The news this week about Prozac and similar antidepressants is... well... depressing. A new analysis of old data has found that these drugs don't work. How can this be? Prozac alone has been taken by over 50 million people and the drug has been in use for over 20 years. The story has attracted a great deal of comment from many angles. In fact, this story has more angles than a geometry problem. My angle is diversity and uncertainty. (OK, maybe that's two angles!)

---

The report that has caused all the fuss is available online.^[1] I liked the story about it in The Guardian,^[2] which included opinions from people who have been helped by antidepressants as well as those whose lives have been damaged. It's difficult to arrive at objective truth but one of the issues here seems to be variable response. This is not a new issue. It has been known for decades that different people respond to drugs in different ways. Many drugs in common use are effective for only a fraction of the people for whom they are prescribed. A 50% response rate is considered good. At the other end of the scale some people have adverse reactions to drugs that benefit the majority.

An interesting study, reported in the British Medical Journal^[3] a few years ago, found that 5% of hospital admissions were due to adverse drug reactions. These were not cases of drug abuse. In each case, the drug had been properly prescribed by a doctor and used as directed. And the drug that was most commonly implicated in this study was not one of the infamous nasties, but low-dose aspirin. Of course, this doesn't mean that doctors should stop prescribing low-dose aspirin. There is convincing evidence that it has significant long-term benfits for many people. But perhaps more care needs to be taken over who should receive it and who should not?

Big Pharma

Many people are quick to blame the pharmaceutical industry but, in my opinion, legislators and medical professionals are best placed to improve the situation. Yes, Big Pharma has questions to answer but Big Pharma is not Big Tobacco. One of the issues in the antidepressant story is that researchers used Freedom of Information legislation to obtain unpublished data about the drugs. The reluctance of drug companies to talk about unflattering research results could be mitigated by legislation: no drug marketing approval unless the studies supporting the application were registered up front.

What about the issue of variable response? It's very difficult to find published data on the subject. It's only to be expected that drug companies want to encourage the perception that a drug works for everyone. They know that's what people want to hear and it's good for profit. Indirectly, it's also good for most patients. What people believe has a significant effect on physical outcomes. This is the placebo effect. The people it's not good for are those who are unusually responsive to the drug, and suffer adverse reactions.

The pharmaceutical industry knows fine that people respond to a drug in different ways. That's why it is spending large amounts of money on pharmacogenetics and tailoring drugs to specific segments of the population. It is only now, when drug companies need to sell the idea of pharmacogenetics, that they start to talk publicly about variable response rates. If we want the industry to be more open about variable response, we need to insist. The medical profession should demand the information in order to be in a position to share it with patients. If necessary, legislation. The first responsibility of the industry is to shareholders, not patients.

And Then?

If we get all this information, what are we going to do with it? Panic? Depression? As well as persuading or forcing Big Pharma to release its secrets, we need to learn how to cope with uncertainty. If we are to make sense of information about drugs that don't work and adverse drug reactions, we must be able to balance benefit against harm. We must also learn how to balance the needs of society with the needs of the individual. Perhaps it's easier not to know? At least we can just blame it all on Big Pharma!

References

1. The antidepressants study by Kirsch et al.
2. The antidepressants story as reported in The Guardian
3. Hospital admissions due to adverse drug reactions as reported in the BMJ