Comparison of ‘Power for Randomized Clinical Trials’ to other public software programs.

 

Date last tested: 7/11/05

 

 

CI%, n1, p1,

n2, p2

Data input

95%, 100, 20%,

200, 10%1

 

95%, 100, 10%,

 200, 20%

95, 474, 45%,

255, 35.29%

95, 474, 35.29%,

255, 45%

 

software

Power

 

Power with continuity correction

Power

 

Power with continuity correction

Power

 

Power with continuity correction

 

Power

 

Power with continuity correction

 

OpenEpi

 

67.23%

60.14%

59.21%

52.41%

71.88%

69.11%

72.91%

70.15%

PEPI

 

65.78%

59.25%

60.09%

52.65%

72.12%

69.32%

72.66%

69.93%

EPI 505

 

67.1%

59.3

59.1%

52.6%

71.7%

69.2%

72.7%

69.8%

Confidence interval CI;  sample size of treatment-1 ‘n1’;  percent with outcome in treatment-1 ‘p1’; sample size of treatment-2 ‘n2’;  percent with outcome in treatment-2 ‘p2’.

 

Reference:

·        PEPI: Abramson JH and Gahlinger PM (2001). Computer program for epidemiologists: PEPI version 4.0. Sagebrush Press, Salt Lake City. Used POWR.EXE version 4.0.

·         Epidemiology 505 - Emory University  (version - 11 October 2000)