds of QCs H_Rotter 2010-01-20 16:21
        
        Two thirds of QCs Ohlbe 2010-01-20 18:49
        
        Outlier tests: forget it!....what if.... moblak 2010-01-25 10:04
        
        Outlier tests: forget it!....what if.... Ohlbe 2010-01-25 23:19