A Consumer’s Guide to Research: I. Observational Studies
I spend a lot of time teaching medical students, residents, fellows, and other health care professionals. Since I think it’s critical that we base our clinical practice on the best evidence available, I like to talk about what research evidence shows. This means that at least some of the time I have to explain a little bit about how research studies are done, what kind of evidence they produce, and why some studies are generally considered to give us better or stronger evidence than others.
One of the most important issues that comes up in discussions of research in the media and with patients and clinicians has to do with research design. Not all research is done in the same way, and we can draw certain kinds of conclusions from one research design but not others. We often want to decide whether something either causes disease (for example, smoking) or helps prevent it (for example, does vitamin C prevent colds?). Two basic kinds of research studies deal with these issues in different ways, and give us different kinds of information.
The most common type of study that gives us information about factors associated with disease is the observational study. These studies are very common, and usually involve collecting information from a group people at one point in time, and then collecting the information on other occasions. We might, for example, ask a group of people in the community about their memory, diet, exercise, and use of vitamin supplements. Later, we would ask them the same questions again. Perhaps some of the people would say that their memory had gotten worse. We could then find out if their diet, exercise, or use of vitamins was different from the people who said their memory was OK.
The good thing about this kind of study is it’s easy to do and therefore less expensive than many other kinds of research. The weakness is we can’t control all the things that might affect the study’s outcome.
Suppose we found that people who took vitamins were less likely to say their memory got worse. Would that mean that taking vitamin C is good for your memory? No. Although this kind of study is often interpreted this way, it’s not true. We don’t know whether something else is going on that is related to taking vitamin C but also affects memory.
A good example are the results of observational studies that ask people about taking drugs called non-steroidal anti-inflammatory drugs (NSAIDS). These are medications like ibuprofen; there are many brand names. Studies have shown that people who took NSAIDS for arthritis are less likely to develop Alzheimer’s disease. So far, so good. But when controlled clinical trials of NSAIDS in people with memory problems were done, they didn’t show any benefit.
Observational studies are really useful for helping us understand what may cause a disease, but they usually aren’t very good at nailing down exactly how to treat one.
