By: Thomas Pickering, MD, DPhil, FRCP, Director of Integrative and Behavioral Cardiology Program
of the Cardiovascular Institute at Mount Sinai School of Medicine, New York.

The gold standard for evaluating any treatment is the randomized controlled clinical trial. If a pharmaceutical company has a new blood pressure-lowering medication it must first prove that it is effective. The simplest way of doing this would be to measure the blood pressure of a group of patients, give them the medication, and then measure their pressure again to see if it goes down. The problem with this approach is that blood pressure tends to be very variable; if you take a group of patients and simply go on measuring the blood pressure, it tends to go down even in the absence of any treatment. To distinguish the effects of the drug from any spontaneous changes of blood pressure clinical trials must include a Control group.

How does this work? Participants in the study are divided into two groups, only one of which is given the treatment. The group getting the drug being studied is called the Active Treatment group, and the other group the Control group. Blood pressure is measured in both groups, so that the effects of the drug can be compared with the changes occurring in the untreated Control group. It is essential that the only difference between the two groups is that one is taking the drug and the other isn't. If one group has higher blood pressure to start with than the other, the results will not be able to be interpreted. To ensure that they are matched, each participant is randomly allocated (that is, by chance) to one of the two groups. If there is a sufficient number of patients in each group, any differences between individuals should even out between the two groups and changes in blood pressure can be attributed to the treatment.

The trial may be 'double-blinded', in which case neither the doctor nor the patient knows until the end of the trial which type of treatment the patient is receiving, or 'open', in which case both parties know from the start. Most of the very large trials we describe are 'open'.