Bioequivalence and Bioavailability Forum • Some testing (Part I)

Some testing (Part I) [Power / Sample Size]

posted by Helmut – Vienna, Austria, 2013-02-10 19:35 (4424 d 15:24 ago) – Posting: # 9997
Views: 10,710

Dear Detlew!

❝ They would call me "Bigamist" and send me to prison in many countries. And I think my spouse wouldn't be very amused if I convert to a Mussulmán in my old days :-D .

I know. Was in the tradition of this post. :-D

❝ 2-stage design combined with RSABE will have another drawback: Where is the point for simulations which represents alpha :confused: ? Simply GMR=1.25 (or whatever BE limit applies) surely not.

Duno. Of course the limits are flexible, but the GMR-restriction cuts in anyway. Have to think about it.

In the following results of some testing:

The sample size converges pretty fast if the ratio is not extreme.

[image]

In the common range nsims=1e+05 should be sufficient.

In some cases the sample size is lower than Lászlós’. For CV 0.5, ratio 0.85, 90% power, 3-period (EMA) I got n=150 even with 100 mio sim’s whereas in Table A1 I find n=158. Likely their number of simulations (10,000) was too low.

EMA, 3-period, 80% power

        0.85    0.90   0.95  1.00  1.05  1.10   1.15     1.20

      PT   TE  PT  TE PT TE PT TE PT TE PT TE  PT  TE  PT   TE

0.30 195  194  54  53 27 27 24 22 27 26 45 45 105 104 429 >201

0.35 129  127  51  51 30 29 27 25 30 29 45 45  84  84 204 >201

0.40  93   90  45  44 30 29 27 27 30 30 42 42  66  68 138  139

0.45  78   77  42  40 30 29 27 27 30 29 39 37  57  57 126  124

0.50  75   75  39  40 30 30 27 28 30 30 39 37  54  53 144  133

0.55  81   81  42  42 33 32 30 30 33 32 39 40  57  56 171  172

0.60  90   88  48  46 36 36 33 33 36 36 45 44  63  63 198 >201

0.65 102   99  54  53 42 40 39 37 39 40 51 50  72  71 225 >201

0.70 117  109  60  58 45 45 42 41 45 45 57 56  78  80 258 >201

0.75 129  136  66  67 51 50 48 46 51 50 63 62  90  89 285 >201

0.80 144  144  75  72 57 54 51 51 54 55 69 68  99  97 315 >201

EMA, 3-period, 90% power

        0.85    0.90   0.95  1.00  1.05  1.10   1.15     1.20

      PT   TE  PT  TE PT TE PT TE PT TE PT TE  PT  TE  PT   TE

0.30 276 >201  75  74 36 36 27 28 36 36 63 62 147 147 609 >201

0.35 180  181  72  70 39 39 33 32 39 39 63 63 117 117 288 >201

0.40 135  130  63  61 39 38 33 33 39 39 57 57  93  94 231 >201

0.45 129  132  57  55 39 37 33 33 39 38 51 51  84  85 273 >201

0.50 150  158  54  55 39 39 36 34 39 38 51 51  84  84 330 >201

0.55 180  178  57  59 42 41 39 37 42 41 54 53  96  97 393 >201

0.60 207  199  66  64 45 45 42 41 45 46 60 60 111 112 456 >201

0.65 237 >201  72  72 51 51 48 46 51 51 69 67 126 125 519 >201

0.70 270 >201  81  82 60 57 51 52 57 57 75 76 144 141 588 >201

0.75 300 >201  93  93 66 66 57 58 63 64 84 85 159 161 663 >201

0.80 333 >201 102 100 72 70 66 63 72 71 93 93 177 176 723 >201

EMA, 4-period, 80% power

        0.85  0.90  0.95  1.00  1.05  1.10  1.15    1.20

      PT  TE PT TE PT TE PT TE PT TE PT TE PT TE  PT   TE

0.30 128 127 36 35 18 19 14 15 18 18 30 30 68 68 280 >201

0.35  88  88 34 34 20 20 18 18 20 20 30 31 58 57 140  140

0.40  66  64 32 31 20 20 18 18 20 20 28 28 46 47  98   98

0.45  56  57 28 29 20 21 18 19 20 21 26 27 40 41  88   90

0.50  54  54 28 28 22 22 20 20 22 21 26 27 38 38  96  100

0.55  56  55 30 30 22 23 22 21 22 23 28 28 38 40 114  116

0.60  60  60 32 32 24 25 22 23 24 25 30 31 42 44 132  124

0.65  70  74 36 37 28 28 26 26 28 28 34 33 48 49 150  155

0.70  78  78 40 40 30 31 28 28 30 31 38 38 54 55 170  167

0.75  88  85 44 45 34 34 32 32 34 34 42 42 60 61 192  186

0.80  98  95 50 50 38 38 34 35 36 37 46 46 66 66 212 >201

EMA, 4-period, 90% power

        0.85   0.90  0.95  1.00  1.05  1.10   1.15    1.20

      PT   TE PT TE PT TE PT TE PT TE PT TE  PT  TE  PT   TE

0.30 182  180 50 49 24 25 18 19 24 24 42 42  96  95 400 >201

0.35 124  123 48 48 26 27 22 22 26 27 42 43  80  80 202 >201

0.40  94   93 44 42 26 26 24 23 26 26 38 39  66  66 158  165

0.45  90   90 40 40 26 27 24 24 26 27 36 37  60  59 184  181

0.50 100  102 38 39 28 27 24 25 26 27 36 36  60  60 222 >201

0.55 118  123 40 41 28 29 26 26 28 29 38 38  64  63 262 >201

0.60 138  139 44 45 32 32 28 29 32 31 40 41  74  71 304 >201

0.65 158  159 48 51 34 36 32 32 34 35 46 46  84  81 348 >201

0.70 180  172 54 55 40 40 36 36 38 40 50 52  96  97 394 >201

0.75 200  195 62 62 44 43 40 39 44 44 56 58 106 106 442 >201

0.80 222 >201 68 69 48 49 44 45 48 49 62 62 118 113 488 >201

PT: PowerTOST (higher due to balanced sequences, higher, lower), TE: Tóthfalusi/Endrényi

—
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:

Sample size for RSABE Anu 2013-02-01 07:01 [Power / Sample Size]
- Sample size for RSABE d_labes 2013-02-01 09:46
  - Sample size for RSABE Anu 2013-02-07 06:58
    - No analytic solution! Helmut 2013-02-07 12:32
      - Coming soon ... d_labes 2013-02-08 11:57
        
        Wow! Helmut 2013-02-08 12:19
        
        Wow! Anu 2013-02-09 09:20
        
        read… Helmut 2013-02-09 09:35
        
        Bigamist d_labes 2013-02-09 18:11
        
        Some testing (Part I)Helmut 2013-02-10 18:35
        Some testing (Part II) Helmut 2013-02-11 01:00
        
        Power testing d_labes 2013-02-11 12:03
        
        the unknown x Helmut 2013-02-11 14:40
        
        Unequal CVs d_labes 2013-02-15 08:47
        
        Typo Helmut 2013-02-15 13:01