Essentials of Nutrition for Sports

In nutrition, medicine and science, when a new diet, drug, or

technique is introduced, initial studies may not be blinded. Treatment effects diminish as the studies go from unblinded to blind to double-blind. Randomized

Randomized

means that subjects are assigned to groups at

random, so that each subject’s treatment is a matter of chance, like flipping a coin. This reduces both conscious and unconscious bias. It helps insure that the groups will be similar not only with respect to factors that are known to effect the outcome, but also to unanticipated or even unknown factors that might influence the outcome had purposeful assignments been used.

When significant differences exist between groups, it not only

calls into question the randomization process, but the rest of the study as well.

Subjects in randomized controlled trials may be selected with

bias. They may be different from the general population to which the conclusions of the study may be directed. Results may not be applicable to subjects and settings different from those in the study.

Asking male endurance cyclists to volunteer for bone density

tests to determine the risk of osteoporosis in male endurance cyclists may sound reasonable on the face of it. This process may not result in a group representative of male endurance cyclists. The volunteer group may be made up of riders

who have reason to worry about

their bones and so volunteer to find out their status.

Take heed when you see a note at the end of research articles

that states: “These results may be due to the particular type of subjects seen in this setting, and may not be applicable to other groups.”

Controlled

Placebo controlled

means that the study involves two

treatments—the treatment under investigation and an ineffective control (a placebo, a “sugar pill”) to which the new treatment is compared. Both groups must be treated the same—except for administration of the treatment.

If the two groups have different outcomes, then there is evidence

that it is due to the single treatment by which the two groups differ.

A control group may seem like a waste of resources. However, a

treatment is not just taking some substance or following a particular exercise program. Placebo effects are often strong, and include: contact with investigators, heightened awareness of the problem, and any changes treatments might produce.

Be alert to: “Subjects were used as their own controls.” Subjects can and often should be used as their own controls if all

treatments can be administered simultaneously (for example, creams A and B randomly assigned to the right and left arms).

“Subjects as their own controls” often means testing a subject,

administering a treatment, and testing the subject again. The difference in measurements may be called the treatment effect.

This can be disastrous. Changes due to the treatment are

confounded with changes that o

ccur over time. We cannot know

without a control group.

To appreciate the danger of a lack of a control group, look at the

results of the control group when you read about placebo-controlled trials. Systematic Reviews & Meta-Analyses

A systematic review is a comprehensive examination of a topic.

Prior studies are identified, evaluated, and summarized according to a specific methodology.

Nutrition for Sports, Essentials of 137