Clinical Trials: Time to call for integrity

Sep 13, 2009 21:18 EDT

When I returned from Barcelona , I told my US colleagues that trials presented at the ESC have a great deal of impact upon daily practice.  As a quasi-journalist, I don’t have to work to make the news interesting because there is very little esoterica, less bias, fewer political agendas, more presenters with no disclosures and  so much fodder for other avenues of interest and testing.  The most sinister undercurrents I see include trials vying for the attention of the EU to act more responsibly on smoke free legislation, monitor the sodium content of food supplies and spend money on prevention. Compare those studies to those reviewed by heartwire and published in PLos Medicine and JAMA .   We need more integrity in our trial designs, our trial interpretation and in implementation.

We need fewer “shotgun” trials like the one we heard a couple of meetings back.  Would anyone in their right mind walk into my ICU today and say, “every patient gets a beta blocker regardless of age, blood pressure or heart rate?”  That trial was much like an open bar call, but instead it was “beta blockers all around”  (no I haven’t started drinking, but it seemed to be an appropriate analogy). Is there any wonder that those patients didn’t do so well when we sprinkled beta-blockers like shotgun scatter ?

What about the trial done just for the sake of doing a trial? Remember the one where an ace inhibitor failed to improve “normal EF diastolic heart failure” yet not one patient got a single EDP measurement or a measure of diastolic filling patterns by echo. It was like we enrolled patients at the local Wal-Mart asking everyone coming out the front door if they were short winded.  If the answer was yes, they got an ace inhibitor and then were just brought back in a few weeks and asked if they felt better.   It was the only trial I’ve ever seen to test the efficacy of a drug on cohorts who probably didn’t even have the disease.

Then there are the cloak and dagger presentations where the presenter is bought and sold by the pharmaceutical company.  His trial is negative and he knows it when he gets up to the podium but he’s found the “needle in a haystack” subset that benefited from the compound.  It might be something like “ patients whose birthdays occur in a month that end in  “er” demonstrated a substantial benefit but the presenter will quickly add that the trial wasn’t powered to address this issue.”  He secretly hopes we’ll rush out of the meeting like minions and prescribe like crazy his compound until the next trial proves it doesn’t work.  Heck, it might even be downright dangerous, but that doesn’t matter as long as part of the investment was recovered.  Thankfully, I don’t think there are a lot of these types of presentations.

Then there are the studies that are the greatest of my pet peeves:  those with the all encompassing and confusing composite end point. The presentation always begins with   the usual statement: ‘We met the composite endpoint (I scribble “positive trial” in the margin of my notepad) which included death, stroke, heart attack, the prevention of 7 consecutive days of subzero temperatures in southern Florida and hospital re -admit rates”.  Then I study the graphs; hmmm…. It met the endpoint by preventing sub zero temps but failed at death, stroke, heart attack and readmit rates.  –(scratch note in margin ). Why not just report on every single endpoint as positive or negative and power the trial to address the very questions that are being asked??

I readily acknowledge that trial designers have a very important, complex task.   We won’t always agree on the implication and  like the bible, outcomes are subject to broad interpretation.  But I do insist that before we spend money and put patients at potential risk, both the premise and the endpoints must make sense and the questions must be worthwhile. Most importantly, improvements in patient welfare (not the pharmaceutical company portfolio) should be the primary goal.

If we as physicians and scientists would all live by the old Alan Simpson quote “ If you have integrity, nothing else matters. If you don’t have integrity, nothing else matters,  “ then henceforth, every single trial would matter.