FDA Factoids

A factoid sounds true, but may not be. The FDA has foisted factoids on us for years.

The FDA requires that drug trials compare the active drug against placebo not only for effectiveness, but also for adverse reactions. Any adverse reaction that occurs on the active drug to an extent greater than a predefined amount (usually 2%) or more often than placebo is required to be part of the package insert, needed to help the consumer make an informed consent to use the product.

Patients are randomized according to underlying demographics and disease status, making the primary outcome (effectiveness vs placebo) a true reflection of reality. However, the adverse reactions do not necessarily reflect reality as the patients are not randomized according to prior life experiences with pharmaceutical products.

For example, a person assigned to the active treatment group might have had dizziness in the past due to a long-forgotten inner ear problem or perhaps had dizziness in response to some other medication in the past. That person is now more likely to experience dizziness during the trial than a person with no prior experience of dizziness. Data is collected concerning hundreds of potential adverse reactions, making it quite likely that the population in the study, usually only a few hundred to a few thousand individuals, will not be completely randomized to past experience with each and every potential adverse reaction. This can lead to a data bias. If the difference between placebo and active drug is quite large, then it is more likely that the difference is real, but a tiny difference can clearly be due to lack of randomization according to the criteria being reported.

Any difference in adverse reactions between placebo and active drug in a study that is not overwhelming should be viewed with suspicion. Many patients on placebo develop adverse reactions, so called nocebo responses. The most common occurring on placebo are GI related (nausea, vomiting, abdominal discomfort, belching, constipation and diarrhea) and CNS related (dizziness, headache, blurred vision, lightheadedness, sleepiness, insomnia). Attribution of cause and effect in the absence of specific randomization to look for cause and effect is pseudo-science and falls into the realm of story-telling and not science. That pharma is required to disclose this information and prohibited from disclosing information concerning serendipitous additional unexpected benefits found with their product as compared to placebo shows the bias of the FDA and is not clearly in the best interest of patients.