1 Boring Old Man

ro·bust     /rōˈbÉ™st/
  adjective:
1. (of a person, animal, or plant) Strong and healthy; vigorous.
2. (of an object) Sturdy in construction.

Why have so many lightweight scientific studies of the efficacy of psychiatric drugs had such a profound impact on prescribing habits of the last several decades? From the present vantage, the second round Anti-depressants [SSRIs] and Antipsychotics [Atypicals] are turning out to be anything but ro·bust, yet they largely replaced their predecessors as the drugs of choice. More than that, they’ve become "blockbusters" in the marketplace, topping the charts in overall pharmaceutical sales – so much so that the specialty of psychiatry has been largely redefined as ‘clinical neuroscience’ [meaning psychopharmacology]. One thing guaranteed. When the definitive history is written, the answer to the "why" of all this is going to turn out to be a lot more complex than the unidimensional and often demonic explanations we hear in our current period of disillusionment. But that’s for our children to figure out.

Speaking strictly for me, I had no idea that the studies were so ‘lightweight." One thing about thinking out loud [blogging], you can look back at what you were thinking about in the past. Two years ago today, I was drawing graphs of the US Income Tax structure trying to understand our financial predicament [taxing thoughts…]. One year ago, I was still at it, trying to understand how Alan Greenspan’s earlier associations with author and polemicist Ayn Rand had affected his sometimes disastrous economic decisions [no freedom without walls…], but I was just beginning to look at modern psychiatry [consequences…] in the wake of the chairman of the department I had been a member of for over thirty years being dethroned in scandal [Dr. Charles Nemeroff]. I suppose that it’s the inertia of motion of a career as a psychoanalyst and psychotherapist. If you spend your every working day poring over the details of patient’s histories trying to parse out the roots of the contemporary problems, it’s how you approach everything else.

So, one year ago, I was just learning that the Atypical Antipsychotics were the top-selling drugs. I didn’t know that LOCF stood for Last Observation Carried Forward, a method of correcting for the fact that people regularly quit clinical drug trials in droves. I had no clue that CRO meant Clinical Research Organization or that they existed. I didn’t know about ghost authors or guest authors. I knew to look for industry funding, but it never occurred to me that the authors didn’t do the study and were often picked after it was done. I didn’t know to go to the acknowledgements to read the Conflict of Interest declarations. Speaker’s Bureaus with industry prepared slides? You’ve got to be kidding. I didn’t know Harvard meant Joseph Biederman, Stanford meant Alan Schatzberg, or Brown meant Martin Keller. I knew that Charlie Nemeroff meant Emory, but I didn’t even know what that really meant. Since last year, I’ve gotten to know clinicaltrials.gov, PubMed, the NIH RePORTER, Ovid, ProPublica, Drugs@FDA, the sites where discovery court documents are archived, and I see Google searches in my dreams. But a year ago, I knew none of that. I’ve had a lot of help along the way [for which I’m eternally grateful]. And I’m a retired old guy with lots of time with a hard science background who is chronically curious. But my point is that a year ago, I was an average expectable psychiatrist, and I was in the dark – even about the chairman of a department I was a peripheral part of. I had no clue that the studies backing up these drugs were so not  ro·bust. I knew the ads were malarky [they always were]. I knew Dr. Nemeroff was full of hot air [he always was]. But I didn’t know about the widespread deceit and deliberate scientific misinformation [that hasn’t always been].

So this year, I’ve learned a lot of things that I shouldn’t have to know. No practicing physician has the time to chase down the FDA reports on drugs, review the appropriateness of statistical tests, learn to calculate effect sizes or NNT [number needed to treat] values, check dropout rates and look to see if they used LOCF or Mixed Model corrections. That’s what Journal editors do [did]. That’s what experts are [were] for, "thought leaders" are [were] to think about. And there’s a limit to what can be done in their absence. I can maybe do a decent job of evaluating a Clinical Trial article now using all those tools mentioned. But I doubt that I could ever evaluate a genomic study with a gajillion SNPs using the statistics involving super-nova sized matrices – not sitting up here in the woods looking at the Internet.

So that’s my message. I shouldn’t have to learn these finer details to know about the drugs, but it’s possible with time and effort. But I’m not even sure that I could learn to evaluate the masses of data generated in a genomic study, even if I gave it my best shot.  That’s why I’ve been so obsessed with this "personalized medicine" business. If our researchers are as corrupt and deceitful as they have been with the drug trials, we’re up a creek with no paddle, because of the volume and complexity of the information. And it matters to our patients because there’s a powerful move afoot to treat people with drugs based on some commercial lab’s printout on their cheek swabs [maybe done at the same time as the waiting room depression screening questionnaire]. That’s no good in my book. We need ro·bust results before any piece of that scenario comes close to happening…