When Should I Make Design Changes?
Posted on November 23, 2011
C. Hendricks Brown (bio) offers advice on the best times to adapt a study's design.
There are a couple of steps in here about when you might make some changes in the design. So, you might be able to specifically go through a number of steps. You might get funded for one particular piece of an intervention that says that you're going to be improving one component of an intervention and adding it on to this one.
So, the question is could you take and use the knowledge that you've already got from your first set of pilot work, for example, and then move it into a next effectiveness study and be able to use that and in combination with adding on another component of an intervention. And is it possible to integrate all three of those kinds of studies -- the pilot study, the effectiveness study, and then maybe an added component of intervention
-- to see whether or not you can combine the results across those different studies.
So this is related to these areas of synthesis of findings across multiple trials and it's another area that we've been interested in from a methologic point of view of how can you theoretically put those together in an efficient manner so that we can combine the results across trials in here, taking into account the changes in the intervention that might take place.
You don't want to just put a bunch of things in at one time. You want to do it very strategically in this one, so one suicide prevention program, for example, that we've had, had a component that was developed for an intervention that was really one year long, for example. So, we've evaluated that for now about 18 schools as we have built up over a period of time in three states the data from a randomized trial from those 18 schools.
But what we found is that we are really needing to look at a two-year program that is built on the exact same model of the first year but extended to that second year. So, the question would be whether or not that second year of program enhances the results or it does it decay over time. So, what we're trying to do is just look at those kinds of changes over time and see whether there is any kind of variations from what we had from original 18 schools.
Excerpted from an interview with the researcher conducted at the 2011 NHSN Conference held in Miami, FL.
Please note that the feedback is viewed only by 4researchers staff and is not intended for communication with individual contributors.
Use the form below to submit feedback about this article. If you would like a response, please be sure to include your e-mail address.