Sample size calculation for non-inferiority study

[Jep]

Hello my fellow SAS users, I need some help with power analysis for a non-inferiority study. I have a diagnostic test that has a sensitivity of 83% and specificity of 96%, which is currently used as the gold standard. There is a new diagnostic test, that is less invasive than the gold standard, and we are trying to evaluate it's accuracy and to see if it is not inferior to the gold standard. A previous study evaluated the accuracy of this new diagnostic tool and found it to have a sensitivity of 90% and specificity of 86%. The hypothesis for the study is that, the new diagnostic tool is similar or even better than the current gold standard. What is the best way to calculate the necessary sample size to address this hypothesis? Also if anyone knows a good book, course or any good materials on sample size calculation I would truly appreciate.    

[Jep]

Hello my fellow SAS users,

I need some help with power analysis for a non-inferiority study. I have a diagnostic test that has a sensitivity of 83% and specificity of 96%, which is currently used as the gold standard. There is a new diagnostic test, that is less invasive than the gold standard, and we are trying to evaluate it's accuracy and to see if it is not inferior to the gold standard. A previous study evaluated the accuracy of this new diagnostic tool and found it to have a sensitivity of 90% and specificity of 86%. The hypothesis for the study is that, the new diagnostic tool is similar or even better than the current gold standard. What is the best way to calculate the necessary sample size to address this hypothesis? Also if anyone knows a good book, course or any good materials on sample size calculation I would truly appreciate.