Side effects are real, and so are the ones fear creates

Key takeaways

• The CDC estimates that harms from medicines lead to more than 1.5 million U.S. emergency department visits a year.
• In a 2020 trial of people who had quit statins over side effects, dummy pills produced about 90% of the symptom burden the real statin did.
• When a drug label shows a placebo column, the gap between the two columns shows roughly how often the drug itself caused a problem.
• Supplements cause harm too: CDC researchers tied them to about 23,000 emergency visits a year, and St. John’s wort can weaken birth control pills.
• Before you stop a prescription, ask your prescriber about the need, the dose and a safe plan to taper.

Sixty people who had given up on statins, the common cholesterol-lowering pills, agreed to an unusual test. A team at Imperial College London gave each of them 12 bottles, one for each month of a year¹.

Four bottles held a statin, and four held dummy pills that looked the same. Four were empty, meaning a month with no pills. Every day, people rated their symptoms on a phone app, from 0 to 100¹.

In the empty-bottle months, scores averaged 8. In statin months, they averaged 16.3. In dummy-pill months, they averaged 15.4¹. The fake pills produced about 90% of the symptom burden the real drug did¹.

The aches and tiredness were real; what people had gotten wrong was the cause. After seeing their own results, half of them were back on a statin six months after the trial ended¹². The trial included only people whose symptoms began within two weeks of starting a statin, so it may not apply to everyone².

If fear of side effects is part of why natural approaches appeal to you, that fear is not foolish. Drug harms are common and well documented. But expectation can make people feel sick too, and plants and supplements carry risks of their own. Here is how to weigh all of it with the same clear eye.

Harms from medicines lead to more than 1.5 million ER visits a year

The Centers for Disease Control and Prevention (CDC) tracks adverse drug events, meaning harms caused by a medicine. It estimates they lead to more than 1.5 million emergency department visits in the United States each year. Nearly 500,000 of those visits end in a hospital stay³. The CDC’s fact page, last reviewed in April 2024, doesn’t say which years of data the estimate comes from³.

People 65 and older make more than 600,000 of these visits a year, at more than twice the rate of younger people³. Blood thinners account for about 1 in 5 visits, diabetes drugs such as insulin for about 1 in 7, and antibiotics for about 1 in 8³.

A CDC-led study in the journal JAMA gives a dated snapshot. Using 2017 to 2019 records from 60 emergency departments, it estimated about 6 visits a year for every 1,000 people⁴. For people 65 and older, the five drugs most often involved were warfarin, insulin, clopidogrel, apixaban and rivaroxaban⁴. All but insulin are used to prevent blood clots.

In that study, about 69% of the visits came from people using a medicine for its intended purpose⁴. Most were not overdoses or misuse but the price of ordinary treatment.

Expectations can cause real symptoms, and scientists call it the nocebo effect

The placebo effect is feeling better after a treatment with no active ingredient. Its mirror image, the nocebo effect, is feeling worse because you expect a treatment to harm you.

A 2007 study in the Journal of Sexual Medicine showed how strong it can be. Researchers gave 120 men a prostate drug, finasteride, presented to them under a code name as a proven treatment for an enlarged prostate. Half were told it might cause sexual side effects, and half were not⁶.

Among the 107 men who finished, 43.6% of those who had been warned reported sexual side effects. Only 15.3% of those who had not been warned did⁶. Same drug, same dose; the only difference was the conversation.

The pattern shows up in large trials, too. A 2022 analysis in JAMA Network Open, led by Julia Haas and Ted Kaptchuk of Beth Israel Deaconess Medical Center and Harvard Medical School, pooled 12 COVID-19 vaccine trials with more than 22,500 placebo recipients and about as many vaccine recipients⁵.

After the first shot, more than 35% of the placebo group reported whole-body symptoms, most often headache and fatigue⁵. The team estimated that nocebo responses made up about 76% of the whole-body symptoms vaccine recipients reported after dose one and about 52% after dose two⁵. For reactions at the injection site, such as arm pain or swelling, the nocebo share was only about a quarter⁵.

Statins give the clearest picture. In 2022, the Cholesterol Treatment Trialists’ Collaboration published an analysis in The Lancet of 19 double-blind trials with about 124,000 people. Double-blind means neither the patients nor their doctors knew who got the real drug⁸.

Muscle pain or weakness was reported by 27.1% of people on a statin and 26.6% of people on placebo⁸. Most of the small excess showed up in the first year⁸. The University of Oxford researchers concluded that more than 9 in 10 muscle complaints among statin users were not caused by the drug⁷.

None of this means people imagine their pain. Aches, fatigue and headaches are part of daily life. More than a quarter of people swallowing dummy pills in those trials reported muscle symptoms⁸. When you watch closely for side effects, ordinary sensations can start to look like proof.

Long side-effect lists mix what a drug causes with what just happened

Unfold the paper insert from a prescription bottle, and the list of possible side effects can run for pages.

The Food and Drug Administration (FDA) separates two ideas in its 2006 guidance on drug labels. An adverse event is anything bad that happens to someone taking a drug, whether or not the drug caused it. An adverse reaction is an effect with some real reason to think the drug was responsible⁹.

The FDA asks companies to show common reactions from clinical trials in a table, with the drug and the placebo or comparison drug side by side. Reactions that happened just as often on placebo are generally left out⁹. Even so, the drug column counts everyone who had a symptom, including people who would have had it anyway.

A separate section lists problems reported after the drug went on sale. Those reports are voluntary, and no one knows how many people were taking the drug when they came in. So, the FDA says, how often such problems happen often can’t be reliably estimated⁹.

The agency also discourages long catch-all lists. Exhaustive lists of every reported event, its guidance says, “tend to obscure the more clinically meaningful information”⁹.

“Doubles your risk” can mean two extra people in 100

Relative risk compares two groups as a ratio, such as “twice as likely.” Absolute risk tells you how many people out of 100, or out of 1,000, are affected.

Picture a hypothetical drug. In its trials, 4 out of 100 people taking it got headaches, compared with 2 out of 100 on placebo. A headline could say the drug “doubles headache risk,” and in relative terms that would be true.

In absolute terms, the drug caused about 2 extra headaches per 100 people. And among the 4 people with headaches on the drug, 2 would likely have had them anyway. If you’re one of those 4, it’s close to a coin flip whether the pill is to blame.

Now the real statin numbers. Over a median of about four years, 27.1% versus 26.6% is a gap of half a percentage point⁸. By our own arithmetic, that’s about 5 extra people with muscle complaints per 1,000 treated.

The researchers counted reports per year instead and put the first-year excess at about 11 per 1,000 people⁸. Either way, it’s roughly 1 in 100 or fewer.

The same math applies to benefits. “Cuts your risk by a third” sounds large, but it could mean 3 fewer heart attacks per 100 people or 3 fewer per 10,000. The simplest question you can ask is: “Out of 100 people like me, how many are helped, and how many are harmed?”

Natural products have side effects too

Plants are chemistry, which is why some of them work and also why some can cause harm.

CDC researchers, writing in the New England Journal of Medicine in 2015, estimated that dietary supplements led to about 23,000 emergency department visits a year. The estimate came from a decade of hospital surveillance, 2004 through 2013¹⁰.

For adults in their 20s and early 30s, visits often involved heart symptoms tied to weight-loss or energy products. For people 65 and older, the problem was often mechanical: 37% of their visits came from trouble getting pills down, including choking¹⁰. About one-fifth of all visits involved children who got into supplements without an adult watching¹⁰.

That total is far smaller than the 1.5 million for drugs, but it isn’t a clean comparison. Far more people take prescription drugs, often for serious illness, and doctors may not think to ask about supplements. That’s our reading; neither source compares the two.

The National Center for Complementary and Integrative Health (NCCIH), part of the National Institutes of Health, says plainly that natural does not necessarily mean safe. It notes that comfrey and kava can cause serious liver harm¹¹. St. John’s wort, a popular herb for mood, interacts with antidepressants, birth control pills and HIV medicines¹¹.

NIH’s MedlinePlus Magazine explains how. Some supplements speed up the way your body breaks down a drug, so the drug works less well. Others slow it down, so the drug builds up and side effects rise¹². St. John’s wort can affect blood thinners and some cancer treatments this way¹².

This is where natural-health values help. Precaution, respect for the body and wanting to know what goes into it are core principles. Applied evenly, they ask the same questions of the herb that they ask of the pill: What is in it? How much? What does the evidence show?

Ask about the need, the dose and the exit plan

Fear of side effects is most useful when it becomes a question: Do you still need every medicine you take? Clinicians call the planned, supervised process of cutting back or stopping a medicine “deprescribing.”

In a trial in Quebec, Canada, researchers worked with 30 community pharmacies and 303 adults ages 65 to 95 who had taken benzodiazepines, a class of sleep and anxiety pills, for a long time¹³.

Half got an eight-page booklet about the drugs’ risks, other options and a step-by-step plan for tapering. They were urged to talk with their doctor or pharmacist¹³. After six months, 27% of the booklet group had stopped the drug, compared with 5% of those who got usual care. Another 11% of the booklet group had cut their dose by at least a quarter¹³.

The researchers concluded that teaching patients directly can prompt shared decisions to cut back on potentially harmful drugs¹³. The key word is shared: the booklet sent people to their prescriber, not around them.

Dose matters too. In the Lancet statin analysis, higher-intensity regimens produced a slightly larger rise in muscle complaints than moderate ones⁸. Asking whether you’re on the lowest dose that does the job is reasonable.

There is a twist. James Howard, who led the empty-bottle statin trial, called switching statins or lowering the dose a reasonable approach. But he cautioned that it can feed the expectation that some dose will make you feel bad². A planned test, done with your clinician, may tell you more than trial and error.

What the evidence shows

Claim	Evidence rating	What it rests on
Prescription drug side effects are common and can be serious	Strong evidence	National CDC emergency department data34
Expectation alone can produce real symptoms (nocebo)	Strong evidence	Placebo groups in large trials and the empty-bottle statin trial158
Most muscle aches in statin users are not caused by the statin	Strong evidence	19 double-blind trials with about 124,000 people78
Being warned about a specific side effect makes people more likely to report it	Promising but early	One randomized trial of 120 men6
Direct patient education helps older adults safely cut back on sleeping pills	Promising but early	One well-run cluster randomized trial13
Supplements can send people to the emergency department	Strong evidence	CDC surveillance data, 2004 to 201310
Natural products can’t cause side effects or drug interactions	Evidence says no, or unsafe	Documented liver harm and drug interactions1112

When to get help, and how to report a side effect

• Call 911 for trouble breathing, swelling of the lips, face or throat, chest pain, fainting or signs of a stroke after any medicine or supplement.
• Call poison control at 1-800-222-1222 if someone takes too much of anything, or if a child swallows a medicine or supplement.
• Call your prescriber or pharmacist about new or worrying symptoms. Don’t stop a prescription on your own, because some medicines need a slow taper¹³.
• Report drug side effects to the FDA through MedWatch, online, by phone or with the MedWatch 3500 form¹⁴.
• Report supplement problems through the federal Safety Reporting Portal at safetyreporting.hhs.gov¹⁴.

The after-market section of a drug label is built from voluntary reports like yours⁹.

Questions readers ask

The bottom line

Side effects are not a myth dreamed up by worried patients, and they are not always what they seem. Drug harms fill emergency departments, mostly among people taking their medicines as directed. At the same time, the expectation of harm can produce real aches, fatigue and headaches, even from a dummy pill.

The fairest path is one standard for everything you swallow. Ask what it does for you in real numbers, what it costs you compared with placebo, whether the dose fits, and how you would know it’s time to stop. That standard protects you from drugs you don’t need and from natural products that aren’t as gentle as they look.

Keep exploring: simple health explanations.

Related reading

• What happens when you don’t tell your doctor about supplements
• What to ask when your doctor recommends a treatment
• How medical decisions are supposed to get made with you
• What gets into supplements that isn’t on the label
• How to judge a cup of herbal tea, which is gentle but not automatically harmless

Sources

1. American College of Cardiology. “SAMSON: Symptom Burden of Statins Compared With Placebo.” ACC.org, 2020 (trial published in the New England Journal of Medicine, 2020; doi:10.1056/NEJMc2031173). acc.org
2. TCTMD. “Both Statins and Placebos Produce ‘Very Real’ Side Effects: SAMSON.” Cardiovascular Research Foundation, 2020. tctmd.com
3. Centers for Disease Control and Prevention. “FastStats: Medication Safety Data.” CDC, 2024. cdc.gov
4. American College of Cardiology. Journal scan of Budnitz DS, Shehab N, Lovegrove MC, et al. “US Emergency Department Visits Attributed to Medication Harms, 2017-2019.” JAMA, 2021;326(13):1299. acc.org
5. Harvard Medical School. “Power of Placebo: Some COVID-19 Vaccine Reactions May Result From Placebo Response” (on Haas JW, Kaptchuk TJ, et al., JAMA Network Open, 2022). HMS News, January 18, 2022. hms.harvard.edu
6. Mondaini N, Gontero P, Giubilei G, et al. “Finasteride 5 mg and Sexual Side Effects: How Many of These Are Related to a Nocebo Phenomenon?” Journal of Sexual Medicine, 2007;4(6):1708-1712. iris.unitn.it
7. University of Oxford. “New Study Shows Muscle Pain Not Due to Statins in Over 90% of Those Taking Treatment.” Oxford University news, 2022. ox.ac.uk
8. Cholesterol Treatment Trialists’ Collaboration. “Effect of Statin Therapy on Muscle Symptoms: An Individual Participant Data Meta-Analysis of Large-Scale, Randomised, Double-Blind Trials.” The Lancet, 2022;400(10355):832-845. spiral.imperial.ac.uk
9. U.S. Food and Drug Administration. “Adverse Reactions Section of Labeling for Human Prescription Drug and Biological Products: Content and Format.” Guidance for Industry, 2006. fda.gov
10. Emory University. “ER Visits Related to Dietary Supplements” (on Geller AI, et al., New England Journal of Medicine, 2015). Emory News Center, 2015. news.emory.edu
11. National Center for Complementary and Integrative Health. “Using Dietary Supplements Wisely: Tips.” NIH. nccih.nih.gov
12. MedlinePlus Magazine. “Did You Know Supplements and Medications Can Interact in Unexpected Ways?” National Library of Medicine, 2025. magazine.medlineplus.gov
13. Tannenbaum C, Martin P, Tamblyn R, Benedetti A, Ahmed S. “Reduction of Inappropriate Benzodiazepine Prescriptions Among Older Adults Through Direct Patient Education: The EMPOWER Cluster Randomized Trial.” JAMA Internal Medicine, 2014;174(6):890-898. pubmed.ncbi.nlm.nih.gov
14. U.S. Food and Drug Administration. “How to Report a Non-Emergency.” FDA. fda.gov
15. Centers for Disease Control and Prevention. “Tips for Medication Safety.” CDC. cdc.gov