In my last post [the illusion of evidence…], I was playing with the phrases evidence of illusion [looking for signs that a scientific paper has used the the techniques of presentation to obscure rather than clarify data] and the illusion of evidence [valuing the mathematical/statistical results of Randomized Clinical Trials over all other sources of information]. I was trying to get at something besides simply the corruption of the Clinical Trial system – but lost track along the way. The corruption theme is so compelling and those two articles in Matter were so powerful, that they carried away my narrative. The thing that got lost, the thing I started off wanting to say, probably deserves its own separate post anyway. In a recent article [unfortunately behind a pay-wall], David Healy makes the point explicitly,
Today, many argue that the growing crisis in health care stems from conflicts of interest and lack of access to clinical trial data. Our view is that small-print disclosure in academic footnotes and open access to trial data, important though these are, will not solve problems that stem from the notion that clinical trials provide a higher form of knowledge than knowledge borne in a clinical encounter – the realm of the experiential and the singular…
and he illustrates it with a poem:
This ‘Trust in Numbers’ story was caught wonderfully when it was unfolding, by George Oppen, a New York poet and member of the communist party, in his 1968 volume Of Being Numerous:
Crusoe we say was rescued.
So we have chosen.
Obsessed, bewildered
By the shipwreck of the singular
We have chosen the meaning
Of being numerous.
I envy that eloquence. For many, this point is counter-intuitive and easy to brush off.
beyond corruption…
Even in the best of circumstances – when protocols are followed precisely, diagnosis is rigorously pursued, analyses are pristine, conflict of interest is absent – a Randomized Clinical Trial is time-limited, and constrained by the measurement instruments selected, proxies like the HAM-D scale substituting for depressive affect. In addition, statistical significance or even effect size measurements don’t necessarily parallel clinical changes that make much of a difference. Subtle adverse effects often become apparent in a time-frame that exceeds the trial’s duration. In many cases, the cleanest of clinical trials can only say that the drug may possibly be effective and may possibly be safe.
beyond limitations…
In the case of psychiatry, the limits of the clinical trial are in bas relief. Unpleasant emotions themselves are not pathological – anything but. They are a part of an essential and elaborate warning system similar to physical pain. Among those presenting with emotional difficulties, primary disorders of the emotional system [Manic Depressive Illness, Melancholia, Anxiety disorders, psychosis, etc] are in the minority. In the majority whose emotional discomfort is related to some aspect of their life or personae, the use of emotion altering medications is highly dependent on the specifics. Likewise, the capacity to manage adverse effects, addiction, withdrawal, the downside of psychoactive medications, is as variable as ripples in the stream. A physical analogy might be the use of corticosteroids. For some, short-term use can be therapeutic. For others, it can be the introduction of a medication that can be over-used with disastrous consequences.
beyond groups…
After medical school and an internal medicine residency in a busy charity hospital where everyone was quite ill, I found myself in a military hospital on a base with healthy soldiers and their dependents – a very different place. My head was filled with facts and algorithms galore. But I had much to learn. The majority of patients had symptoms but rarely any of the dire illnesses that populated my mind. And I realized that a lot of my job was informed reassurance – a different role I had to learn to fill. Somewhere along those early days there, I realized that I was thinking in terms of case studies, talking less about diseases. A particular case comes to mind that goes with this topic:
She had been diagnosed with Idiopathic Thrombocytopenic Purpura, a condition where the platelets, cells involved in clotting, all but disappeared leading to easy bleeding. The wisdom of the time was that taking out the spleen would cure cases in kids under 13, but wasn’t helpful in kids over 13. She was 13 at diagnosis, and a splenectomy had been curative. When I saw her, she was thirty something, mother of two, and had consulted a gynecologist for heavy periods. She was referred because her blood smear showed no platelets and her platelet count was essentially zero. We did the workup and started her on the recommended anti-metabolite. We were rewarded with a rising platelet count. One day, she came in "itching," a sign of liver toxicity and we watched with dismay as she became jaundiced and quite ill from our treatment. Her platelet count rose to normal, but it took a month for her liver function to return [to our great relief].
We sent her to the Air Force teaching hospital in Texas. When she returned, she was on another anti-metabolite [cousin to the one we used]. Our consulting hematologist sent us a stack of articles about treatment. Her platelets rose, but the itching returned, along with signs of liver failure. This time we stopped the medication more quickly and she didn’t get quite so sick. One day, I asked her about bleeding, and she said that except for heavy menses, it was no problem. I had been following protocol, and had given her two cases of liver failure chasing an abnormal lab test in an otherwise asymptomatic person, abetted by our best and brightest. We stopped the medications. Depot hormone shots eliminated her periods. And she did fine for the two years I followed her after that.
I know it’s not a Randomized Clinical Trial, but the case has always stuck with me because I was young and wanted to do the right thing, so I checked with the experts and followed recommendations precisely, but I was treating a surrogate, an abnormal lab value, and not attending to the clinical reality of the patient. I have no idea how this patient lived with such a low platelet count and no bleeding problems – but that was the way the case went for years after I stopped assaulting her with chemotherapeutic agents. I could tell such stories for hours – stories where well studied guidelines, clinical trials, or algorithms weren’t right for a given patient. I learned that the case at hand can often end up trumping the group recipe when they are at odds. Clinical medicine mirrors the Zen saying:
you can’t get it from books,
you can’t get it without books.
Without the evidence-base from history, group studies, treatment guidelines, we’re lost at sea. But without careful attention to the individual case in front of us, we can wander just as far off track. And so to the point being made by Dr. Healy:
…small-print disclosure in academic footnotes and open access to trial data, important though these are, will not solve problems that stem from the notion that clinical trials provide a higher form of knowledge than knowledge borne in a clinical encounter…
beyond clinical trials…
The necessarily contained environment of a Randomized Clinical Trial is only a distant surrogate to that of the patients encountered in clinical practice, analogous to the relationship between the model airplane and the jumbo jet – a useful step, but hardly the full story. The cogent question is how did it come to be that the clinical trial has been escalated to some kind of gold standard for evidence based medicine rather than what it is – a minimum safety and efficacy requirement for FDA approval.
The point of the push for accurate Randomized Clinical Trials is to give practitioners a clear picture of the safety and efficacy of available medications, not to direct treatment or initiate an ad campaign. The real clinical trial is in the individual patient, and sometimes, the best medicine is in none at all.
With the patient mentioned above, we saw her frequently. We made sure her husband’s next assignment was to a base stateside with a full hematology department [affiliated with a medical school]. He sold his Porsche and bought a Volvo after we cautioned them about car wrecks. But other than stopping her menstrual periods, it was what’s called benign neglect or watchful waiting.
Unfortunately, many college educated people— the ones I know from political forums— are so pro-science and so convinced that the double-blind placebo study is a genuinely gold standard, that they don’t stop to consider that the drug companies are the money-hungry corporations they like to critique. I’ve often seen these people facetiously say, “Oh yeah, the evil Bigpharma” to mock what they perceive to be an ignorant denial of science. They have no idea that most of what they believe to be true about mental illness is marketing.
I’m working on ways to challenge this and one of the problems is that their responses to individual testimonies is “anecdotal”; when the fact that different drugs effect different people differently, behave differently with other drugs, and a whole slew of other factors. It’s not “anecdotal”, it’s the effects of the drugs.
At times, they’ll dismiss a study because of small sample size, but doesn’t larger samples make it easier to come up with statistically significant numbers?
Another thing that makes it difficult for me to adequately address this problem is the fact that I can’t afford the research and do not have time, skills, or access to, for example, determine how many and what studies have concluded that people who do not take drugs for first psychosis are more likely to recover and be restored to full-functioning (after two years).
Thanks for pointing out again that the FDA is not there to regulate the practice of medicine – it is there to regulate the claims of the corporations that want to sell drugs and medical devices. That said, there are good reasons for controlled clinical trials before a treatment is adopted.
When David Healy critiques “the notion that clinical trials provide a higher form of knowledge than knowledge borne in a clinical encounter – the realm of the experiential and the singular…” he draws from the distinction between disease and illness. Clinical trials deal with operationally defined diseases, whereas treating clinicians deal with singular illnesses. As you say, relevant factors that operate in the singular illness may not be considered in the clinical trials – age; co-morbidity; concurrently required medications; insight; capacity for a treatment alliance; duration of required treatment; family stress; economic stress; and many others. So, the knowledge gained in clinical trials is needed but is not necessarily generalizable to or determinative in the management of an individual patient. So, it’s not a question of a higher form of knowledge so much as it is addressing a different question.
Overall, we can be sure that a truly useful new drug will sell itself without a lot of marketing and with no need for deception, and without the need for experimercials. And, the best contribution of clinical trials nowadays may be to keep junk out of circulation because the trials failed – two good recent examples are CRH receptor blockers for depression and the glucocorticoid receptor blocker mifepristone for psychotic depression. Clinical trials have a legitimate place in drug development – we just need to recapture the field from the commercial distortions.
Sir William Osler
“The good physician treats the disease; the great physician treats the patient who has the disease”
Steve Lucas
Yikes!
… nine Canadian judges reviewed Eli Lilly’s patent applications and ruled that they failed to meet the standards for patentability. Yet in June 2013, the company served notice that it planned to file a complaint under the North American Free Trade Agreement claiming that in light of the decisions, Canada is not compliant with its patent law obligations under the treaty. As compensation, Eli Lilly is now seeking $500 million in damages.
The Eli Lilly claim is winding its way through the legal process – the Canadian government filed its defence several weeks ago – but the implications are being felt far beyond the specifics of the case.
If the pharmaceutical giant succeeds, it will have effectively found a mechanism to override the Supreme Court of Canada and hold Canadian taxpayers liable for hundreds of millions in damages in the process. The cost to the health care system could be enormous as the two Eli Lilly patents may be the proverbial tip of the iceberg and claims from other pharmaceutical companies could soon follow.
http://www.michaelgeist.ca/2014/08/20-year-old-patent-application-end-canadas-biggest-trade-deal/
Mickey — Do you make your e-mail available at all?
Steve–
If Osler were alive and living in my neck of the woods, his statement would have to read: “The physician treats the symptoms (and treats them and treats them), the good physician treats the disease; the great physician treats the patient who has the disease.”
Melody
Melody,
Yep, and sometimes you do not even need symptoms:
http://www.healthnewsreview.org/2014/07/one-physicians-story-an-egregious-example-of-ordering-unnecessary-tests/
Steve Lucas
The role of pivotal clinical trials is to demonstrate efficacy as the basis of FDA approval, not to demonstrate utility or use of any agent in any given patient. The results of the pivotal trials that support a particular indication are described in the package insert with the drug and it is our responsibility as physicians to know the basis of a drugs approval, versus what we and our patients need in a treatment. An antidepressant approved on the basis of an 8 week trial demonstrating a 50% reduction in Hamilton score or an antipsychotic with a 30% reduction in PANS score over 8 weeks is very different than what we want to accomplish with our patients. As Dr Carroll mentions the FDA does not regulate or mandate or inform clinical practice, it regulates claims of efficacy and safety of medications. It is our responsibility as physicians to know as much as we can about any treatment and gather that information from a variety of sources, pivotal trials are only one such source, along with other literature (sorting through the experimertials), our experience and that of our trusted colleagues. But we have an obligation to understand the basis of drug approval and the trials that went into that approval. I do agree having results of all trials available would make our job somewhat easier in assessing the utility and safety of a medication. Keeping in mind a company can do as many trials as they need to get 2 positive trials, knowing how many and the results of all the other trails would be helpful to say the least. As I teach my trainees, pivotal trials demonstrate efficacy in a very narrow sense, they teach us something about the drug but they do NOT teach us how to best use that drug in any specific patient.