I certainly agree with MacGyver’s take on these questions. Here’s what I would say in addition.
Medicine is about balancing the risks and benefits of our therapies. This balance is influenced by how much uncertainty there is about the risks and benefits of a specific treatment. It is also influenced by the urgency with which we need to act. The value of scientific research is that it reduces uncertainty, allowing us to more effectively balance risks and benefits. The problem with anecdotes, such as this one, is that they appear to reduce this uncertainty, but they really don’t. The exact same kind of experience that made it seem marijuana helped this child has made people believe in the healing power of prayer, Lourdes water, homeopathy, and many other therapies we know can’t and don’t actually work.
So it’s not a simple question of saying, “What would you do if you knew something worked even if it hadn’t been demonstrated scientifically,” because you can’t know it works without scientific evidence. The real question is how do you balance the need to act against the uncertainty about the effects of your actions in each case. Let me illustrate with a couple of examples.
I am often asked about unproven remedies for tear staining in white dogs. This is a benign condition in which a natural pigment in the tears causes the hair under the eye to turn reddish brown. People with show dogs hate this, and they are often willing to try any kind of remedy that promises to fix it. Since none of these have been studied scientifically, we don’t know if they work or if they are safe. I tell people that since the condition is purely cosmetic, the need to act isn’t very great. Balancing this against the great uncertainty about safety and efficacy, I wouldn’t use these remedies.
Another easy example is when a dog comes in having lost a great deal of blood from some acute trauma. The need to act is very great. And there is tremendous evidence supporting the safety and effectiveness of a blood transfusion in this situation, so there is not much uncertainty. It is easy to decide to use the therapy.
A situation like the child you describe is much more ambiguous. Obviously, a high frequency of seizures does not permit an acceptable quality of life, so there is clearly a strong urgency to act. I am not a neurologist nor a physician, so I don’t know anything about the treatment options, or how much uncertainty there is about their effects. Presumably, the best choice would be to begin with the therapies that have the most scientific evidence to support them. If these fail, it would be appropriate to consider therapies with less evidence.
In the case of marijuana, I’m not aware of any evidence to suggest it ought to be a good therapy for seizures, so I don’t know why it was tried in this case. But for the sake of argument, if there are only anecdotes to suggest it might be useful, I would expect the child’s doctors to have a very clear conversation with the parents about the great uncertainty. Unproven remedies don’t just fail to work sometimes, they also sometimes cause real harm. This is the risk we take when we act without good evidence. Sometimes, if the situation is desperate enough, it may be right to take this risk, but it should always be done with a clear understanding that we are, to a large extent, rolling the dice with our patient’s health.
Finally, if we use such an unproven remedy and the patient improves, do we conclude the remedy works? No, because again this same logic supports every magical therapy out there. Instead, we add the information to our list of anecdotes, which would support more controlled research to find out if it really does predictably help patients with this problem without unacceptable side effects. If the patient is doing well, it is reasonable to consider continuing the therapy with close monitoring for potential harm. But it wouldn’t be appropriate to conclude the therapy is effective and recommend it freely for others.
So there isn’t a simple yes or no answer. It’s a fluid, subtle balancing of risks/benefits/uncertainty that has to be individualized for every patient/problem/therapy.
I hope this helps.