1 Boring Old Man

fyi

http://www.philly.com/philly/blogs/phillypharma/JJ-CEO-Alex-Gorsky-talks-Risperdal-via-deposition-with-lawyers-nearby.html#ixzz26lAZmNxP

Monday, September 17, 2012

“J&J CEO Alex Gorsky talks Risperdal, via deposition, with lawyers nearby
Johnson & Johnson CEO Alex Gorsky

Johnson & Johnson Chief Executive Office Alex Gorsky was proud enough of helping to boost Risperdal sales to include the numbers on his resume at one point, according to court filings, but he is trying to hard to avoid talking about the antipsychotic drug on the witness stand in front of a jury in a trials scheduled for the next several weeks in Philadelphia.

On Sept. 10, with a jury waiting in Philadelphia Court of Common Pleas, J&J settled a case with a man who claimed he grew breasts after being prescribed Risperdal as a child, rather than risk having a judge rule that Gorsky must testify.”

blog post contains a link to 60-page transcript of Gorsky’s deposition in the individual suit filed by Aron Banks in Philadelphia was a publicly-available exhibit in the case

1BOM,

Thank you.

It’s better to know the truth, than to remain naive—even when it changes things for the worse. There is no comfort in that truth. It makes me sad too…Thank you for sharing the ugly truth. There is comfort in knowing you see it too.

As an adult psychiatrist in training to become a child and adolescent psychiatrist, I often wonder at the extent and prevalence of off-label prescriptions in my new field. I think that a distinction should be made between two different kinds of off label prescriptions, and this distinction is blurred in the different professional guidelines and “talking points” you have recently quoted. The first kind is when a medication has not been tested yet on a specific population, but there are reasonable empirical and scientific grounds to suppose that it may be helpful for that population; in this case, and when better tested alternatives have either failed or are not available, it is logical to try an off-label drug. The other kind of off-label prescription, which is illegitimate yet prevalent, is when a drug has been tested on adequate samples of the target population and has not been found efficacious enough or safe enough. In contrary to what you wrote in a previous post (http://1boringoldman.com/index.php/2012/09/13/eyes-wide-shut-open-iii/) about fluoxetine: ” It was one of the two studies used by Lilly to get approval for Prozac in the treatment of MDD in Adolescents [the only SSRI approved for kids since it got through the FDA before the black box warning was imposed].”, many other of the commonly prescribed antidepressants have been tested by the manufacturers on depressed children and adolescents, but did not qualify for an FDA indication for that population because they did not clearly show an advantage over placebo ( not just paroxetine, but also sertraline, citalopram and venlafaxine, to name a few). I think that medications of the latter kind should not be prescribed for adolescent MDD (some of them do have FDA indications for adolescent anxiety disorders).

Dr. Mayer,

You are going to be a child and adolescent psychiatrist. Do any of the concerns raised here about AACAP resonate with you?

Do you plan to join AACAP when you complete your training?

How did you learn of 1BOM?

Annonymous: I am certainly concerned about the biased representation of these issues by the AACAP, because it is looked upon as a good source of information by both physicians and parents.
I am not American so will not join the AACAP. I’ve been a devoted reader of 1BOM for a while, but don’t remember how I first found him.
If by any chance Dr. Bernard Carroll reads my previous comment, I would be very interested to learn what he thinks about the issue.

It’s remarkable, but I cannot name a single American child psychiatrist, prominent or otherwise, who has spoken out about how pill focused AACAP has become. That streak continues.

1BOM, it is my argument that if AACAP is not effectively that the impact on child mental health down the line will be significant.

I would make the points that
1. There are alternatives to child psychiatrists for expertise on child mental health issues. A blog such as “Child in Mind” by Dr. Claudia Gold provides a example of this. As does an organization such as “Zero to Three.”
2. It is possible to believe that child mental health issues are real, and have neurobiological underpinnings, and require intervention, while not necessarily believing that the term “treatable” centers primarily around the pill du jour. Dr. Bruce Perry, who I believe testified at the Texas TMAP trial, is an example.
3. The big shift is into child psychiatrists becoming diagnostic/pharmacotherapy advisors to pediatricians. Is you think you’ve seen lots of kids medicated so far you ain’t seen nothing yet.
http://www.aacap.org/galleries/PracticeInformation/Collaboration_Guide_FINAL_approved_6-10.pdf
4. AACAP appears keenly interested in influencing certain constituencies they can court but cannot control. Among these appear to be Pediatric and Family Medicine primary care providers and Government. Of course this also includes the majority of child psychiatrists who are not KOLs. There was a description on another site that in terms of becoming a KOL that you could be the player or your could be the ball. So, the balls, or the played, if you will.
5. There is a cost to challenging the status quo. The “hardball” Keller refers to, the professional consequences Ryan expresses concern over, what comes across in Dulcan’s language. If there remains little to no cost in the status quo there will not be change.
6. That cost might be minimal for some groups that can impact AACAP but for whom this is not on their radar and/or right now their attitude is that they don’t have much interest in exercising their leverage. It would be of value for that to change.
7. If there exists a high yield approach, it might be in finding a way to effectively galvanize those groups to, with minimal effort, want to impact AACAP.
8. E.g., since there are alternatives to AACAP, congressional staffers could choose to stop accepting/soliciting visits from AACAP representatives. They could make suggestions that opinion papers from alternate groups (e.g., pediatricians or psychologists or even child portion of the APA be used instead) if they want expert input. They could let the AACAP representative know why.
9. Child psychiatrists are planning to be expert advisors to pediatricians. AACAP is supposed to vet these consensus committee experts? Wagner was on the 2007 depression one. Why in 2002 was she pushing a short half life drug like paroxetine over fluoxetine? Keller in his deposition makes the poin he’s not a pediatric payhopharmacology expert, he’s a clinical trial expert. Wagner didn’t seem to have much to do with running the trial, analyzing the data, the poster she presented was ghostwritten by SKB. Was she seeing many child patients at the time? To follow up on 1BOM’s point: how many kids had she personally followed longitudinally face to face on paroxetine, on fluoxetine, who had MDD? How many, before he started giving multiple talks promoting paroxetine’s first line use over fluoxetine? Does anyone doubt whether SKB would have continued paying her for all those talks of she hadn’t? Pediatricians can find alternatives to AACAP for guidance. It’s one thing to push for off label use, it’s another to take 750,000 dollars from one company for one drug (2 if you count it’s enantiomer) as the chair of their CME committee has done. If you are a pediatrician making 125,000 dollars a year with a full time practice you know that organizations do not just hand you close to 125,000 a year just for the hell of it. Pediatrics still has well child visits, they don’t need to assign a dsm code to get paid, and they still have a literature where a pill can fail.
10. Maybe if the reputation of the academy suffers in a meaningful way, the AACAP you see on their website that advocates for kids with pills in the picture (rather than being the picture, rather than being the stand-in whenever they say “treatable”) will see a reason to push against the AACAP we’re talking about here. To deal with the vindictiveness, the “hardball.” Right now there appears to be none.
11. If people like Thacker, who can maybe influence the government side re AACAP, and other who can effectively and consistently reach primary care providers and front line child psychiatrists (the “balls”), don’t take action then I fear that a few years from now we will be seeing am ugly marriage between the “talking points” you describe and this:
https://aacap.confex.com/aacap/2012/webprogram/Session9312.html

I realize that, as your feed states, this site is “All Mickey, All the Time” not “All AACAP, All the Time.” I also doubt there are people out there with bith the motivation an ability to make such interventions come to pass, and these issues about the academy will return to being anechoic. But I do hope that you continue to keep the academy in mind.

AACAP is driven by a model of alleviation of childhood suffering coming in a pill. That the dangers of over generalizing this pale in comparison to the dangers of under detection. This model is so ingrained that it will maintain for the generics as well as brand name medication. The importance of reflection on the behavior of these KOLs, still steadfastly embraced by AACAP, is because it reflects on the judgment of the organization. If you are someone worried that a sad child is going to be seen by a provider for 10 minutes and put on a pill and somehow the system will count them as one of the common and treatable who has been helpled (with the evidence base developed by people like Ryan and DelBello and prosteletized by people like Wagner and Bostic), then you should care very much about the nature and strength of AACAP’s voice. Whether you’re a child psychiatrist, a primary care provider, a child researcher, or an advocate who has seen how much difference can be made with more sophisticated and costly interventions (which are so often subtly downplayed by the Amercian Academy of Child and Adolescent Psychopharmacologists in their talking points).

Please let my earlier comment post when you have the chance.

Primary care providers and legislators need to realize that AACAP is attempting to transition from an age old message of “First do no harm” and the pediatrician’s viewpoint that from a perspective pf primarily normalizing childhood from a MEDICAl perspective to a “We’ve got a pill that works for that” and “1 in 5 Kids have something wrong with them for which there is a pill.” And that the judgment exercised in the Study 329/JAACAP paper affair is representative. There were many expert child psychiatrists involved in promoting paroxetine as first use for adolescent MDD over fluoxetine, including AACAP. In a way Dulcan was right. And that should frighten those who care about the mental health of kids. Including those who feel that medications have an important role in the treatment of some kids. Because you care about the scholarship behind the recommendations you are getting too.

KOLs, including some who fly under the radar such as Dr. Carlson, had their power greatly boosted by ther relationship with pharma. Now that some of the dust is settling, they are not going to allow easily for a genuine reflection on this era, or the attitudes it ha spawned. This will. It not occur without leverage being applied to AACAP. Time will tell if anyone will take on that task and pursue it effectively. So that the genuine medical needs of some proportion of kids is not generalized to 1 in 5 (hearing towards 1 in 4 and counting).

How different will the care of the mental health care of kids be if these voices have a greater presence:
http://claudiamgoldmd.blogspot.com/?m=1
http://www.childtrauma.org/
http://www.zerotothree.org/
1BOM
Rather than AACAP and these people:
http://www.thebalancedmind.org/about/staff/cabf-scientific-advisory-council
Including Dr. Luby’s extension of mood disorders to toddlers. Dr. Carlson as chair of the current AACAP conference. Balanced Mind being one of two sites linked to by the AACAP bipolar disorder information page.

How different will it be when the Siren’s call of that group of KOLs, when their influence within AACAP, grows even more? Look at the direction of Dr. DelBello’s Work on “prevention” in younger kids. Soon enough we’ll have a BP med or statin model for PREVENTATIVE psychopharmacogy in young kids. Then the pediatric market can really become worthy of pharma’s time.

And those KOLs on the balanced mind page, and AACAP, can show the same level of ethical behavior. And Drs. Ryan and Wagner can bring the same level of scholarship and clinical acumen as before. As bright to you by the people who pushed Paxil over Prozac as first line in depressed teens.

I wonder how those months of “vetting” by Dr. Drell and his executive council is going.

The same kinds oF institutional hubris in child psychiatrists that brought us the paradigm Of the refridgerator mother now has some new paradigmatic oversimplifications tailored for the 21st Century.

The old is new again.

You’ve probably seen this, but just in case:
http://www.pharmalot.com/2012/09/the-op-ed-a-suggestion-to-restore-faith-in-pharma-studies/
Dr. Drazen’s comment speaks volumes about the state of peer review in the field. But this commentary in Pharmalot, and the original paper itself, are encouraging.

To continue my beating of a dead horse: the argument is that if they don’t do something to restore trust that pharma studies will lose their ability to influence. This discussion is happening because there is that fear. That is the impetus. The damage to the brand. I think this is at the heart of what I’m arguing for professional organizations. If the trust in them by primary care providers and government appears to be at risk they will care. Pharma is getting beaten up, but the professional organizations that should have been tempering their influence instead rejoiced in their new found resources.

Dr. Krumholz argues that pharma has a problem. They need to instill trust. This may or may not lead to greater transparency. But at least there has been enough success in impacting trust that it is an issue.

This has not happened to the professional societies, nor my great white whale in the comments section: AACAP.

Also, is there any child psychiatrist with a similar stature in the states in the field to Dr. Krumholz saying anything similar?

Why not?

P.S. 1BOM, you’ve communicated directly with Mr. Jones. Have you ever communicated directly with Paul Thacker. He does appear to be at the Saffra center. The center doe appear to have an interest in institutional ethics. It would be fascinatin to see that center produce a white paper on AACAP in relation to Study 329 and the use of paroxetine in adolescents, the Texas algorithms and their use of risperidone in foster care youth, the emerging field of mood disorders in 0-5 yer old kids, …etc.

THAT would provide a platform for approaching pediatricians and legislators. And, that might send a signal to professional organizations that they should be supporting True and Practical and Meaningful transparency and should be adding their voice to that chorus.

Rather than trying to create more defensive statements, and create more transparency portals, and more consensus committees (it’s not even clear why AACAP disn’t use their extant ethics committee for that instead of creating a new consensus committee with Ryan and DelBello).

What day of the AACAP “vetting the issue” watch are we on now? What decade if we start counting from the publication of the 2001 paper?

Maybe here is a more constructive approach:
I wonder about an open letter to the Safra center suggesting that they approach AACAP and offer to produce an external review of their institutional ethics a a way of strengthening the work of the organization. I wonder if the center would seriously consider that. With their funding they could do it free of charge. And, if the academy remained convinced that their internal resources have always been, and continue to be sufficient to that task, perhaps the center could choose to go ahead with such an analysis anyway.

In any event, the Pharmalot op-Ed is a good read.

An open letter to Paul Thacker and the Edmond Safra Center for Ethics at Harvard re AACAP might actually be of interest …

P.S. 1BOM, you’ve communicated directly with Mr. Jones. Have you ever communicated directly with Paul Thacker. He does appear to be at the Saffra center. The center doe appear to have an interest in institutional ethics.

Yes. He is at the Safra Center. “But I’m not working on pharma issues. I’m doing a book project about the hill.” There aren’t enough Paul Thackers to go around, it seems.

Does he have a recommendation for another “Paul Thacker” at the center who might have an interest?

The Future:
https://aacap.confex.com/aacap/2012/webprogram/Session9023.html

From Ben Goldacre’s new book Bad Pharma http://www.guardian.co.uk/business/2012/sep/21/drugs-industry-scandal-ben-goldacre?

@ Gaddy Meyer: thanks for the query, which I saw just today. I see where you are coming from with the distinction between two kinds of off label prescribing. However, I come down on the side of not absolutely proscribing either of them, while recognizing that there are dangers with both of them.

Regarding your first type, it all hinges on the plausibility of the rationale for going to an off label drug in a particular case. Remember, these are N = 1 experiments. When Robert Post and James Ballenger introduced the off label use of carbamazepine for bipolar disorder, reasoning from the kindling theory and from the good reputation of carbamazepine in temporal epilepsy, I regarded that as a reasonable thing to try. However, if the off label use of the drug is based on just naked empiricism or commercial pressure, then we need to pause. One of the scariest moments I can recall as an educator was in a discussion with a primary care physician who assured me he had found the answer to refractory depression – lamotrigine plus quetiapine in combination. Likewise, over on Health Care Renewal I have written extensively in criticism of using atypical antipsychotic drugs in nonpsychotic depressed patients.

Regarding your second type of off label use, I would allow it if it makes sense for a particular case. So, for a late teenager with an adult appearing melancholic depression it might well be reasonable to use a non-approved antidepressant other than fluoxetine from the adult formulary if the physician is comfortable with using that drug already in adults.

In her critique last year in New York Review of Books, Marcia Angell opined that all off label prescribing of psychotropic drugs should be forbidden. I wrote to Dr. Angell privately to say I disagreed, and we had a cordial exchange on this point, agreeing to disagree. We do need to keep in mind that the charter of the FDA does not include regulation of the practice of medicine, but rather regulation of the claims made by pharmaceutical corporations.

@Bernard Carroll: Thank you for replying. I agree with your point of view. My division was too rigid and simplistic. But I do find that not just in common practice but even in the different “guidelines”, there is a tendency to disregard negative studies. Sometimes this is because the negative studies are just not published (as is shown in the Goldacre piece linked above by Altostrata), but at other times it seems to be because of a bias, a view that if a study is negative, then the study itself must be at fault, not the medication/intervention.

The “success” of the “n of 1” experiment with the child in front of you always seems to trump a negative finding in a larger study. Look at the findings in the pediatric literature of antibiotics not helping with certain coughs or sore throats or ear pain. That the feeling they help is an illusion. Or the HRT findings? Where are the equivalents in psychiatry? The equivalent studies that say one should do NO intervention because you don’t have one that works?

In reply to Gad Mayer and Annonymous: Here is a further discussion of some of these issues. http://hcrenewal.blogspot.com/2009/04/in-defense-of-psychiatric-diagnoses-and.html

There is a good deal of pragmatism in the practice of medicine. Mostly, we manage individuals rather than populations. So, it’s not that the N of 1 experiment trumps large clinical trials. It’s that we try to match the large clinical trial data to the case at hand, with due consideration of the faults and the gaps and the non-generalizability of the large trials. I don’t care what the meta-analyses say, if I am caring for a 17 or 18 year old with severe retarded depression and a loaded family history of mood disorder and suicide then I will treat vigorously with an antidepressant, which for sure won’t be fluoxetine, and I will use every adjunctive intervention that I can, as well.

Dr. Caroll, So if you were a geriatric psychiatrist and your clinical experience is that atypicals are helpful in the setting of severe behavioral disturbance in dementia you go forward with that, the results from studes notwithstanding? And if you are a child psychiatrist and you have a non-autistic very young foster child in front of you with severe behavioral disturbance in the context of significant past trauma no number of negative or equivocal studies would hold you back from neuroleptics and other “adjunctive” treatments? “Pragmatism” is all well and good, but what formal mechanisms are you using to control for the cognitive biases and traps inherent in the “tailoring to the individual” approach you are describing? If “pragmatism” means you are able to discern and compensate for these traps and biases just through clinical judgment, that black box called the “art” of medicine, and you can’t point to formal mechanisms, then it would seem like you’re just saying “trust me because I’m a doctor.”

Sorry, meant to say Dr. Carroll.

P.S. why for sure wouldn’t it be fluoxetine?

To your question, Annonymous, I think I would frame the issue differently. I don’t believe in “atypicals” as a class – that is just a marketing trope. And I don’t believe in “severe behavioral disturbance” as an automatic indication for any class of drug. As I said before, it all depends on the specifics of the case. But, yes, I might very well use an antipsychotic drug for an elderly agitated patient who is delusional and hallucinating while I work him up for causes of delirium. Likewise for a patient with Lewy body dementia who has clear organic delusions on which she is acting out. I know studies say that may shorten her life expectancy, but my goal is to improve her quality of life while she is still with us. The longitudinal N = 1 experiment allows us to evaluate the progress toward that goal, with dose adjustments and trial withdrawals.

As for your example of the hypothetical younger patient, I should not comment because I have not worked with such patients professionally. But in general my preference is to avoid such drugs in young patients, especially considering that the positive trials are experimercials.

Is there an “art” of medicine? Yes. Is there a skill called “clinical judgment”? Yes. I am proud to affirm both. Can we reduce clinical decision making to formal mechanisms? Not yet. Look at the train wrecks called STAR*D and the like. The practice of medicine requires us to make decisions under conditions of incomplete information. Will we always be right in hindsight? No. Get used to it.
Do we say, in effect, “Trust me because I am a doctor?” Yes. That pretty much captures the essence of the Hippocratic Oath, which of course obliges us to put the interest of the patient first while removing all personal conflict of interest in our professional work. We cannot have the former without the latter, and lately the latter has been in short supply among our so-called leaders.

I found Dr Carroll’s exchange with “Anonymous” (who insists on anonymity because he can’t spell) intriguing because it brings to life again the old debate between medicine as a science and as an art. The FDA holds up the science end in insisting that agents pass RCTs before being licensed. Yet the indications in these RCTs are always limited, or bogus. Dr Carroll is saying that years of clinical judgment and experience acquired at the bedside are at least as valid as the knowledge acquired in these RCTs, and that off-label prescribing represents — if I may extend his argument a bit — a further exercise in the art of medicine. I find this argument quite congenial.

Drs. Carroll and Shorter,

I appreciate you each having taken the time to reflect upon these questions.

ART VS SCIENCE?

I would argue that the “science” vs “art” dichotomy is not the most useful lens to use.

Clinicians are trying to achieve the best outcomes for each of the individuals they are treating. To me it is more a question of all of us looking critically at the heuristics used to try to ascertain the path to these best outcomes. The need to consider the characteristics of these heuristics, and most importantly how to judge how to most effectively coalesce them, applies equally whether one is extracting conclusions from the experience with 1,000 patients with a particular problem that one has treated personally, or from the experience with 1,000 patients with a particular problem treated across several sites as part of a clinical trial.

Being aware of a bias is obviously not equivalent to being able to parse its impact in such a way that you can draw an accurate conclusion. There are some very real problems with intuitions derived from clinical judgment and experience acquired at the bedside and the part of developing approaches such as randomization, double-blind, intent to treat, apriori establishment of primary and secondary endpoints, was to have a way to double check these intuitions and compensate for some of those biases. Those studies have some serious problems of their own (not the least of which is having a patient population physicians can generalize to those they treat), but if there are few to no situations in psychiatry where a clinical trial can be viewed as sufficiently well designed and implemented so as to overturn one’s own clinical intuition about a treatment (people are always happy to point to the clinical trial that disproves that which they believe should be disproven) then I continue to find that concerning.

I am NOT arguing that clinical decision making should be reduced to formal mechanisms. They are train wrecks, and 1BOM attended part of a trial dedicated to what was wrought from one of the most egregious examples, TMAP. I am very used to physicians not always being right in hindsight.

But how are we going to be able to even judge whether or not you were right in hindsight?
CATERGORY A+?

I am in full agreement with Dr. Carroll that addressing the blatant conflicts of interest are very necessary. I do not agree with the implication that that will be sufficient. You both seem to be arguing that the wisdom gained from a combination of direct clinical experience and one’s clinical deductions, unblinded and warts and all, should be allowed to take its rightful place alongside RTCs, and not be relegated far down the alphabet while RTCs sit alone in category “A.” That’s more than fair.

But, I don’t believe that it is fair to every single time place that direct clinical experience and deduction in category “A” above RTCs. Category A+ so to speak. To say, I’m always going to try the treatments if the patient in front of me is in need, even if the RCTs argue it’s useless. To always say: Because I’m treating a patient, not a population. That seems to be what is subtly (and not so subtly) happening. Because then you’ve defeated the whole raison d’etre of what the RCTs were originally meant to be able to do, before they became experimercials (which is a wonderful term by the way). They were meant to be able to overturn beliefs that had grown through direct clinical experience but where the belief was primarily based in uncompensated for confounds.

I know I’m stating the obvious with that, but let’s say a psychiatrist cannot name a single RTC that he/she considered sufficiently well designed and implemented so as to have overturned his/her previously held belief, based on their own clinical experience, about the effectiveness of a treatment. Overturned to an extent that they even if it one of the adjunctive mechanisms they can use, they choose not to. And also cannot name systematically applied heuristics that they have used to compensate for the biases inherent in drawing from their own clinical experience. Do we suppose then that they just intrinsically are able to parse all of the pitfalls that led to the development of the clinical trials as tools over the past 50-60 years. Because then effectively the n=1 trial is inspired by the RTC findings, but then do always trump them.

FIRST, DO NOT FAIL TO HELP

It really seems that psychiatry has transitioned to “A chance to give a pill, is a chance to help,” a version of the surgeon’s “A chance to cut is a chance to cure,” coupled with “First, do not fail to provide help.” Coupling this with patients with poorly defined conditions and high placebo/natural history improvement rates, along with an attitude that RTCs cannot trump one’s own experience, and that would seem a recipe for a lot of prescriptions being handed out.

I had asked “So if you were a geriatric psychiatrist and your clinical experience is that atypicals are helpful in the setting of severe behavioral disturbance in dementia you go forward with that, the results from studes notwithstanding?” Dr. Carroll rightly pointed out that the limited utility of the terms “atypical” and “severe behavioral disturbance.” He also provides some genuinely thoughtful clinical examples: “But, yes, I might very well use an antipsychotic drug for an elderly agitated patient who is delusional and hallucinating while I work him up for causes of delirium. Likewise for a patient with Lewy body dementia who has clear organic delusions on which she is acting out. I know studies say that may shorten her life expectancy, but my goal is to improve her quality of life while she is still with us. The longitudinal N = 1 experiment allows us to evaluate the progress toward that goal, with dose adjustments and trial withdrawals.” But RCTs do not suggest that an antipsychotic should not be used short-term in the setting of delirium. Nor in relation to the second example do the (I believe) suggest that they would not impact her delusions and possibly improve her quality of life. They both sound like examples of Dr. Carroll being a thoughtful clinician. However, some geriatric psychiatrists were using antipsychotics wholesale to try to treat relatively poorly defined chronic behavioral issues in nursing home residents with likely Alzheimer’s dementia. If they say that their clinical experience is that this is effective and safe, and large RCTs argue otherwise, that is closer to the issue I was trying to raise.

8,000 LITTLE KOLs AND 25 MILLION CHILDREN 0-5

Here is the context for why I am so concerned about what can otherwise sound like an academic discussion:

1. I am beginning to suspect that child psychiatry is attempting to position itself as a specialty more akin to radiology. As specialists who will be presented with a distillate (e.g., clinical information and maybe short video clips of a child), or meet very briefly with a child one time after a review of information, and will much spend more time is such review and interface with other clinicians, than in direct contact with children and their families (or therapists and teachers). In a sense the field will be reformed into the image of its leaders (who as 1BOM rightly points out are often having little contact with children and their families, and even less detailed longitudinal contact).

8000 little mini-KOLs.

2. I am also seeing increasing signs of the expansion of psychopharmacologic treatment into the 0-5 year old age group. To those who would argue that this is alarmist, and will be done only with utmost caution, I would say look at the 5 and up group. If 30 years ago you had suggested that 5 year olds from chaotic/traumatizing home environments would be brought in to see child psychiatrists and, after a total of 1 hour of record review and contact with the child and family, be sent out with prescriptions for antipsychotic or antiepileptic medication people might have looked at you funny. Now in many places it is standard of care. Given the mindset that appears to be in place at AACAP regarding medications, it will be easy for child psychiatrists to believe that the alternate non-medication approaches suggested in the practice parameter have either been exhausted or are not available, that there is a strong family history of bipolar disorder after all (by now this trend has been so long in place that it is skewing the family histories of young parents of young kids), and that the situation is too dangerous to not intervene. They’ve been hitting their sibling after all. There is a young baby in the home, …etc. So the n=1 experiment of cycling through meds for a 0-5 kid will begin. Let’s say you take your 2 year old to see a pediatrician, say you are convinced he has a seizure disorder, and say you want them on meds for it. Maybe you had some prenatal/birth factors that are concerning. Maybe an uncle has an ill defined seizure disorder. You say something has to be done or you don’t know if you can take it anymore. Pediatrician briefly consults a neurologist and an EEG is scheduled, then maybe a sleep deprived video EEG study is done. At the end of the day if there are insufficient findings on those (and the clinical findings are not overwhelming) the neurologist is usually not going to green light the pediatrician giving the child and antiepileptic drug. He/She might actively discourage it. This may end up in undertreatment of some kids, but makes a huge huge difference to overtreatment. There are likely going to be many more instances where no medication is provided at all than when medication is provided. Everyone gets paid even if no neurological illness is diagnosed. Contrast this to what will happen when you take your 2 year old to see a pediatrician, say you are convinced he has an electrical abnormality/bipolar disorder/something that it leading to violent outbursts than none of your other kids have. Maybe you had some prenatal/birth factors that are concerning. Maybe an uncle has some ill defined bipolar spectrum illness. You say something has to be done or you don’t know if you can take it anymore. Pediatrician briefly consults a child psychiatrist. I’ll concede that perhaps if this were Dr. Carroll’s chosen population, and he were being consulted, that there would be very extensive evaluation and intervention before turning to medications. I harbor strong doubts that will be the standard of care in the not too distant future. Your post at Health Care Renewal at the address you provided is a good one and I would recommend it to anyone who hasn’t already clicked on the link. But, I would still argue that the kind of pragmatism you describe, which seems to often involve going ahead and giving the best pill you can think of (rather than a pragmatism that more often involves not giving any pills), is more endemic to psychiatry than some other fields of medicine. Particularly pediatrics.

This is not even counting those voices, already present within the child psychiatric community, calling for prophylactic intervention with medication for young children of parents with mood disorders.

GIVING CLINICAL EXPERIENCE AND RCTs THEIR DUE
Drs. Carroll and Dr. Shorter. I said that both seem to be arguing that the wisdom gained from a combination of direct clinical experience and one’s clinical deductions, unblinded and warts and all, should be allowed to take its rightful place alongside RTCs, and not be relegated far down the alphabet while RTCs sit alone in category “A.” I also said that is more than fair, but that adequately designed RTCs should not always be trumped (and if none are adequately designed then that should be addressed).

I would add that it is also fair to then ask what heuristics can be used to better systematically address the biases that arise from direct clinical experience. That these are deserving of as much discussion as those used in the context of clinical studies. Particularly by those who rightly champion the importance of direct clinical experience and also rightly point out that studies will never be able to directly address all the clinical problems physicians face. Please note that I am speaking of systematic heuristics that allow for refinement of clinical judgment over time, NOT heuristics that are directly used as a stand in for clinical judgment (such as treatment algorithms).This may come naturally to you Dr. Carroll, but I suspect that years of genuflecting in front of the gold standard idol of RCTs has left many psychiatrists with a weaker natural sense than yours of how to adequately derive clinical wisdom from raw clinical experience.

I would close with again emphasizing that there should be a place for studies that are adequately designed to be able to influence physicians not to use a treatment at all with some populations. Dr. Shorter, as a scholar of the history of medicine, is this not the mechanism by which surgeons moved away from radical mastectomies that were much more frequently performed at one point in time based on the shared clinical wisdom of the time? I am not familiar enough with that story, so perhaps the change came not from studies but more from direct clinical experience. Either way, it might be an instructive example and one that may or may not have parallels in modern psychiatry. Perhaps the use of unilateral vs bilateral ECT? My impression is that the shift from the former to the latter has been more driven by studies, despite what had been felt to be the direct experience of some practitioners. Again, I would defer to you on that since it is an area of particular expertise for you.

Sorry, meant to say the shift from bilateral to unilateral ECT, not the other way around.

Anonymous’s thoughtful comments certainly deserve a reply, if only a brief one. Until now in this thread, we have been going on the assumption that there are only two kinds of knowledge in clinical medicine: (a) scientific, based for example on RCTs; and intuitive, based on rich clinical experience. But there is a third kind of knowledge that guides many clinicians: fashion. Whatever is on the field’s radar is often what steers individual clinicians. We don’t abandon agents, such as the barbiturates or the amphetamines, because RCTs have ruled them ineffective; we abandon them because it is no longer fashionable to prescribe them (and doing so can expose one to legal action). Diagnoses such as melancholia go off the boards not because they have been demonstrated not to exist, but because such terms have become unfashionable, receding in the face of “major depression.” This third kind of knowledge — fashionable knowledge — is worse than either of the first two because it has no scientific footing at all, and because clinicians are often obliged to deny the reality of their own experiences in order to conform to it. There are many clinicians who firmly believe there are two very different kinds of depression but feel constrained by fashion to prescribe SSRIs for both. This is a real defeat for a field that aspires to clinical science.

I appreciate Dr. Shorter’s reply and will reflect further upon it as there is much of worth there. 

I remain concerned that if a physician with a lot of experience downplays the importance of RCTs to their own practice, pretends like they can read the literature with a learned enough eye to make sense of it, and filter out the effect of the culture of the field within which they practice (often impacted by these “landmark studies”), then he/she is more at risk of missing this key point from Ben Goldacre’s Introduction to “Bad Pharma”:
” While the deceit of a marketing drive can be ignored by an ethical doctor, the problems caused by distorted evidence affect everybody, without exception. ”
To me, it is only when the majority of physicians, from neophyte to master, truly internalize that “everybody, without exception” part, and feel outrage around it, that physicians will become a force for real systemic change. 

Reading Ben Goldacre’s introduction also made me truly wish that individuals in Zero to Three, in the American Academy o Pediatrics, and others committed to infant mental health, would all read some or all o that book and then read these statements from Dr. Hendren during his recent tenure as president of AACAP:
http://www.aacap.org/cs/root/AACAP.Transparency
– while placing Dr. DelBello and Dr. Ryan on conflict of interest conflict committees and Dr. Bostic as head of the CME committee
&
http://psycnet.apa.org/psycinfo/2007-18374-001
– Dr. Dulcan and letting the science shine through
And ask themselves, if they come to be impressed by even a few of Goldacre’s points, if Dr. Hendren’s AACAP is really what they want to ally with in their mission to serve infants. 
As Ben Goldacre seems to be saying, everyone has choices to make, and those choices will have real impact on the lives of people, including infants seen by pediatricians. 
If AACAP’s stance to date is indicative, do you trust them to advise pediatricians about the use of pharmacology in infants. 
Most importantly, when NOT to engage in its use?

1BOM, please let last comment post.

Annonymous, you clearly have your teeth into a slew of issues that will keep you busy for a long time. I wish you well with your efforts.

Dr. Carroll, I hope you were sincere in your comment. Either way, by accident or by intent, you highlight the little impact that my pontificating will have. In addition, since those efforts are unlikely to pay the mortgage, I will most likely simply be able to reuse these comments in another blog several years hence for all the good it will do. This, in the end, is the rub. I disagree with you on some of your points, but very much agree that there has been good clinical research work done in psychiatry and wish you luck in trying to continue to promote that work.

I seem to agree with a lot of folks that there are different kinds of ways to generate knowledge the various knowledge gleaned can be generally weighted based on how they were developed. I do believe that strong, meaningful RCTs should rank higher than clinical intuition. There is something to be said about the rarity of meaningful RCTs in psychiatry, but it doesn’t mean when we get one we shouldn’t consider it important. There is also something to be said about strong research that indicates how variable, inconsistent, affected by extraneous information, etc. clinical intuition is. At the end of the day, I do believe it is sometimes the best knowledge we have, but I do not believe we should be content in that. It should humble doctors if they come to a decision that is really only informed by intuition (sometimes also known as chance), not glorified as an example of the artistry of medicine. In fact, I often see when decisions that weren’t based on sound research lead to good outcomes, it’s all “what a masterful and artistic practice of medicine, how did you become such a genius!?” When decisions that weren’t based on sound research lead to bad outcomes, it’s all excuses “We never could have predicted this. Hindsight is 20/20. this was just so complicated. Doctors are bound to get in wrong sometimes.” The first situation should be tempered with the idea that even by chance, doctors or possibily unrelated factors will help get things right sometimes (as opposed to giving credit to an artistic doctor), and in the second, doctors should be clear and upfront with patients about the limits of their knowledge, build a case for what they plan to do with the best evidence available (aka build trust), recognize the risks and uncertainties from the get go, plan some contingencies, and not that a lot of contingencies can’t be planned for.

The trust me I’m a doctor comments really rang untrue for me. I trusted doctors precisely because I thought they were adept at synthesizing and adapting clinical research of various quality to particular situations, with attention to their own judgement (and systemic inquiry of their own judgement) and patient values/access/funds/etc. I really no longer believe this. Doctors, like all people, make decisions based on whim, erroneous factors, bad data, bias, social pressure, personal/financial interests, and whatever. The idea that we should just trust them more than anyone else seems hallow to me, regardless of their oathtaking (which really could just obfuscate the realities at hand). I really am ok with the idea that doctors will get stuff wrong a lot. Like it or not, they are the best positioned to help in a lot of situations. I just wish then they were better at assessing and expressing uncertainties of treatment outcome before treatment is agreed upon. While of course we cannot exactly predict the course of a particular treatment (not even close), because we have so much “hindsight” data, we can calculate probabilities of outcomes based on past data. We can be better at prediction if we had better RCTs too. As a patient, I would like to be informed about this before I “trust” anyone or agree to anything.

I am not ok with the idea that I should trust someone just because they are a doctor when I can assess that their thought process about clinical decision-making is flawed. I value the Hippocratic Oath, but I don’t think that it should be used as justification of trust. Even with the best intentions, doctors can and do often not put the interest of their patients first or disentangle their personal interests. Many just don’t utilize research well, or aren’t able to integrate it with their personal beliefs/interests and those of their patients.

This has been kind of rambly, but I’ll end with the idea that this emphasis on trust is actually the basis from which particularly psychiatrists have “healing power.” I think it is why doctors promote and defend the idea that they should be trusted without too much questioning. I do believe that that trust has led to a lot of benefit for some people (though I might critique the kind of social arrangements we have to solidify in order to allow for such unquestioning trust), as 1BOM has demonstrated repeatedly on this blog, the scientific data on which trust in doctors is greatly build upon is poor and doctors engage in behavior (intentionally or not) that leads to personal and/or professional gain at the expense of patients/research participants. To me, lack of applicable research findings and customary behavior that is not in line with the Hippocratic Oath does erode trust in medicine, and because at the end of the day it seems trust is a big factor in medicine’s healing armament, this is sad news.

All good points. To drill down further on a couple:
1. What would you name as two or three “meaningful” RCTs in psychiatry that suggest (or have suggested) a change from practice derived from clinical intuition?
2. Listings are plentiful of the potential biases inherent in the clinical wisdom directly derived from a physician’s own experience. HOWEVER, RCTs will never be able to replace this is the majority of instances. In those instances, what do you suggest as mechanisms that would allow physicians to better filter those biases while GATHERING the data from their own experience (not just at the point of decision making). Studies can more straightforwardly use blinding, randomization, intent to treat, preset endpoints. What are the equivalents you believe an individual physician should use in the gathering stage? Because, by the time they reach the decision makin stage, if there isn’t a sufficiently on point meaningful RCT, they have to go off what is essentially opaque (to extent even to them) pattern recognition process. Put another way: studies apply most of their techniques for compensating for bias PRIOR to the analysis of results stage. No one seems to address possible techniques for individual physicians to be able to apply (albeit imperfectly) in their own direct clinical practice because all of that is supposed to be taken care of through the use of larger studies such as RCTs. You’ll seeany discussion of how one RCt is significantly methodologically superior to another RCT. When is the last time you saw a discussion of how one case series from one physician was significantly methodologically superior to another and this should be accorded more weight? The number of permutations in medicine precludes the direct application of RCTs in many instances. My argument with Drs. Shorter and Carroll is that there should be at least a few instances where the field can have RCTs that eliminate interventions. E.g., you shouldn’t be lobotomizing feisty young women, rather than having o wait for it to go out of fashion. It can be easy to overromaticize the food old days in psychiatry. However, I am in agreement with them or the majority of instances that seem to be out there. I.e. it’s not an algorithm you can easily nail down at the decision making point and it appears strongly influenced by the physician’s personal database. Their own internal non-published papers if you will. For older physicians this can outstrip the n of even some of the metanalyses out there. It seems though, that at the gathering stage for that data, physicians operate very similarly to the way they did 60-70 years ago. If one believes that the PROCESS (obviousloften content of that knowledge is different now) of acquiring and acting upon clinical wisdom in clinical psychiatry in the 40s and 50s was just fine thanks, then there is no problem. This can include using studies to try to better elucidate underlying processes that can inform clinical practice, such as Dr. Carroll’s work with the DST. If you believe that RCTs are going to do all the heavy lifting, then you also don’t care if the techniques by which physicians gather data from their own experience (outside of clinical trials) hasn’t become much more sophisticated in the past 50 years.

Much as I am impressed by your overall argument, and strongly agree there have to be some sufficiently meaningful psychiatric RCTs out there (would love if you could name a few as I mention above), it seems to me that in the majority of instances physicians can’t currently use them as a direct guide and there is likely enough opacity to their own understanding of the pattern recognition they use (many high level decisions are this way) that to some extent they are saying to themselves “I trust myself because I’m a doctor.” You rightly point out that there is going to be bias in how they attribute tx success vs tx failure (sometimes in the opposite direction from which you describe). OK, now what? Have you seen any suggestions on techniques they can apply as they encounter these examples that would allow them to come out the other end with a less biased set of results/conclusions to be used during the pattern recognition stage. If not, then in a way I think we sort of are stuck with Dr. Carroll’s statement that we’re just going to have to get used to it and pray that physicians at least start to deal more with the commercial biases Dr. Carroll rails against.