Understanding study designs for nurses who ‘don’t do’ epidemiology

Nova Corcoran,  University of South Wales.

It’s that word ‘epidemiology’. Possibly it makes people think another other unpleasant word – ‘statistics’. Usually it takes you back to being a nursing student again and those research methods lectures that were full of words like p-values and confidence intervals.  In practice, nurses (and other health care practitioners) often lack confidence in their ability to understand statistical studies either because they never really understood research or simply just because they have not set foot in a classroom for years.  Epidemiological studies inform nursing practice on a day to day basis from what dose of drug to give, to the best ways to care for a patient. So perhaps it may be time to revisit some of the basics of epidemiology, and you don’t need a calculator to do it.

This post will introduce three common study designs; the randomised control trial (RCT), the case control study and the cohort study. From years of experience in teaching evidence based practice, the first common problem is that no-one can remember which study is which – hence the purpose of this post.  So I am going to try and explain these three study designs in their most simplistic forms:

a)     A RCT. This involves two or more groups, one of who is randomised to an intervention (i.e. a drug), and another who receives either nothing or a placebo (i.e. a sugar tablet). They are the control group. Both groups are measured at the start and end of the study to see if the intervention (i.e. drug) or placebo has any effect on the chosen study outcome (i.e. do they develop orange spots). This is considered the gold standard of research, so most drug trials and vaccines have been tested by a RCT.

b)     A case control study. This usually follows two groups of people, one who has something interesting like orange spots on their faces (cases), and one group who does not have the interesting thing (i.e. they have no orange spots). This study is usually retrospective, which means it looks backwards, or into the past, to try and work out why one group has the orange spots, and why the other group doesn’t.

c)     A cohort study. This follows a group of people over time to see what happens to them. This means it is prospective, so it follows people over a set period i.e. 10 years. The group of people usually don’t have what the study wants to investigate yet (i.e. no orange spots), but they have been exposed to something which might mean they develop something interesting later (maybe orange spots or even purple spots).

Now remember, this is the simplistic version. In practice you will find many anomalies such as more than two groups in studies or different sampling techniques (basically how people ended up in the intervention groups).  The best way to get to grips with these three study designs is to look for them in nursing practice journals and have a read of the articles so you can see how they work in practice.  For example in the evidence based nursing journal last month 17 (1) 2014 you will find examples from all three of these studies, and they are all clearly labelled as to what type of study to make your life a little easier. Have a read and if you understand the study design then you can give yourself a pat on the back as you now know how these studies were designed.


(Visited 750 times, 1 visits today)