Designing Trials Efficiently: Reducing Sample Size: Part 3

How are we going to actually make these trials more efficient? So, there's really four ways to actually do this, and we'll go through them one at a time. The first is to have a more focused and relevant research question. So, the research question we already talked about needs to be one that's worth answering but focusing it on a more homogenous

population, actually allows you to have less variability in the data. So, if you only do a study in people over 65, the data's going to be more homogenous than I do a study that goes from neonates all the way to 95-year-olds. So, the other issue here is that it's really important to calculate

a sample size after you come up with the research question, not start off with the sample size and back calculate to the research question. Because what happens when people do that is you end up usually overestimating the effect size, coming up with the sample size that's too small and you can't really answer the question to begin with.

The second thing you could do is change those ingredients of the sample size. Remember, I said there's a Type 1 and a Type 2 error, that is not suggested that you actually do this. But if you allow more error in the study, you will get away with a smaller sample size. But when you think about it, a p-value of 0.05

or a Type 1 error 0.05 means you're wrong 1 in 20 times. If you don't think that's a lot, go on to clinicaltrials.gov and see how many thousands and thousands and thousands of studies there are out there. 1 in 20 is a lot. If you are wrong that many times, we'd be really getting things wrong.

The other thing I joke with the fellows is, if you don't think patients care about an error rate that high, just hang a sign on your office door that says, "Guess what? I'm wrong 1 in 20 times, " and watch people run out of the waiting room. That is a pretty high level of error.