Making Sense of Medical Studies
You’ve heard about a medical breakthrough and you want to learn more. But trying to decipher the published study gives you a headache. Learn how to translate the latest medical findings into plain English.
You often hear news reports about medical studies that could affect you or a loved one. But sometimes, these brief reports can leave you with more questions than answers.
Was the study a “clinical trial?”
A clinical trial is a medical research study done with human volunteers. The results of these trials apply more to humans than studies done in the laboratory or that are done on animals. Laboratory and animal studies represent earlier stages of medical research. There are three main types of clinical trials:
- Treatment trials test the safety and efficacy of a new drug, medical device, or procedure (such as surgery).
- Prevention trials determine if a lifestyle change (such as exercise or quitting smoking) or some other intervention lowers the risk of developing a certain disease.
- Diagnostic trials are done to find better tests or procedures for diagnosing disease.
What is a “randomized trial?”
There are several types of studies. The most reliable is the randomized control trial (RCT). An RCT has:
- A treatment group. One group of people receives experimental treatment, like a new medication.
- A control group. The control group doesn’t get the new medication or therapy. Instead, they may receive the standard therapy or a placebo (sugar pill).
- Randomly chosen participants. Neither the subjects nor the researchers choose who goes into which group. Subjects are assigned to the control or treatment group by chance. Randomization helps to eliminate bias, which can skew the results.
An RCT will often have these features:
- It is doubled masked. Neither the subjects nor the researchers know which patients are in which group.
- It has a large sample size. Many people are tested.
- Subjects are followed over a long period of time. This can be for years – or even a lifetime – depending on the study.
What are case-control and cohort studies?
Case-control and cohort studies are a step below the RCT. A case-control study looks for a common link by comparing people who have had a certain disease with people who haven’t. For example, if some passengers on a cruise ship came down with food poisoning and others didn’t, a case-control study would compare the types of foods eaten by each group.
A cohort study looks at people who have been exposed to certain risks and compares them to people who haven’t. The group is followed to see what happens to them in the future. For example, researchers may study children who have been exposed to secondhand smoke and children who haven’t – and later compare the cancer rates between the two groups.
What are some other types of studies?
The evidence obtained in the following studies is not as strong as in the RCT, case-control study, or cohort study.
- Cross-sectional study. Looks at a single point in time.
- Example: Comparison between overweight and thin adults to determine which group more often develops diabetes
- Ecological study. Explores the link between occupation and environment on disease. It may compare disease rates in different populations.
- Example: Comparison of cavity rates among children in two different towns with two different fluoride levels in the water
- Meta-analyses. Instead of working with patients directly, researchers review the medical literature for similar studies, then combine and analyze the results.
- Case reports. Describes the symptoms of a patient for learning. When AIDS first appeared, so did numerous case reports describing its symptoms.
What is “relative” vs. “absolute” risk?
Test results are often explained in terms of “relative” or “absolute” risk. Let’s say a randomized controlled study compared two groups, each with 1,000 subjects that have high cholesterol. The treatment group received a cholesterol-lowering drug and the control group got a placebo. After five years of following each group, 25 people in the treatment group had a heart attack, while 50 people in the control group had a heart attack.
- The relative risk would tell us how much larger or smaller the chance of heart attack is while using the test drug. The relative risk for the placebo group in this clinical trial was 1.5. That means that the chance of someone in the placebo group had a heart attack is 50 percent greater than someone who gets the experimental cholesterol-lowering drug.
- Absolute risk uses whole numbers. In this example, in terms of “absolute” risk, for every 1,000 people not taking the drug (taking the placebo instead), there would probably be 25 more heart attacks.
Some people find absolute risk more useful because it is an actual number as opposed to the relative risk, which is expressed as a ratio.