Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Firstly, thanks for answering my last question. I appreciate the help and effort put in.

I have another question to put forth. Again, it is related to Medical Statistics.

There are one third of male prisoners who have a history of heroin injection. One in 200 of them will die from heroin overdose. To prevent plausibly 30% of these deaths, a major trial will randomise individuals to control and intervention.

And we need to calculate how many consented prisoners need to be randomised for the proposed trial to have 80% power to discern a 30% reduction in heroin overdoses within 4 weeks of release for prisoners.

What I have been taught is to do something like this:

Target reduction is 30%. Initial ratio to start with is 0.5% and have to go to 0.35%.

So, (0.5*99.5)+(0.35*99.65) and divide this by 30^2 to get the value of 'n', i.e., the number in each group that should be taken to conduct the test.

The value of n is coming out to be 0.009 something, which I think is wrong. Am I reading the question wrong? How else can I do this question?

Thank you very much again!

share|improve this question
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.