SAMSON Answers the Statin Side Effect Question

John Mandrola, MD


November 15, 2020

The novel and elegant SAMSON trial largely answers the lingering question of statin side effects.

This has major implications because statins may be the most studied class of any drug. More than 130,000 patients have been enrolled in placebo-controlled trials; the results show a consistent 25% relative reduction in future cardiac events.

Yet many patients discontinue the drug because of adverse effects, most often myalgia. Large observational studies report a 20% to 50% higher rate of myalgia in statin users vs nonusers.

But in blinded randomized controlled trials (RCTs), people taking statins have almost the same rates of minor adverse effects as those taking placebo.

One way to explain the discrepancy between the observational studies and the RCTs is that statin trials have run-in periods that exclude patients who report adverse effects. Another explanation is the nocebo effect—when a negative expectation produces a negative outcome.

An Italian group has published one of the best examples of nocebo effects. They gave the β-blocker atenolol to three groups of men: one group was given no information, another group was informed about the drug but not any side effects, and a third group was told about possible erectile dysfunction. The rates of erectile dysfunction in each group were 3%, 15%, and 31%, respectively.

The SAMSON Trial

In the SAMSON trial, presented at the 2020 American Heart Association (AHA) Scientific Sessions and simultaneously published in the New England Journal of Medicine, James Howard and colleagues at the Imperial College London tested the nocebo effect of statins.

They used the so-called n-of-1 trial design in which each patient serves as their own control. This design is nifty because it mirrors what clinicians do in practice. You start a drug and see if it makes the person feel better; if you aren't sure, you stop it, observe for a time, and then restart.

The key difference in SAMSON is that patients and doctors were blinded to the treatment. The trial randomly assigned 60 patients who had previously stopped statins because of side effects to 12 one-month periods consisting of no medications, placebo, and statin. All but 11 patients completed the 12-month trial.

Patients recorded symptom intensity (0-100) each day on a smartphone app.

The primary outcome of the study was the ratio between excess symptom intensity caused by the placebo and excess symptom intensity caused by the statin.

The graph below depicts how the nocebo proportion was calculated. First, the researchers recorded the difference in symptom scores between taking no pills and a placebo. This is the nocebo proportion (ie, the ill effects from the negative expectations of the pill). The authors then recorded the symptom score from the statin, which includes both the nocebo effect and the pharmacologic effects of the drug.


Three Main Results

The mean symptom score for all patients was 8.0 during no-pill months, 15.4 during placebo months, and 16.3 during the statin months. The difference in symptom scores between the placebo and statin months and the no-pill months was highly significant (P < .0001).  However, the difference in symptom scores between the placebo and statin months was not significant (P = .39). To calculate the nocebo effect, they divided the symptoms score of the nocebo component by the statin proportion, for a nocebo ratio of 0.9:

symptom intensity on placebo (15.4) – symptom intensity on no pills (8.0)


symptom intensity on statins (16.3) – symptom intensity on no pills (8.0)

The authors concluded that patients really do get symptoms from taking statin tablets. But, crucially, 90% of these symptoms are elicited by placebo pills too.

Statin side effects, therefore, are mainly caused by the act of taking pills. Six months after completing the trial, 30 of the 60 patients who were shown their personal results had successfully restarted statins.

My Comments

The results of any trial ought to be considered in the context of prior data.

Before SAMSON, the evidence for nocebo effects of statins was strong. The most convincing statin RCT in this regard was the lipid-lowering arm of the ASCOT trial, which compared atorvastatin vs placebo in 10,000 patients. ASCOT-LLA had both a blinded phase and open-label continuation. In the blinded phase, there were no significant differences in muscle-related symptoms, but in the open-label phase, significantly more patients in the statin arm reported muscle symptoms.

We clinicians can use the SAMSON results the same way the Imperial College researchers used them: to convince some patients previously intolerant to statins to restart a drug that lowers cardiac events. That is a big deal.

Another general message of SAMSON is for clinical trialists. If expectations can have this large of an effect, imagine the expectation differential between those in the procedure arm of, say, an atrial fibrillation ablation vs drug trial. Or in a percutaneous coronary intervention (PCI) vs medical therapy trial? Expectations are especially relevant with measurement of subjective bias–prone outcomes, such as quality of life.

My favorite message of SAMSON is how it informs the art of medicine. Experienced clinicians well know that getting the best outcome from a drug or procedure transcends pharmacology or physiology. It means managing the complex neurophysiology of placebo and nocebo effects. Such effects may stem from simple expectations of reward or harm, social learning (think Internet), and even Pavlovian conditioning.

Clinicians must be mindful of our words and actions. Yes, patients ought to be informed, but it also makes sense to use all the tools of human nature to get a positive outcome. The practices of showing a patient and her family a beautiful angiographic result after a PCI or being optimistic after an AF ablation should not be discounted. And when we prescribe a treatment, there is every reason to be positive.

The word "doctor" originated from the Latin verb "docere"—to teach. I will save the SAMSON trial on my phone or computer and use it to teach patients about the power of expectations and the nocebo effect of statins. When, for instance, a patient reports statin side effects, rather than reduce the dose of the statin, which reinforces that the drug is causing the symptom, a better tack would be to pull out the SAMSON trial and teach about the nocebo effect.

Yet empathy remains vital. Patients who report statin side effects are having real symptoms; they shouldn't be dismissed. SAMSON simply says the side effect isn't from the statin but from the statin pill.

The SAMSON investigators have beautifully shown that being a healer means more than writing a prescription or moving a catheter. For that I say, thank you.

John Mandrola practices cardiac electrophysiology in Louisville, Kentucky, and is a writer and podcaster for Medscape. He espouses a conservative approach to medical practice. He participates in clinical research and writes often about the state of medical evidence. 

Follow John Mandrola on Twitter

Follow | Medscape Cardiology on Twitter

Follow Medscape on Facebook, Twitter, Instagram, and YouTube



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: