DESIGNING OF TRIALS - POWER CALCULATIONS BASED ON MATCHING AS A SURROGATE FOR RANDOMIZATION

In the design of clinical trials we strive to extract the maximum amou nt of information from a minimum number of patients and yet retain sta tistical validity. Matching has become acknowledged as a useful and va lid surrogate for the randomized trial. By using data from matched tri als along with power calculations we can accurately predict the number of patients required to achieve statistical significance in a related randomized trial.