Welcome to Data Based Medicine

January, 12, 2012 | 19 Comments

Share this:

Adverse drug events are now the fourth leading cause of death in hospitals

It’s a reasonable bet they are an even greater cause of death in non-hospital settings where there is no one to monitor things going wrong and no one to intervene to save a life. In mental health for instance drug-induced problems are the leading cause of death — and these deaths happen in community rather than hospital settings.

There is also another drug crisis — we are failing to discover new drugs.

Companies are sawing off the branch on which they are sitting. Doctors are failing to recognize the most treatable cause of death today.

These two crises may be linked in that detecting adverse events on drugs is still the best way to discover a new use for a drug, and new drugs. But there are fewer and fewer incentives for anyone to recognize adverse events. Companies are blocking efforts to detect problems and in so doing are sawing off the branch on which they are sitting. Doctors are neutering themselves by failing to recognize and treat what should be the most recognizable threat to life and moreover the most eminently treatable cause of death in the world today.

Part of the problem is cultural

A century ago Freud drew our attention to the many ways in which speech could be biased. Half a century ago clinical trials drew our attention to the biases that both doctors and patients bring to therapy. Just as Freud’s insights once made it difficult for anyone to accept things that were said at face value, so clinical trials and evidence-based medicine have created a culture that makes it increasingly difficult for doctors or patients to spot what is right in front of our own eyes. Ultimately Freud ended up being used to explain away or deny claims of abuse that we now know were happening, and in much the same way companies and doctors are now using trial data, or the lack of it, as a drunk uses a lamppost — for support rather than illumination. Just as a point came at which claims of abuse could no longer be denied, we may be nearing a point where treatment-induced problems will have to be recognized.

This blog aims at raising the profile of this interlocked set of problems and the need for Data Based Medicine.

It will do several things:

1. First — the key thing is to pave the way for RxISK.org, a website for doctors and patients to enter reports of treatment related problems. For reasons outlined below and in my latest book, Pharmageddon, having another talking head is not a way to solve problems. It’s time to do rather than talk or write, and RxISK.org is about doing. It will be the only site on the web where patients and patient-doctor teams on a global basis can enter the details of what happens on treatment and can discover both new things about treatments and also whether a particular treatment is actually producing a problem for this patient.

It’s time to do rather than talk or write, and RxISK.org is about doing.

2. Second — I will post a set of questions and mysteries for which I have no answers. These are not questions without answers, or mysteries that cannot be solved, but some will likely remain unanswered and unsolved. Some because they are quite old and the trail may be going cold. Others because they are enduring. But some because the answers are hidden behind barriers of attorney-client privilege – a barrier that few people are aware of. The best response to these questions and mysteries will be data or documents that provide answers to the questions or information that sheds light on the mysteries.

The best response to these questions and mysteries will be data.

The style in which these questions are raised will often be quite stilted and the posts will not come thick and fast. There is a reason for this. Several years ago I put in a Freedom of Information request to Eli Lilly. The 103rd document that came back read as follows:

103 “Healy long term strategy.

Thank you for the message outlining your strategy to counteract Dr David Healy’s claims re: Prozac and violence.

Send a letter to Healy designed to get him to stop discussing a study that he has never done. Have a third party expert in the audience at BAP to ask Healy questions when he presents.

Just last Thursday Healy was quoted in a Cincinnati paper saying Prozac causes violence and suicide…X has asked that we go back to legal and determine if we can sue Healy under UK law.”

Anyone responding to this blog therefore needs to take care with their choice of words.

Anyone responding to this blog therefore needs to take care with their choice of words and ideally needs to stick with data, and documents. This is not a forum for prejudices or unsupported opinions; it’s for moving questions forward or solving mysteries.

(Note Post Hoc: Some colleagues have wondered if I am or was being sued by Lily. The answer is No. If you want to read more on this episode, see the Academic Stalking section on HealyProzac.com or download this article on Academic Stalking.)

Other documents show that at least one public relations agency linked to Lilly was tasked with developing a Healy management plan. Given that the marketing departments of each pharmaceutical company employ several PR agencies and I’ve given several marketing departments an incentive to manage me, this seems a reasonable way to explain for instance getting an email from a colleague in Japan who had just met psychiatrists from the US who on hearing he knew me advised him to have nothing to do with me as I was trouble and I would soon be in trouble. The interesting aspect of this was that the American psychiatrists he named had never met me, and had never engaged with me in any way.

A UK PR rep told me I was doing more for sales of Prozac than anyone else!

But the picture is complex. Many years ago a woman who claimed to be responsible for PR for Prozac in the UK introduced herself to me at a meeting saying she was really pleased to meet because I was doing more for sales of Prozac in the UK than anyone else.

This is not a simple picture with the good guys (the clinicians) on one side and the bad guys (industry) on the other. Most of the grief I’ve experienced has come from clinical colleagues, as the website healyprozac.com, especially the academic stalking section, shows. The solid help I’ve had has mainly come from within industry.

A Japanese colleague was advised to have nothing to do with me by an American psychiatrist who had never met me.

The challenge in this world managed by public relations agencies and marketing departments for both scientists in industry and insurance companies and the rest of us is to find a way to turn this dynamic around and use efforts to hide problems as a way to make things safer. If there are just two boxers in the ring, it’s possible to try Muhammad Ali’s rope-a-dope strategy, but there are at least three in this case. The third party is organized medicine — doctors. The fact that medicines are available on prescription only has made doctors, rather than companies, the chief deniers that a drug might be causing problems. Insurance companies, regulators, and others are also parties to these events and their interests in the outcome shift around from time to time.

3. Third — this site will in due course have a category of posts for people who have been through the system, people who have had partners, parents, children, or friends injured by treatments and who have found themselves trapped in a Kafkaesque world when they have sought help from doctors, regulators, or others who seem to be there to help us.

Their stories will highlight the lunacy of the system but will also show how we can change everything.

Their stories are aimed at highlighting the lunacy of the current system but also showing how someone who is determined can change everything. These stories will likely migrate to RxISK.org when it is up and running.

4. Fourth —the site will have a myth-fable-fairy tale slot. Academic lectures or articles seem to make little difference, perhaps even increase sales of whatever drug is being discussed. But very often people trying to describe the issues turn to fairy-tales or myths to explain what’s going on. If the Emperor’s New Clothes turns up once, it does so hundreds of times. This suggests trying to pull together a collection of alternate ways to describe things. I’ll take a stab at posting one or two examples as starters over the next few weeks and then throw the slot open to others.

If the Emperor’s New Clothes turns up once, it does so hundreds of times.

5. Finally — the main theme of the early posts is missing data and how impossible it is today to practice Data Based Medicine. If the data on which we base our treatments is missing, what might data-based position papers on various treatments look like? I will post a draft position paper on the Antidepressants. The principle underpinning these position papers is a set of questions to which the answer will commonly be: We have no data. It would be excellent to have input from anyone who can provide comparable statements for other drug groups or who can see ways to improve on what has been done here for the antidepressants.

If the data on which we base our treatments is missing, what might data-based position papers on various treatments look like?