What You Need to Know About Clinical Trials

Great news! You can drink coffee again. It really is good for you. It provides antioxidants and gives you energy. It may even cut the rate of liver cancer and lower your risk of developing Type 2 diabetes. But wait. It wasn't too long ago that studies were showing coffee wasn't good for you . Drinking coffee was associated with interfering with your body's ability to keep homocysteine and cholesterol levels in check and with an increased risk of stroke and rheumatoid arthritis. But, hey, that was yesterday. Today, it's good for you. Tomorrow, who knows?

Every day consumers are bombarded with a confusing variety of new medical studies , which often contradict each other. Currently, there are approximately 50,000 clinical trials going on in the United States. Often, instead of giving us important information about new medications, procedures or nutritional information, these studies end up leaving us with a myriad of questions. Is a low-fat diet the best, or would I do better on a low-carbohydrate one? Is this new drug for high blood pressure the right one for me, or is the one I'm taking better? It's hard to sort through all that information to come up with the right answer.

What information can you trust?

All clinical trials are not equal, and learning how to identify those that offer significant and trustworthy findings can be difficult. In fact, some supposed clinical trials are actually just case series, which are results that come from a review of patients who have followed a common treatment plan and are not part of any standardized study.

If you want to find out if the clinical trial results you are hearing about are valid, here are a few questions you should ask:

• Who sponsored the study?
  Often, clinical trials are sponsored by the manufacturer of the drug or device being used in the study. This is true even of the studies that come out of teaching institutions. There can be a conflict of interest when this happens, because negative results may end up not being published. A May 2004 study of drug trials reported in the Journal of the American Medical Association (JAMA) found that 65 percent of findings on harmful effects from the medication weren't completely reported.
• Is it a randomized control trial?
  The very best trials are those that are multi-sited, fully-blinded, randomized control studies. Plainly stated, the study is carried out at several sites, no one but the researchers who have set up the trial know who is getting what treatment, and subjects are randomly distributed (by age, sex, other medical conditions, etc.) into groups which are either subjected to the experimental procedure or serve as the control group.
• How many people were studied?
  If you have a randomized control trial for a condition such as coronary artery disease, and the trial only involves 50 people, the findings will not be very significant at all. However, if you have a rare condition that affects only 200 people a year and the trial involves 20 people, then the results may be very significant.
• Who was studied?
  If a trial consisted of studying women between the ages of 18 and 40 and you are a 65-year-old male, you cannot assume that the results will be applicable to you.
• What is the probability that the findings occurred by chance?
  You want that probability to be low, less than 5 percent. You can find out the probability by looking at the "P Value" and seeing if it is less than 0.05.
• Are the results significant and of any clinical benefit?
  If less than 7 percent of the people see their cholesterol levels decrease by 20 percent, is that really of any significance? It may be, depending on other factors such as the risks of taking the drug. Also ask how much of an improvement was seen. For example, one study found that people with poor blood circulation, who for three weeks wore specially-designed trousers, were able to exercise more than their counterparts. How much longer? Less than a minute. That may be scientifically significant, but it isn't of any genuine benefit to patients.
• Where was the study conducted?
  The countries known for having the best studies are Great Britain, the United States, Australia, Japan, France and Germany. U.S. studies are often sponsored by the manufacturer, so check carefully who is funding the trial.
• Do the results match up with other similar studies?
  If radical new findings aren't substantiated by other trials, be even more vigilant in asking all of the previous questions.

The media doesn't always get it right either

Just because you read about a new clinical trial in your newspaper or hear about it on the nightly news, don't assume you've gotten the full report. Researchers from Harvard University, the University of California and the University of Newcastle, reviewed media reports of medical developments and reported in the June 2000 New England Journal of Medicine that journalists tended not to mention possible side effects of treatments. In addition, they found that journalists didn't usually disclose who was funding the study, and they often used information from the drug company's own press releases, which tended to distort research findings.

Still confused?

Talk to your health care professional. Ask your doctor what he or she knows about the study and where you can get more information. If you believe the results of a clinical trial may be beneficial to you, ask your physician to review the data and make recommendations that are right for you.

For Further Information You May Consult the Following:

• The National Institutes of Health sponsors http://www.clinicaltrials.gov/
• CenterWatch by Thomson Corporation can be found at http://www.centerwatch.com/patient/trials.html
• The Office of Rare Diseases at the NIH sponsors: http://ord.aspensys.com/asp/resources/rsch_trials.asp