New Immunizations

Many people are often confused when they hear about different types of medical research. There are different terms that are used with different meanings and significances. This is a mini-primer to help you when you read about various reports.

Randomized study: In a randomized study, each participant has an equal chance of getting any of the treatment options in the study. The patient is randomly assigned to one of the options. Usually there are two options in a study (treatment A vs. treatment B or Treatment A vs. placebo), but occasionally there are more than two options (arms) in the study. Randomized studies are preferred because they reduce the chances that the characteristics of the population getting one treatment might be very different from the characteristics of the other population. Such a difference would be a bias that would affect the outcomes. If some one enrolls in a randomized study, s/he should know that there is a 50/50 chance of getting the treatment vs. placebo if a placebo is being used as the second arm. Some times the second arm is no treatment at all ( this was the case in the Women’s Health Initiative Study of HRT).

Double Blind Study: In a double blind study, neither the practitioner nor the patient knows whether or not the patient is getting treatment A, treatment B, or a placebo.  In other words, both the doctor and and the patient are “blind” to what arm of the study s/he is in. This is to neutralize the placebo effect as a reason for improvement and also to neutralize observer bias from the doctor. Both the placebo effect and the observer bias have been shown to falsely affect studies. When possible, randomized double blind studies are considered the gold standard. Sometimes it is impossible to disguise the treatment arms from either the patient or the doctor. When one knows which treatment arm, it is called a single blind study. When both know which arm the patient is in, it is called an open label study.

Placebo effect: Is has been shown that when a person gets a treatment and believes that this treatment might be helpful, it is likely that the person will experience improvement, even if the treatment does not have any biological way og improving the health of the patient. The placebo effect is very powerful. Often it can produce 30-50% improvement. The effect is more than just the the patient imagines that s/he is improving. Investigators have measured real physiologic effects when people receive a placebo. Nor does one have to trick the patient. In one study patients were told that the treatment they were receiving did not have any biologically active ingredient, they still had benefits from the placebo. These are all clear indicators of the power of the mind-body connection. We should not minimize the power of believing that a treatment is helpful. Nevertheless, when one is trying to figure out is a treatment is benefical (especially is the treatment has potential side effects), the placebo effect must be controlled for- usually via a randomized study.

Retrospective Study:In retrospective studies, researchers obtain information about a population that has already received treatment ot has some other characteristic that the researchers want to study. The researchers then analyze the data. Retrospective studies are often the first studied that point researchers in the direction of a new answer. They are inexpensive and quick to do because the data is already in. On the other hand, it is often difficult to compare groups of individuals in retrospective studies so you could be comparing apples to oranges. Retospective studies are rarely conclusive and usually are followed, if felt to be promising, by prospective randomized studies, where individuals are recruited into the study before they receive the treatment.

[Robert Eidus, MD] [Our services] [About Dr. Eidus] [Health Topics] [Research Primer] [Newsletter]