“Important Conversations” Are Needed to Explain the Nocebo Effect. Anita Slomski. JAMA, February 3, 2021. doi:10.1001/jama.2020.25840
Roger needed no convincing that taking a statin could prevent his early death. At age 52 years, he had mixed lipidemia, severe peripheral vascular disease, obesity, fatty liver disease, and a previous femoral artery occlusion. But as he explained to the investigators of the SAMSON (Self-Assessment Method of Statin Side Effects or Nocebo) trial, he’d already tried 3 different statins and discontinued each one due to the dreadful muscle pain he felt while taking them.
“He was totally gobsmacked when we unblinded the results of SAMSON and showed him that his worst months—including muscle pain so bad he couldn’t get out of bed—were from placebo,” said cardiologist James P. Howard, MB BChir, clinical research fellow at Imperial College London and co–first author of the SAMSON report published in the New England Journal of Medicine. After discovering that he reported feeling fine during the months of the trial that he received a statin, Roger resumed statin therapy with no symptoms for the 4 years since receiving his personal results.
The novel n-of-1 trial validated what physicians have long observed: patients’ negative expectations for statin therapy rather than the drug’s pharmacological action are often responsible for intolerable adverse effects. SAMSON, in fact, found that 90% of adverse effects from statins were explained by this nocebo effect. “The nocebo effect is a massive burden; in our 60 patients, side effects were so bad that they had to come off the tablets on 71 occasions,” Howard said in an interview.
The 60 study participants, all who had previously discontinued statin therapy because of intolerable adverse effects, received 4 bottles each of 20-mg atorvastatin and placebo, and 4 empty bottles. Each month for a year, participants took pills or nothing in a random sequence and recorded their daily symptom intensity on their smartphones.
“To work out the nocebo effect, it’s imperative that you have a nontreatment arm where the patient takes nothing so you can subtract the background symptoms that are ever-present, such as the aches and pains of getting older or of arthritis, for example,” Howard said. “As far as we know, this is the first time anyone has done such a trial.”
At the end of the trial, patients saw how they rated their symptoms during the 3 treatment sequences, which was compelling enough to convince half of them to resume statin therapy. “Only 18 of the original 60—less than one-third—told us that they weren’t restarting statins because they still believed they caused side effects,” Howard said.
Although trials in other journals including JAMA and The Lancet have reported nocebo effects in statin therapy, SAMSON stood out because the study design demonstrated to patients themselves that the nocebo effect is real.
The study’s participants had first-hand evidence that “just the simple act of taking a pill, where they might have been expecting side effects, explained much of the symptoms,” Donald Lloyd-Jones, MD, ScM, president-elect of the American Heart Association (AHA), told JAMA last fall during the AHA’s virtual Scientific Sessions conference.
However, some experts question the magnitude of the nocebo effect in SAMSON’s results. “It’s easy for me to believe that 50% to 60% of statin side effects are nocebo, but not 90%,” Steven E. Nissen, MD, chief academic officer and Lewis and Patricia Dickey Chair in Cardiovascular Medicine at the Cleveland Clinic, said in an interview. “Some patients who have tried very hard to take statins have a real disorder” that prevents them from taking statins.
Howard agrees that his results shouldn’t be extrapolated to all patients who take statins. “In a larger trial, you might find a 70% or 95% nocebo effect,” he said. What’s important “is that the nocebo effect dominates in a majority of patients on a statin and that real side effects are much rarer than we thought.”
For physicians, that means explaining the nocebo effect. “We have to have very important conversations with our patients rather than just writing a prescription, actually telling them what to expect,” said Lloyd-Jones, also chair of preventive medicine at Northwestern Medicine Feinberg School of Medicine.
A Bad Rap Fuels the Nocebo Effect
Although much less studied than the placebo effect, the nocebo response has been demonstrated in a variety of therapies in experimental and real-world settings. A recent review article in the New England Journal of Medicine cited several striking examples. When New Zealand pharmacies switched to a new formulation of thyroid hormone replacement medication, reports of adverse events increased 2000-fold, even though the drug’s active ingredient remained unchanged. Nearly a third of study participants taking the β-blocker atenolol for cardiac disease and hypertension developed sexual adverse effects and erectile dysfunction when they were warned of the potential side effects compared with 16% who weren’t informed of possible adverse effects. Patients have blocked the analgesic effects of the potent opioid remifentanil when falsely told it would increase pain.
“The nocebo effect has been described in biosimilars used in autoimmune diseases, when patients believe the drugs are less effective than the original biologics,” Luana Colloca, MD, PhD, the review’s first author and associate professor in the Department of Pain and Translational Symptom Science at the University of Maryland School of Nursing, said in an interview.
“We know that allergic reactions can be amplified by nocebo, such as people continuing to have symptoms of gluten-intolerance even after receiving a negative diagnosis,” Colloca added. And 30% of women receiving chemotherapy for breast cancer developed anticipatory nausea from previously neutral environmental cues, such as meeting an oncology nurse at the grocery store or being in a room painted the same color as the infusion room, her review noted.
The greater a patient’s negative perception of a therapy, the stronger the nocebo response. “The patients we give statins to are the same patients who get prescriptions for angiotensin-converting enzyme inhibitors, hypertension treatment, and aspirin,” Howard said. “Patients don’t start ramipril for hypertension and say they feel terrible. People view statins much more negatively and with more skepticism.”
Concerns about statins began when the US Food and Drug Administration (FDA) required statin labels to list rhabdomyolysis as a potential serious adverse effect after an early statin was withdrawn from the market. “That made some doctors edgy about statins,” and, in turn, patients, Howard said.
Then as professional societies advised that statins could benefit a greater number of people, a baseless claim began circulating “that statins were developed to enrich the pharmaceutical industry and that doctors are in bed with big pharma, pushing cholesterol drugs,” Nissen said. Companies selling “natural” products to lower cholesterol have also contributed to perceptions that statins are harmful.
A new study that tracked statin adverse effects reported to the FDA’s Adverse Event Report System found that significantly more nocebo-related subjective adverse events than harms substantiated by clinicians have been reported in the last decade. Complaints of nocebo-effect symptoms—but not objective adverse events—peaked whenever the FDA issued a statin warning. One such warning occurred in 2010 when an increased risk of myopathy was observed with high-dose simvastatin.
Bad publicity has also dogged bisphosphonates after reports emerged of women developing esophageal ulcers after taking the drug to treat osteoporosis. “These patients took the bisphosphonate incorrectly—dry swallowing it or taking it while lying down—and they refluxed alendronic acid into the esophagus,” David Karpf, MD, adjunct clinical professor of endocrinology, gerontology, and metabolism at Stanford University School of Medicine, said in an interview.
In the large population-based fracture-prevention trial that Karpf led, serious gastric adverse events were higher in the placebo group than in the bisphosphonate group. “We told participants that the drug is effective in preventing fractures and is generally well tolerated, and lo and behold, we had excellent compliance in the trial,” he said.
“I think the nocebo effect demonstrated in the SAMSON study is generalizable to any drug that has been studied in large populations and shown to be well tolerated but with some side effects, like bisphosphonates,” Karpf added. Drugs approved to treat asymptomatic chronic diseases have passed a high bar for safety, and, therefore, should be more tolerable to patients, he said. But at the same time, the nocebo effect may be stronger for drugs used to prevent disease in asymptomatic patients. “People aren’t getting any therapeutic satisfaction from taking a statin, but they are reading about muscle damage when they Google statins,” said Howard.
These Symptoms Aren’t Phony
Clinicians generally have an inkling about which patients may be vulnerable to the nocebo effect, such as those with a history of anxiety or depression. Other tip-offs are patients who say they’re very sensitive to medications or hate taking them or who mention a long list of symptoms that their previous physicians couldn’t diagnose, according to Arthur Barsky, MD, professor of psychiatry at Harvard Medical School.
Lack of trust in the clinician can also prompt a nocebo response. “A patient reporting side effects can often be a commentary on the doctor-patient relationship,” Barsky said in an interview. “If you aren’t sure your doctor has made the right diagnosis or you aren’t comfortable with your doctor, it’s easier to say you’ll stop taking a drug because it causes headaches than to say, ‘I don’t trust you.’”
Patients who report nocebo symptoms are feeling real distress—but misattributing it to the drug. In reality, their symptoms may be caused by aging, not eating well, stress, or the underlying disease itself. “For patients with difficult lives, side effects to statins can be a nidus for their emotional pain,” Jennifer Robinson, MD, MPH, professor of epidemiology at the University of Iowa College of Public Health and lead author of statin guidelines for the National Lipid Association, said in an interview.
It’s important for clinicians to acknowledge nocebo symptoms as real but “to discount their medical significance by telling patients that the symptoms they are experiencing aren’t harmful or aren’t an indication that the drug is dangerous,” Barsky said. “The more you are worried about what a drug will do to your body, the more you will monitor side effects and the more intense they will become.”
If patients appear hesitant about starting or continuing a particular drug, clinicians should ask what their worries are, Colloca suggested. “The nocebo effect can occur if a patient has incorrect information about a drug or has had prior negative medication experiences,” she said. Physicians can point to trials of the drug showing that participants in the placebo group had similar adverse effects as those on the active drug. “Allaying the patient’s concerns can make the drug more tolerable,” Colloca said.
Physicians may also be able to head off a nocebo effect by emphasizing a drug’s efficacy, tolerability, and safety rather than mentioning rare adverse effects. “I tell patients that statins have been studied in a quarter of a million people and are safer than aspirin,” Robinson said.
When switching patients with rheumatoid arthritis from a biologic therapy to a less costly biosimilar, Roy Fleischmann, MD, clinical professor of medicine at the University of Texas Southwestern Medical Center, takes pains to explain the efficacy and safety of the biosimilar and that the vast majority of patients respond well to it. “There is a perception—among physicians, too—that if a drug is cheaper, the quality is not as good as the original biologic,” Fleischmann said in an interview. “It’s important for physicians to assure themselves and convey to their patients that the biosimilar has been manufactured according to FDA standards.”
Although clinicians must disclose a drug’s potentially dangerous adverse effects, patients can decide if they also want to be informed about potential minor adverse effects. So-called authorized concealment avoids priming patients to experience adverse effects through the potent power of suggestion.
Among patients who initially say they cannot tolerate a statin, up to 90% can successfully return to daily moderate or high-intensity statin therapy when physicians use strategies to mitigate the nocebo effect, according to Robinson. It may take trials of a few different statins or starting a patient on a 5-mg dose once a week and gradually increasing the dose to overcome nocebo adverse effects, she said.
SAMSON’s Howard said he’ll rechallenge patients with a different statin but disagrees with the strategy of inching patients along on low doses to increase tolerability. “You can’t tell patients that the side effects aren’t caused by statins and then start them at a low dose of another one,” he said. “Either you believe that the side effects are due to the nocebo effect, or you believe they are biochemical and then you go with a low dose. Sending mixed messages isn’t helpful.”
For patients at low risk of myocardial infarction or stroke who continue to experience muscle pain after trying 2 different statins, Howard will switch them to the nonstatin ezetimibe. “But if the goal of the statin is for secondary prevention, you are duty-bound to try a lot harder with these patients, whether that means rechallenging with another statin or using a PSCK9 [proprotein convertase subtilisin/kexin type 9] inhibitor,” Howard said.
“We have effective drugs to treat major diseases that have a huge societal impact, such as diabetes, heart disease, osteoporosis,” said Stanford University’s Karpf. “But we need to work harder to improve patients’ adherence to these lifesaving therapies. I think the SAMSON study is one step in that direction.”
Posts
