The Skinny on Studies
So far, I’ve been talking about health studies as if they’re all the same—but they aren’t. There are several different kinds of studies, each designed to uncover various kinds of information.
For example, cohort studies follow large groups of people over time, watching to see what happens. The researchers don’t do anything; they simply observe the members of the group, then report what they find. For example, the researchers might track 10,000 people over 20 years, then look at the data and report that those who exercised more than three times a week developed less heart disease. This is very useful information, but it doesn’t prove that exercise reduces the risk of heart disease. It could be that those who exercised more were healthier to begin with, or were more likely to consume healthy diets, which was what really protected them against heart disease. It’s also possible that the link between more exercise and less heart disease was just a statistical coincidence.
Yoga Exercises For Chronic Back Pain Photo Gallery
We doctors prefer to base our treatments on prospective, double-blind studies, which are felt to be the “gold standard” of studies. These studies are
• Prospective, which means that at least two groups of people are followed over time to see how a certain treatment fares. One group receives the treatment, while the other group, called the “control group,” does not. The control group may receive a different treatment, or no treatment at all.
• Double-blind, which means that neither the researchers nor the participants know which group is getting the treatment and which is not. For example, if the study required the participants to take a new medicine, everyone would take pills that looked exactly the same so that no one would know which participants were taking the “real thing.” There are good reasons for this: If the researchers know which participants are getting the treatment, they might be tempted, consciously or subconsciously, to rate their outcomes higher than deserved. And they might do the reverse if they know which participants are in the “control group,” taking the placebo. Meanwhile, the participants are also “blinded” to reduce the influence of the placebo effect, which can cause improvements simply because the participants believe the “medicine” will make them better.
• Randomized, which means that participants are arbitrarily assigned to receive either the treatment or something else. For example, those in the “treatment group” might participate in art therapy led by a trained art therapist, while those in the “control group” might be asked to sit and draw whatever they wish, or listen to a lecture on good health habits. By randomly assigning participants to treatment or control groups, no one can “stack the deck” by putting certain patients into a group more likely to produce a certain result.
Prospective, double-blind, randomized studies published in respected medical journals carry a lot of weight. But a meta-analysis is even heftier. This is a “study of studies,” created when numerous studies on the same topic are combined and, using statistical tools, create a larger study. For example, separate studies with 100 participants, 250 participants, 75 participants, and 300 participants may be combined into a meta-analysis of 725 people. Larger studies are considered more valid, as they’re less likely to be influenced by chance events.
When considering how much weight you should give to the results of a study, find out what kind of study it is. Cohort studies may provide useful information, but only a prospective, double-blind, randomized study puts the treatment to the test. Of course, even these studies have their limits. But ideally, many prospective, double-blind, randomized studies will come to the same conclusion, as shown by a meta-analysis that includes these studies.