I hope you will forgive the mini-barrage. I don’t usually clutter inboxes, but the timely publication of a trial that fits, hand-in-glove, with the current series was too much temptation.
To understand what’s gone wrong in cholesterol research, and what my 4-part series is about, you don’t need a theory. You just need a trial.
The one published last night in the New England Journal of Medicine, and delivered straight to my inbox, is as clean an example as you’ll find—and it’s right on time.
Patients with heart disease were randomized to more intensive lipid lowering (as recommended by the AHA) or usual care. The result, we’re told, was a resounding success: More lipid lowering led to fewer coronary events, YAY!!
Another win for cholesterol reduction?
Let’s look inside the numbers.
Mortality—what patients care about—was 2% in the treatment group and 1.9% in the control group. No difference. Not just insignificant, untouched. (In fact, numerically higher in the lowest lipids group).
So why is the trial a ‘success’, with a fawning editorial that calls the results “firm support” for current AHA guidelines?
Because the primary endpoint wasn’t death, or anything else people taking statins care about. It was a composite of fluff, with a ‘benefit’ driven by two things: Nonfatal MI, and coronary revascularization.
Sound familiar?
The study’s final tally was 100 coronary events in the intensive group versus 147 with usual care—a difference of 47 events. That difference included 14 fewer nonfatal MIs and 41 fewer revascularizations. This constituted the entire difference, and then some.
Sound familiar?
Now pause. (I need a breath, or my head will pop off).
If those events had represented meaningful heart injury, as they were supposed to—heart attacks that damage myocardium, or lead to heart failure, or shorten life—then a difference of that magnitude should easily have left a trace on mortality.
It didn’t.
That’s because modern ‘nonfatal MI’ in trials is a troponin-based construct (see part 2) unassociated with cardiac peril.
And revascularization is even worse.
We have decades of trial evidence showing that, in stable disease, stents DO NOT prevent heart attacks, strokes, or death. Yet here it is—a false surrogate for those outcomes, and therefore the primary driver of a false benefit.
In other words, this trial changed people’s blood tests and billable procedures, not their lives.
Sound familiar?
This is not an anomaly. It is the system working as intended.
The AHA has made it clear that what matters to them is not whether people live longer, but whether meaningless endpoints move. Researchers respond accordingly. Trials are built on outcomes that are easy to manipulate, and count—and easy to publish in the vaunted NEJM.
And so we’ll continue to see studies like this. Technically correct, methodologically sound, and clinically fraudulent.
Because once you accept nonfatal MI and revascularization as surrogates for death or disability, what patients care about no longer matters.
And that is the Big Mistake.
