Bioequivalence and Bioavailability Forum

Meta-analysis vs. naïve pooling [General Sta­tis­tics]

posted by Helmut  – Vienna, Austria, 2010-04-09 22:33 (5922 d 12:43 ago) – Posting: # 5070
Views: 10,984

Dear D. Labes & earlybird!

❝ SUGI 27: SAS® Tools for Meta-analysis

Is it really that easy? See the example at page 2:

Suppose we have [...] two identical studies, following the same protocol, using the same treatments, using the same response variable, and suppose once again that we can obtain the data. We could then fit a mixed model (if we wanted to treat study as a random effect), using PROC MIXED, or an ANOVA model (if we wanted to treat study as a fixed effect), using PROC GLM.

It took me while to find suitable data in my files - who repeats a study without any modifications? This was the closest I could get: AUC-data, same protocol (clinical site, sampling schedule, analytical method,...). Different batches of test and reference were used (actually the reference came from two countries).

Single studies
Fixed: sequence+treatment+period
Random: subject(sequence)
Study 1: n=12, 99.93% (94.46% - 105.71%), CVintra 7.61%
Study 2: n=13, 91.00% (85.26% -  97.12%), CVintra 9.24%

Naïve pooling (study not in the model)
Fixed: sequence+treatment+period
Random: subject(sequence)
Pooled 1: n=25, 95.05% (90.86% -  99.43%), CVintra 9.31%
Interesting: The CV is higher than in either study despite the higher sample size.

OK, let's add the study to the model...
Fixed: sequence+treatment+period
Random: subject(sequence)+study
Pooled 2: n=25, 95.05% (90.85% -  99.43%), CVintra 9.31%
Aha No.1

And code all as fixed...
Fixed: sequence+treatment+period+subject(sequence)+study
Pooled 3: n=25, 95.05% (90.86% -  99.43%), CVintra 9.31%
Aha No.2

Besides some minor differences in the second decimal of the lower CL all models came up the same. Why all the fuzz with sequential designs, alpha-adjustements and the like - if I can simply throw all data into the big pot, stir it up, and voilá...

Again: Is it really that easy?
Both studies were already evaluated, the alpha is 'consumed' - any professional statisticians out there?

—
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:

• Meta-analysis earlybird 2010-04-09 07:54 
  • Meta-analysis in SAS d_labes 2010-04-09 08:35
    • Meta-analysis vs. naïve poolingHelmut 2010-04-09 20:33
      • Meta-analysis vs. naïve pooling ElMaestro 2010-04-09 22:02
        • Meta-analysis vs. naïve pooling Helmut 2010-04-09 22:11
          • Meta-analysis vs. naïve pooling ElMaestro 2010-04-09 22:20
            • Meta-analysis vs. naïve pooling Helmut 2010-04-10 00:49
              • Meta-analysis vs. naïve pooling ElMaestro 2010-04-10 08:13
                • Yes, but... Helmut 2010-04-10 14:46
                  • Yes, but... ElMaestro 2010-04-11 11:44
                    • Yes, but... Helmut 2010-04-11 14:27
      • Meta-analysis alpha d_labes 2010-04-12 13:32
        • Meta-analysis alpha Helmut 2010-04-12 14:25
          • 2-Study Meta-analysis d_labes 2010-04-12 15:03
            • 2-Study Meta-analysis Helmut 2010-04-12 15:18
              • Anatomical miracle d_labes 2010-04-12 15:33
                • Anatomical miracle Helmut 2010-04-12 15:40