c:\ian\git\artbin\package\artbin.ado
*!version 2.1.0 IW 04jan2024
c:\ian\git\artbin\package\art2bin.ado
*!version 1.01 09june2022
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .2
Alternative hypothesis H1: pi2 - pi1 < .2
Anticipated event probabilities 0.100 0.100
Alpha 0.100 (two-sided)
(taken as .05 one-sided)
Power (designed) 0.900
Total sample size (calculated) 78
Sample size per group (calculated) 39 39
Expected total number of events 7.8
Expected number of events per group 3.9 3.9
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .05
Alternative hypothesis H1: pi2 - pi1 < .05
Anticipated event probabilities 0.300 0.300
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (designed) 0.900
Total sample size (calculated) 3532
Sample size per group (calculated) 1766 1766
Expected total number of events 1059.6
Expected number of events per group 529.8 529.8
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .15
Alternative hypothesis H1: pi2 - pi1 < .15
Anticipated event probabilities 0.150 0.150
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (designed) 0.900
Total sample size (calculated) 240
Sample size per group (calculated) 120 120
Expected total number of events 36
Expected number of events per group 18 18
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .1
Alternative hypothesis H1: pi2 - pi1 < .1
Anticipated event probabilities 0.200 0.200
Alpha 0.200 (two-sided)
(taken as .1 one-sided)
Power (designed) 0.800
Total sample size (calculated) 290
Sample size per group (calculated) 145 145
Expected total number of events 58
Expected number of events per group 29 29
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .05
Alternative hypothesis H1: pi2 - pi1 < .05
Anticipated event probabilities 0.100 0.100
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (designed) 0.900
Total sample size (calculated) 1514
Sample size per group (calculated) 757 757
Expected total number of events 151.4
Expected number of events per group 75.7 75.7
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .2
Alternative hypothesis H1: pi2 - pi1 < .2
Anticipated event probabilities 0.250 0.250
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (designed) 0.900
Total sample size (calculated) 198
Sample size per group (calculated) 99 99
Expected total number of events 49.5
Expected number of events per group 24.75 24.75
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .15
Alternative hypothesis H1: pi2 - pi1 < .15
Anticipated event probabilities 0.200 0.200
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (designed) 0.900
Total sample size (calculated) 300
Sample size per group (calculated) 150 150
Expected total number of events 60
Expected number of events per group 30 30
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .05
Alternative hypothesis H1: pi2 - pi1 < .05
Anticipated event probabilities 0.150 0.150
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (designed) 0.900
Total sample size (calculated) 2144
Sample size per group (calculated) 1072 1072
Expected total number of events 321.6
Expected number of events per group 160.8 160.8
------------------------------------------------------------------------------
Estimated sample size for two-sample comparison of proportions (non-inferiority)
Null hypothesis: p2 - p1 ³ delta (inferior), where:
pi (entered in command) is the overall proportion of participants
expected to experience the outcome if the treatments
are non-inferior, and delta is the smallest change in
proportions between groups (p2 - p1) which would still
be clinically important.
Alternative hypothesis: p2 - p1 < delta (non-inferior)
Note: a non-inferiority analysis is one-sided.
Assumptions:
power = 0.9000
alpha = 0.0500 (one-sided)
pi = 0.0500
delta = 0.0500
estimated required sample size (per group) = 326
The total estimated sample size is 652
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .05
Alternative hypothesis H1: pi2 - pi1 < .05
Anticipated event probabilities 0.050 0.050
Alpha 0.100 (two-sided)
(taken as .05 one-sided)
Power (designed) 0.900
Total sample size (calculated) 652
Sample size per group (calculated) 326 326
Expected total number of events 32.6
Expected number of events per group 16.3 16.3
------------------------------------------------------------------------------
Estimated sample size for two-sample comparison of proportions (non-inferiority)
Null hypothesis: p2 - p1 ³ delta (inferior), where:
pi (entered in command) is the overall proportion of participants
expected to experience the outcome if the treatments
are non-inferior, and delta is the smallest change in
proportions between groups (p2 - p1) which would still
be clinically important.
Alternative hypothesis: p2 - p1 < delta (non-inferior)
Note: a non-inferiority analysis is one-sided.
Assumptions:
power = 0.8000
alpha = 0.0500 (one-sided)
pi = 0.2000
delta = 0.1000
estimated required sample size (per group) = 198
The total estimated sample size is 396
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .1
Alternative hypothesis H1: pi2 - pi1 < .1
Anticipated event probabilities 0.200 0.200
Alpha 0.100 (two-sided)
(taken as .05 one-sided)
Power (designed) 0.800
Total sample size (calculated) 396
Sample size per group (calculated) 198 198
Expected total number of events 79.2
Expected number of events per group 39.6 39.6
------------------------------------------------------------------------------
Estimated sample size for two-sample comparison of proportions (non-inferiority)
Null hypothesis: p2 - p1 ³ delta (inferior), where:
pi (entered in command) is the overall proportion of participants
expected to experience the outcome if the treatments
are non-inferior, and delta is the smallest change in
proportions between groups (p2 - p1) which would still
be clinically important.
Alternative hypothesis: p2 - p1 < delta (non-inferior)
Note: a non-inferiority analysis is one-sided.
Assumptions:
power = 0.7000
alpha = 0.0250 (one-sided)
pi = 0.3000
delta = 0.1000
estimated required sample size (per group) = 260
The total estimated sample size is 520
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .1
Alternative hypothesis H1: pi2 - pi1 < .1
Anticipated event probabilities 0.300 0.300
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (designed) 0.700
Total sample size (calculated) 520
Sample size per group (calculated) 260 260
Expected total number of events 156
Expected number of events per group 78 78
------------------------------------------------------------------------------
Estimated sample size for two-sample comparison of proportions (non-inferiority)
Null hypothesis: p2 - p1 ³ delta (inferior), where:
pi (entered in command) is the overall proportion of participants
expected to experience the outcome if the treatments
are non-inferior, and delta is the smallest change in
proportions between groups (p2 - p1) which would still
be clinically important.
Alternative hypothesis: p2 - p1 < delta (non-inferior)
Note: a non-inferiority analysis is one-sided.
Assumptions:
power = 0.9000
alpha = 0.0250 (one-sided)
pi = 0.5000
delta = 0.3000
estimated required sample size (per group) = 59
The total estimated sample size is 118
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .3
Alternative hypothesis H1: pi2 - pi1 < .3
Anticipated event probabilities 0.500 0.500
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (designed) 0.900
Total sample size (calculated) 118
Sample size per group (calculated) 59 59
Expected total number of events 59
Expected number of events per group 29.5 29.5
------------------------------------------------------------------------------
Estimated sample size for two-sample comparison of proportions (non-inferiority)
Null hypothesis: p2 - p1 ³ delta (inferior), where:
pi (entered in command) is the overall proportion of participants
expected to experience the outcome if the treatments
are non-inferior, and delta is the smallest change in
proportions between groups (p2 - p1) which would still
be clinically important.
Alternative hypothesis: p2 - p1 < delta (non-inferior)
Note: a non-inferiority analysis is one-sided.
Assumptions:
power = 0.8000
alpha = 0.0500 (one-sided)
pi = 0.6000
delta = 0.0500
estimated required sample size (per group) = 1188
The total estimated sample size is 2376
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .05
Alternative hypothesis H1: pi2 - pi1 < .05
Anticipated event probabilities 0.600 0.600
Alpha 0.100 (two-sided)
(taken as .05 one-sided)
Power (designed) 0.800
Total sample size (calculated) 2376
Sample size per group (calculated) 1188 1188
Expected total number of events 1425.6
Expected number of events per group 712.8 712.8
------------------------------------------------------------------------------
Estimated sample size for two-sample comparison of proportions (non-inferiority)
Null hypothesis: p2 - p1 ³ delta (inferior), where:
pi (entered in command) is the overall proportion of participants
expected to experience the outcome if the treatments
are non-inferior, and delta is the smallest change in
proportions between groups (p2 - p1) which would still
be clinically important.
Alternative hypothesis: p2 - p1 < delta (non-inferior)
Note: a non-inferiority analysis is one-sided.
Assumptions:
power = 0.7000
alpha = 0.0250 (one-sided)
pi = 0.8000
delta = 0.1000
estimated required sample size (per group) = 198
The total estimated sample size is 396
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .1
Alternative hypothesis H1: pi2 - pi1 < .1
Anticipated event probabilities 0.800 0.800
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (designed) 0.700
Total sample size (calculated) 396
Sample size per group (calculated) 198 198
Expected total number of events 316.8
Expected number of events per group 158.4 158.4
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .2
Alternative hypothesis H1: pi2 - pi1 < .2
Anticipated event probabilities 0.100 0.100
Alpha 0.100 (two-sided)
(taken as .05 one-sided)
Power (calculated) 0.903
Total sample size (designed) 78
Sample size per group (designed) 39 39
Expected total number of events 7.8
Expected number of events per group 3.9 3.9
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .05
Alternative hypothesis H1: pi2 - pi1 < .05
Anticipated event probabilities 0.300 0.300
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (calculated) 0.900
Total sample size (designed) 3532
Sample size per group (designed) 1766 1766
Expected total number of events 1059.6
Expected number of events per group 529.8 529.8
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .15
Alternative hypothesis H1: pi2 - pi1 < .15
Anticipated event probabilities 0.150 0.150
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (calculated) 0.902
Total sample size (designed) 240
Sample size per group (designed) 120 120
Expected total number of events 36
Expected number of events per group 18 18
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .1
Alternative hypothesis H1: pi2 - pi1 < .1
Anticipated event probabilities 0.200 0.200
Alpha 0.200 (two-sided)
(taken as .1 one-sided)
Power (calculated) 0.802
Total sample size (designed) 290
Sample size per group (designed) 145 145
Expected total number of events 58
Expected number of events per group 29 29
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .05
Alternative hypothesis H1: pi2 - pi1 < .05
Anticipated event probabilities 0.100 0.100
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (calculated) 0.900
Total sample size (designed) 1514
Sample size per group (designed) 757 757
Expected total number of events 151.4
Expected number of events per group 75.7 75.7
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .2
Alternative hypothesis H1: pi2 - pi1 < .2
Anticipated event probabilities 0.250 0.250
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (calculated) 0.901
Total sample size (designed) 198
Sample size per group (designed) 99 99
Expected total number of events 49.5
Expected number of events per group 24.75 24.75
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .15
Alternative hypothesis H1: pi2 - pi1 < .15
Anticipated event probabilities 0.200 0.200
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (calculated) 0.901
Total sample size (designed) 300
Sample size per group (designed) 150 150
Expected total number of events 60
Expected number of events per group 30 30
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .05
Alternative hypothesis H1: pi2 - pi1 < .05
Anticipated event probabilities 0.150 0.150
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (calculated) 0.900
Total sample size (designed) 2144
Sample size per group (designed) 1072 1072
Expected total number of events 321.6
Expected number of events per group 160.8 160.8
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial substantial-superiority
Number of groups 2
Favourable/unfavourable outcome favourable
Inferred by the program
Allocation ratio 1:3
Statistical test assumed unconditional comparison of 2
binomial proportions
using the score test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 <= .15
Alternative hypothesis H1: pi2 - pi1 > .15
Anticipated event probabilities 0.200 0.500
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (designed) 0.800
Total sample size (calculated) 391
Sample size per group (calculated) 98 293
Expected total number of events 166.1
Expected number of events per group 19.6 146.5
------------------------------------------------------------------------------
+----------------------------------------------------------------------------+
| Sample size calculation for non-inferiority trials using a risk difference |
+----------------------------------------------------------------------------+
p1 = 0.7
p2 = 0.9
delta = 0.2
alpha = .025 (one-sided)
power = .9
allocation = 1:1
not assessable = 0
Sample size using observed values (approximate):
Group 1: 20
Group 2: 20
Total: 40
Sample size using maximum likelihood estimation (more accurate):
Group 1: 23
Group 2: 23
Total: 46
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .2
Alternative hypothesis H1: pi2 - pi1 < .2
Algorithm used 2-arm
Anticipated event probabilities 0.300 0.100
Alpha 0.025 (one-sided)
Power (designed) 0.900
Total sample size (calculated) 40
Sample size per group (calculated) 20 20
Expected total number of events 8
Expected number of events per group 6 2
------------------------------------------------------------------------------
+----------------------------------------------------------------------------+
| Sample size calculation for non-inferiority trials using a risk difference |
+----------------------------------------------------------------------------+
p1 = 0.75
p2 = 0.85
delta = 0.1
alpha = .025 (one-sided)
power = .9
allocation = 1:1
not assessable = 0
Sample size using observed values (approximate):
Group 1: 83
Group 2: 83
Total: 166
Sample size using maximum likelihood estimation (more accurate):
Group 1: 86
Group 2: 86
Total: 172
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .1
Alternative hypothesis H1: pi2 - pi1 < .1
Algorithm used 2-arm
Anticipated event probabilities 0.250 0.150
Alpha 0.025 (one-sided)
Power (designed) 0.900
Total sample size (calculated) 166
Sample size per group (calculated) 83 83
Expected total number of events 33.2
Expected number of events per group 20.75 12.45
------------------------------------------------------------------------------
+----------------------------------------------------------------------------+
| Sample size calculation for non-inferiority trials using a risk difference |
+----------------------------------------------------------------------------+
p1 = 0.8
p2 = 0.7
delta = 0.15
alpha = .05 (one-sided)
power = .9
allocation = 1:1
not assessable = 0
Sample size using observed values (approximate):
Group 1: 1268
Group 2: 1268
Total: 2536
Sample size using maximum likelihood estimation (more accurate):
Group 1: 1265
Group 2: 1265
Total: 2530
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .15
Alternative hypothesis H1: pi2 - pi1 < .15
Algorithm used 2-arm
Anticipated event probabilities 0.200 0.300
Alpha 0.050 (one-sided)
Power (designed) 0.900
Total sample size (calculated) 2536
Sample size per group (calculated) 1268 1268
Expected total number of events 634
Expected number of events per group 253.6 380.4
------------------------------------------------------------------------------
+----------------------------------------------------------------------------+
| Sample size calculation for non-inferiority trials using a risk difference |
+----------------------------------------------------------------------------+
p1 = 0.85
p2 = 0.8
delta = 0.1
alpha = .025 (one-sided)
power = .9
allocation = 1:1
not assessable = 0
Sample size using observed values (approximate):
Group 1: 1209
Group 2: 1209
Total: 2418
Sample size using maximum likelihood estimation (more accurate):
Group 1: 1217
Group 2: 1217
Total: 2434
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .1
Alternative hypothesis H1: pi2 - pi1 < .1
Algorithm used 2-arm
Anticipated event probabilities 0.150 0.200
Alpha 0.025 (one-sided)
Power (designed) 0.900
Total sample size (calculated) 2418
Sample size per group (calculated) 1209 1209
Expected total number of events 423.15
Expected number of events per group 181.35 241.8
------------------------------------------------------------------------------
+----------------------------------------------------------------------------+
| Sample size calculation for non-inferiority trials using a risk difference |
+----------------------------------------------------------------------------+
p1 = 0.9
p2 = 0.9
delta = 0.05
alpha = .05 (one-sided)
power = .9
allocation = 1:1
not assessable = 0
Sample size using observed values (approximate):
Group 1: 617
Group 2: 617
Total: 1234
Sample size using maximum likelihood estimation (more accurate):
Group 1: 631
Group 2: 631
Total: 1262
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .05
Alternative hypothesis H1: pi2 - pi1 < .05
Algorithm used 2-arm
Anticipated event probabilities 0.100 0.100
Alpha 0.050 (one-sided)
Power (designed) 0.900
Total sample size (calculated) 1234
Sample size per group (calculated) 617 617
Expected total number of events 123.4
Expected number of events per group 61.7 61.7
------------------------------------------------------------------------------
+----------------------------------------------------------------------------+
| Sample size calculation for non-inferiority trials using a risk difference |
+----------------------------------------------------------------------------+
p1 = 0.7
p2 = 0.75
delta = 0.15
alpha = .025 (one-sided)
power = .9
allocation = 1:1
not assessable = 0
Sample size using observed values (approximate):
Group 1: 105
Group 2: 105
Total: 210
Sample size using maximum likelihood estimation (more accurate):
Group 1: 105
Group 2: 105
Total: 210
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio equal group sizes
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .15
Alternative hypothesis H1: pi2 - pi1 < .15
Algorithm used 2-arm
Anticipated event probabilities 0.300 0.250
Alpha 0.025 (one-sided)
Power (designed) 0.900
Total sample size (calculated) 210
Sample size per group (calculated) 105 105
Expected total number of events 57.75
Expected number of events per group 31.5 26.25
------------------------------------------------------------------------------
+----------------------------------------------------------------------------+
| Sample size calculation for non-inferiority trials using a risk difference |
+----------------------------------------------------------------------------+
p1 = 0.7
p2 = 0.9
delta = 0.2
alpha = .025 (one-sided)
power = .9
allocation = 1:2
not assessable = 0
Sample size using observed values (approximate):
Group 1: 17
Group 2: 34
Total: 51
Sample size using maximum likelihood estimation (more accurate):
Group 1: 15
Group 2: 30
Total: 45
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio 1:2
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .2
Alternative hypothesis H1: pi2 - pi1 < .2
Algorithm used 2-arm
Anticipated event probabilities 0.300 0.100
Alpha 0.025 (one-sided)
Power (designed) 0.900
Total sample size (calculated) 51
Sample size per group (calculated) 17 34
Expected total number of events 8.5
Expected number of events per group 5.1 3.4
------------------------------------------------------------------------------
+----------------------------------------------------------------------------+
| Sample size calculation for non-inferiority trials using a risk difference |
+----------------------------------------------------------------------------+
p1 = 0.75
p2 = 0.85
delta = 0.1
alpha = .025 (one-sided)
power = .9
allocation = 1:3
not assessable = 0
Sample size using observed values (approximate):
Group 1: 61
Group 2: 182
Total: 243
Sample size using maximum likelihood estimation (more accurate):
Group 1: 50
Group 2: 149
Total: 199
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio 1:3
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .1
Alternative hypothesis H1: pi2 - pi1 < .1
Algorithm used 2-arm
Anticipated event probabilities 0.250 0.150
Alpha 0.025 (one-sided)
Power (designed) 0.900
Total sample size (calculated) 243
Sample size per group (calculated) 61 182
Expected total number of events 42.55
Expected number of events per group 15.25 27.3
------------------------------------------------------------------------------
+----------------------------------------------------------------------------+
| Sample size calculation for non-inferiority trials using a risk difference |
+----------------------------------------------------------------------------+
p1 = 0.8
p2 = 0.7
delta = 0.15
alpha = .05 (one-sided)
power = .9
allocation = 1:4
not assessable = 0
Sample size using observed values (approximate):
Group 1: 728
Group 2: 2912
Total: 3640
Sample size using maximum likelihood estimation (more accurate):
Group 1: 687
Group 2: 2745
Total: 3432
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio 1:4
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .15
Alternative hypothesis H1: pi2 - pi1 < .15
Algorithm used 2-arm
Anticipated event probabilities 0.200 0.300
Alpha 0.050 (one-sided)
Power (designed) 0.900
Total sample size (calculated) 3640
Sample size per group (calculated) 728 2912
Expected total number of events 1019.2
Expected number of events per group 145.6 873.6
------------------------------------------------------------------------------
+----------------------------------------------------------------------------+
| Sample size calculation for non-inferiority trials using a risk difference |
+----------------------------------------------------------------------------+
p1 = 0.85
p2 = 0.8
delta = 0.1
alpha = .025 (one-sided)
power = .9
allocation = 1:2
not assessable = 0
Sample size using observed values (approximate):
Group 1: 873
Group 2: 1745
Total: 2618
Sample size using maximum likelihood estimation (more accurate):
Group 1: 838
Group 2: 1675
Total: 2513
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio 1:2
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .1
Alternative hypothesis H1: pi2 - pi1 < .1
Algorithm used 2-arm
Anticipated event probabilities 0.150 0.200
Alpha 0.025 (one-sided)
Power (designed) 0.900
Total sample size (calculated) 2618
Sample size per group (calculated) 873 1745
Expected total number of events 479.95
Expected number of events per group 130.95 349
------------------------------------------------------------------------------
+----------------------------------------------------------------------------+
| Sample size calculation for non-inferiority trials using a risk difference |
+----------------------------------------------------------------------------+
p1 = 0.9
p2 = 0.9
delta = 0.05
alpha = .05 (one-sided)
power = .9
allocation = 1:4
not assessable = 0
Sample size using observed values (approximate):
Group 1: 386
Group 2: 1542
Total: 1928
Sample size using maximum likelihood estimation (more accurate):
Group 1: 333
Group 2: 1332
Total: 1665
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio 1:4
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .05
Alternative hypothesis H1: pi2 - pi1 < .05
Algorithm used 2-arm
Anticipated event probabilities 0.100 0.100
Alpha 0.050 (one-sided)
Power (designed) 0.900
Total sample size (calculated) 1928
Sample size per group (calculated) 386 1542
Expected total number of events 192.8
Expected number of events per group 38.6 154.2
------------------------------------------------------------------------------
+----------------------------------------------------------------------------+
| Sample size calculation for non-inferiority trials using a risk difference |
+----------------------------------------------------------------------------+
p1 = 0.7
p2 = 0.75
delta = 0.15
alpha = .025 (one-sided)
power = .9
allocation = 1:3
not assessable = 0
Sample size using observed values (approximate):
Group 1: 72
Group 2: 215
Total: 287
Sample size using maximum likelihood estimation (more accurate):
Group 1: 62
Group 2: 186
Total: 248
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome unfavourable
Inferred by the program
Allocation ratio 1:3
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 >= .15
Alternative hypothesis H1: pi2 - pi1 < .15
Algorithm used 2-arm
Anticipated event probabilities 0.300 0.250
Alpha 0.025 (one-sided)
Power (designed) 0.900
Total sample size (calculated) 287
Sample size per group (calculated) 72 215
Expected total number of events 75.35
Expected number of events per group 21.6 53.75
------------------------------------------------------------------------------
ART - ANALYSIS OF RESOURCES FOR TRIALS (binary version 2.1.0 04jan2024)
------------------------------------------------------------------------------
A sample size program by Abdel Babiker, Patrick Royston, Friederike Barthel,
Ella Marley-Zagar and Ian White
MRC Clinical Trials Unit at UCL, London WC1V 6LJ, UK.
------------------------------------------------------------------------------
Type of trial non-inferiority
Number of groups 2
Favourable/unfavourable outcome favourable
Inferred by the program
Allocation ratio 1:2
Statistical test assumed unconditional comparison of 2
binomial proportions
using the wald test
Local or distant distant
Continuity correction no
Null hypothesis H0: pi2 - pi1 <= -.1
Alternative hypothesis H1: pi2 - pi1 > -.1
Anticipated event probabilities 0.700 0.750
Alpha 0.050 (two-sided)
(taken as .025 one-sided)
Power (designed) 0.800
Loss to follow up assumed 20 %
Total sample size (calculated) 398
Sample size per group (calculated) 133 265
Expected total number of events 233.48
Expected number of events per group 74.48 159
------------------------------------------------------------------------------
*************************************************************
*** ARTBIN HAS PASSED ALL SOFTWARE TESTING POINT 1 *********
*************************************************************