Map of life expectancy at birth from Global Education Project.

Wednesday, September 10, 2014


One of the difficulties with pharmaceutical regulation is the length of follow-up in clinical trials. Typically, drugs are approved based on trials of no more than six months duration. This is true even for drugs that people may take for a long time, even for many years. Sometimes the FDA requires post-marketing surveillance or longer term follow-up studies after a drug has been approved, but they often don't, and these requirements aren't even enforced in many cases.

Benzodiazepines are very commonly prescribed tranquilizers. Very common brand names are Xanax, Librium, and Ativan. (Rohypnol, also in the class, is a popular date rape drug. Sorry to bring that up.) They are addictive, and are often drugs of abuse. Many people doctor shop for them just as they do for opioids, they are often sold on the street, and they are often found in conjunction with opioids in overdose victims. They are prescribed for anxiety, but as I often say, anybody who isn't paranoid nowadays is nuts.

Well now, it turns out that a carefully done case-control study links even moderate-term consumption of benzos to Alzheimer's disease. By moderate term I mean more than three months worth. By links I mean an elevated odds ratio of 1.84 for more than six months of consumption.

The odds ratio is not the same as the risk ratio -- since math is hard I don't want to digress to explain this here. A case control study does not allow for computation of the actual relative risk in the population, but in this case, we can figure out that it translates to an approximately 50% increased risk. There are other studies that point to the same conclusion; the strength of this study is that it attempts to account for the possibility that people who have symptoms that might be associated with a higher risk of Alzheimer's are more likely to be prescribed benzos. (That's called confounding by indication.)

This is a very disturbing finding. If someone is anxious for a reason, such as the fact that we live in a dangerous world, taking a pill is not going to fix the real problem. If there is in fact a disease that causes people to feel anxious all the time and we don't understand why and can't do anything else about it, having them take a pill for the rest of their lives that causes drowsiness, dizziness, lack of coordination, decreased libido and amnesia -- even without the Alzheimer's disease --- is probably not a good idea. In fact, they are recommended only for short term use -- alcohol withdrawal, termination of seizures, brief (2 - 4 weeks) treatment of anxiety. There is no condition for which long-term use is helpful or indicated. Yet these investigators had no trouble finding 22% of people in a sample from Quebec who had been prescribed benzos for more than six months. (That was the rate of long-term prescribing in the controls, in other words. The rate in the cases was 33%.)

To reassure you, there was no finding of increased risk of Alzheimer's associated with short-term use. That indication is not affected by this study. The point is, doctors should not be doing this.

No comments: