There are really only two main ways to determine if a drug does what it is supposed to do. One way is to give the drug to people and observe what happens. This method has been used for as long as we have taken drugs. Naturopathy and practices such as traditional Chinese and Ayurvedic medicine are based on centuries of observations recorded by practitioners. Observation remains an important component of modern, Western medicine.

There are problems with relying on observation alone. Observers may be biased. Patients taking the drug may be biased. There is no way to be certain that it was the drug and not something else that caused an observed effect or lack of effect. For example, we may give a drug to people with headaches and find that most report reduction in headache pain. But what if most of these people were also getting acupuncture or massage or on special diets? Was it the drug or some other treatment or maybe a combination of the two that led to reduced pain? Perhaps factors such as age, sex or the presence of other medical conditions might be important in a particular drug’s effect. Observation alone may not reveal this.

In order to deal with such problems, we need to go beyond simple observation and start testing drugs. This involves making sure that all those taking the test drug are either getting no other treatment or are all getting the same treatment. It also involves comparing the responses to a test drug with responses to a placebo (something with no physiological effect). Making sure that observers and patients don’t know whether the drug or the placebo was taken can eliminate bias. The patients should be matched in terms of age, sex, general health and use of other drugs that may influence results.

Testing drugs in this way allows greater certainty in saying that our hypothetical headache drug, and not something else, is responsible for reducing headache pain. Currently, any pharmaceutical product sold in Canada must undergo such testing. Unfortunately, many remedies that are not classified as pharmaceuticals can be marketed without testing. When you use such remedies therefore, the available information about the remedy usually comes from observation alone. As I’ve just argued, this information is inherently less reliable. Therefore, whatever results you may get, you will be less certain that the remedy was indeed responsible for those results.

I believe that before you ingest any substance you should always ask, “what is the evidence for the usefulness of that substance?”. Is that evidence based on opinion, observation or testing? Evidence based on testing is the most reliable we have and I urge you to keep this in mind before deciding what to put in your body.