34 Do not let medicines kept at home cause serious harm

Over 6 years, 22 hospitals in the United States registered 662 cases of acute liver failure in a row. Of these, 275 (42%) were caused by acetaminophen, and the median dose swallowed was 24 grams. Of these 275 people, 131 (48%) had taken too much by accident, not on purpose. Among the people who took too much by accident, 38% were taking more than a single medicine that contained this ingredient at the same time. Of these 275 people, 27% died before a liver transplant, and 8% received a transplant. In 2020 the National Medical Products Administration required over-the-counter leaflets to state that using too much causes severe liver damage, and that the suggested maximum in a day is not more than 2 grams. The leaflet must also say to avoid, as far as possible, combining the drug with medicines that contain acetaminophen or other fever and pain drugs. US rules require the opening line of the warning on this kind of over-the-counter drug to name three liver-harm situations: going over the maximum for a day, taking it together with another medicine that contains acetaminophen, and drinking 3 or more cups of alcohol a day during treatment. The same notice also requires the leaflet for an extended-release product to state that if an adult or a child takes too much, they should be seen at once, whether they feel unwell or not. The antidote is called N-acetylcysteine. The leaflet says to use it as early as possible: within 12 hours the effect is satisfactory, and after 24 hours the effect is poorer. A multicenter study in the United States from 1976 to 1985 followed 2540 overdose patients who were given this antidote. Among patients whose blood level marked them as likely to develop liver injury, liver injury appeared in 6.1% of those who started the antidote within 10 hours, and in 26.4% of those who started it only from 10 to 24 hours. Started within 8 hours after the overdose, it protected the patient no matter how high the blood level was.

Reye's syndrome causes brain disease and fatty change in the liver, and it usually appears after influenza or chickenpox. In US national surveillance from 1981 to 1997, 1207 cases were reported in people 18 or younger. Salicylate, the ingredient in aspirin-type drugs, could be measured in the blood of 82% of the sick children. The overall share who died was 31%. Warnings about salicylate drugs began in 1980. That year 555 cases were reported, the peak. From 1987 on, each year has stayed at no more than 36 cases. In 2011 the national drug regulator banned oral nimesulide for children 12 or younger. In 2020 it banned metamizole tablets for children and adolescents 18 or younger. Compound artemisia and metamizole tablets (复方青蒿安乃近片), Zhongganling tablets (重感灵片), and Zhongganling capsules (重感灵胶囊), which also contain metamizole, are likewise banned for ages 18 or younger. The measuring question uses US figures. In 2011 the US drug regulator stated that infant acetaminophen drops are 80 mg per 0.8 milliliters, and the drinkable liquid for older children is 160 mg per 5 milliliters. From those amounts, the same 1 milliliter of drops holds about 3 times as much drug as the drinkable liquid. The US drug regulator said that having both strengths at the same time is easy to mix up and can make a child take too much by accident, and that dosing devices from different products must never be mixed. A study in Greece measured 71 teaspoons and 49 tablespoons from 25 households. Filled with water, the volume ran from 2.5 milliliters to 7.3 milliliters. In a US emergency-department study, parents who used teaspoon or tablespoon units measured wrong 42.5% of the time. Parents who used only milliliters measured wrong 27.6% of the time. The adjusted OR was 2.3, 95% CI 1.2–4.4 (the chance of measuring wrong is about more than double).

These medications are classified as nonsteroidal anti-inflammatory drugs (NSAIDs); common examples include ibuprofen, naproxen, and diclofenac. A meta-analysis encompassing 280 randomized trials of these drugs against placebo involving 124,513 participants demonstrated that compared to placebo, ibuprofen raised the risk of upper gastrointestinal perforation, obstruction, or bleeding by an RR of 3.97 (95% CI 2.22–7.10; roughly about 4 times). Naproxen showed an RR of 4.22 (2.71–6.56), while diclofenac resulted in an RR of 1.89 (1.16–3.09). All NSAIDs also approximately double the risk of heart failure. U.S. regulations mandate that over-the-counter NSAIDs carry warnings about stomach bleeding; these warnings specifically identify six high-risk groups: individuals aged 60 or older, those with prior stomach ulcers or bleeding, people on anticoagulants or steroids, individuals also taking another NSAID, drinkers consuming three or more alcoholic beverages daily during treatment, and users exceeding recommended dosages or duration.

In its Announcement No. 57 of 2021, the National Medical Products Administration updated the instructions for 14 such medicines, including paracetamol, pseudoephedrine and dextromethorphan oral solution, pediatric paracetamol, artificial cow-bezoar and chlorpheniramine maleate granules, and pediatric paracetamol and amantadine hydrochloride granules. The new warning reads: “It is not recommended for parents or guardians to give these products to infants and young children 2 years old or younger on their own; they should be used under the guidance of a doctor or pharmacist.” An additional note was added: “These products must be used strictly according to the dosage instructions on the package to avoid overdosing.” Another note states: “Concurrent use of other cold medicines containing the same or similar active ingredients should be avoided.” A Cochrane systematic review included 29 randomized controlled trials, 10 of which were conducted on children. For all these medicine types, no better effect than placebo was observed in pediatric trials: cough medicines, antihistamines (the anti-allergy component), antihistamines combined with decongestants (the ingredient that relieves nasal congestion), and cough medicines combined with bronchodilators. Adverse reactions were reported in 21 studies, with formulations containing antihistamines and dextromethorphan showing a higher incidence of side effects.

In October 2020, the U.S. Food and Drug Administration (FDA) issued a safety warning. Use of nonsteroidal anti-inflammatory drugs after roughly 20 weeks of pregnancy may cause kidney problems in the fetus, leading to low amniotic fluid levels. This warning applies to both prescription and over-the-counter drugs, including aspirin, ibuprofen, naproxen, diclofenac and celecoxib. After reviewing adverse event reports up to July 2017, the FDA identified 35 cases of low amniotic fluid or neonatal kidney dysfunction, all with serious outcomes; 5 of these newborns died, each with accompanying kidney failure. In most documented cases, amniotic fluid levels returned to normal within 72 hours to 6 days after medication use stopped. The FDA advises pregnant women not to take these drugs after week 20 unless explicitly directed by a doctor. Low-dose aspirin (81 mg) prescribed for specific pregnancy-related conditions is exempt from this restriction. When pain relief or fever reduction is needed, the FDA mentions that other medicines such as acetaminophen can be used; consult a pharmacist or doctor first.

Announcement No. 68 of 2022 from the National Medical Products Administration switched omeprazole enteric-coated tablets from prescription to over-the-counter and issued a model leaflet. The listed use is short-term relief of heartburn and acid reflux caused by too much stomach acid. The precautions say: do not use it for more than 7 days, and if the symptoms do not ease, ask a doctor or a pharmacist. Do not take it again within 2 months. If the symptoms come back, get medical care at once. Do not use it when swallowing is difficult or painful, when you vomit blood, when there is blood in the stool, or when the stool is black. Those can be signs of something serious. When alarm symptoms appear, such as clear weight loss, repeated vomiting, trouble swallowing, vomiting blood, or black stool, a malignant tumor has to be ruled out first. The reason is that treatment can hide the symptoms and delay the diagnosis. A person aged 55 or older with a new symptom, or with a symptom that has changed, should ask a doctor. People taking clopidogrel should avoid taking omeprazole at the same time. The reason is in the US leaflet for clopidogrel (Plavix). Omeprazole inhibits the enzyme in the body that converts clopidogrel (CYP2C19). The leaflet says "omeprazole was shown to reduce significantly the antiplatelet activity of Plavix when given concomitantly or 12 hours apart". Taken together, or taken 12 hours apart, clopidogrel's effect against blood clots is clearly weaker either way. The uses written in that leaflet are acute coronary syndrome, a recent heart attack, a recent stroke, and peripheral arterial disease.

A Cochrane systematic review pooled 6 randomized trials and 1147 people. Compared with a placebo, antibiotics for a cold made no clear difference in the share who had not recovered or whose symptoms persisted (RR 0.83, 95% CI 0.60–1.14, and the difference is not significant). The risk of adverse reactions was 1.8 times the placebo group (95% CI 1.01–3.21). For adults it was 2.62 times (1.32–5.18). For children it did not rise (0.91, 0.51–1.63). For acute purulent rhinitis with a pus-filled runny nose lasting under 10 days, antibiotics also had no clear benefit (RR 0.73, 0.47–1.13). Adverse reactions were more common, though (RR 1.46, 1.10–1.94, about 46% higher). The authors' conclusion is that routine antibiotics are not recommended in either situation.

The diarrhea-treatment manual of the World Health Organization (WHO) states that "anti-diarrhea" drugs and anti-vomiting drugs have no real benefit for acute or prolonged diarrhea in children. They cannot prevent dehydration, and they cannot improve nutrition. The Chinese edition of the manual says, in its own words, that some drugs have dangerous and sometimes fatal side effects, and that these drugs must never be used in children 5 or younger. Drugs such as loperamide, which slow the movement of the bowel, can cause severe paralytic obstruction of the bowel. That can kill, and it can also drag the infection out. The manual also states that drinks with too much sugar, such as soft drinks and fruit drinks sold in shops, pull water from the body into the intestine and cause hypernatremic dehydration (dehydration in which the blood sodium is too high). What the WHO manual recommends is a reduced-osmolarity formula, with a total osmolarity of 245 mOsm/L (milliosmoles per liter). Compared with the older formula, it cut unplanned intravenous fluids by 33%. A Cochrane review pooled 8 trials and found a similar result (OR 0.59, 95% CI 0.45–0.79, about 40% less).

The 3rd edition of the International Classification of Headache Disorders, from the International Headache Society, gives the diagnostic criteria. The person already has a headache disorder, has headache on 15 or more days a month, and has regularly overused a medicine for headache for more than 3 months. For non-opioid painkillers such as acetaminophen, aspirin, and ibuprofen, the overuse line is 15 days a month or more. For combination painkillers the line is 10 days a month or more. A combination painkiller means more than a single class of pain-relieving drug combined, or a painkiller with an added ingredient such as caffeine, for example acetaminophen plus codeine. If you rotate several non-opioid painkillers, the days are added together. The classification states that epidemiological evidence from several countries shows that, among people with headache on 15 or more days a month, more than half have this kind of headache. Most patients improve after they stop the overused medicine, and treatment to prevent headaches also works better.

US emergency-department surveillance estimates that from 2004 to 2013, children under 6 who swallowed a medicine without an adult watching made about 640000 emergency visits (the estimate is 640161). In 2013 alone there were about 59000 (59092). From 2010 to 2013, the most common form was an oral prescription tablet, 45.9%. Among prescription drugs, the most common were opioid painkillers (13.8%) and benzodiazepines, sedative drugs used for sleep and calming (12.7%). In 2017, 5 US poison-control centers asked parents who called for help. When a child swallowed a prescription drug, 30.7% of the time it was a grandparent's medicine. More than half (51.5%) of the prescription drugs had already been taken out of the original packaging by an adult before the child got them. China's National Catalogue of Hazardous Wastes (2025 edition) lists waste medicines produced in ordinary household life, and their packaging, as hazardous waste within domestic garbage. Shanghai's sorting guide puts waste medicines in hazardous waste. The guide states that waste medicines include expired drugs, drug capsules, tablets, and the inner packaging, and that "waste medicines should preferably be put out together with the packaging." Shanghai also requires that expired household medicines collected by pharmacies go to a temporary transfer point set up by the market supervision administration of each district. The US FDA says the best way to deal with expired and unwanted medicines is to hand them to a take-back point.

This is called a disulfiram-like reaction. A 2023 medication-safety article from the Fujian drug regulator says that drinking before or after some cephalosporins can set off this reaction, and that it "can seriously harm health and can even threaten life." Symptoms include a flushed face, sweating, and palpitations, and even difficulty breathing, a drop in blood pressure, and anaphylactic shock. Following the Clinical Medication Notices of the Pharmacopoeia of the People's Republic of China, the drugs it lists include cefazolin, ceftriaxone, cefoperazone, metronidazole, tinidazole, and furazolidone, among others. It states that drinking alcohol, and taking or eating drugs or foods that contain alcohol, is forbidden for 7 days after the dose. The examples include liqueur chocolates and Huoxiang Zhengqi oral liquid. The reason given is that the enzyme the drug inhibits often takes 4 to 5 days to recover. A hospital in Guangzhou reviewed 82 cases over 2 years. Cephalosporins caused 57 cases (69.51%). Nitroimidazoles, the class that includes metronidazole and tinidazole, caused 10 cases (12.20%). The alcohol came from alcoholic drinks in 40 cases (48.78%) and from ethanol-containing medicines in 31 cases (37.80%). Medicines taken by mouth included Huoxiang Zhengqi oral liquid, cough syrup, and motherwort fluid extract (益母草流浸膏). Of the 82 cases, 68 were mild, 9 were moderate, and 5 were severe. They eased within 1 to 3 hours. Nobody died. The US leaflet for metronidazole requires no alcohol during treatment and for at least 3 days after stopping, and no products that contain propylene glycol.

A 2010 adverse-reaction bulletin from the National Medical Products Administration says that vitamin C Yinqiao tablets (维C银翘片) contain chlorphenamine maleate, also called chlorpheniramine (扑尔敏). The adverse reactions are mainly drowsiness, a feeling of weakness, sleepiness, and a dry mouth. The bulletin advises doctors and pharmacies to tell the patient: while taking the drug, do not operate an aircraft, a vehicle, or a boat, and do not do work at a height, mechanical work, or operate a precision instrument. A 2020 public medication notice from the National Medical Products Administration says that combination cold medicines often contain an anti-allergy ingredient such as chlorphenamine or diphenhydramine, and that taking them causes drowsiness. People who drive, operate machinery, or work at a height should be cautious. A US driving-simulator trial enrolled 40 drivers with seasonal allergic rhinitis, aged 25 to 44. Each person switched treatment a week apart and drove for 1 hour under each: diphenhydramine 50 mg, fexofenadine 60 mg, alcohol up to a blood alcohol concentration of about 0.1%, and a placebo. Fexofenadine is an anti-allergy drug that does not cause drowsiness. The ability to match the speed of the car ahead was worst after diphenhydramine, and clearly better both after alcohol and after fexofenadine. The ability to hold the lane got worse both after alcohol and after diphenhydramine. When the car ahead suddenly blocked the way, the reaction was slowest after alcohol, a mean of 2.21 seconds, and after fexofenadine it was 1.95 seconds. Driving on fexofenadine was about the same as driving on placebo. The authors' conclusion is that diphenhydramine affects driving more than alcohol does. The drowsiness a driver reported did not show how well they drove, so "whether I feel drowsy" cannot decide whether they can drive.

A 2013 review by Canadian pharmacology researchers says that more than 85 drugs are known or expected to interact with grapefruit. Of these, 43 may cause serious adverse reactions. The serious reactions include torsades de pointes, a heart rhythm disorder that can cause sudden death. They also include rhabdomyolysis (large-scale muscle breakdown, which can damage the kidneys), bone marrow suppression, respiratory depression, gastrointestinal bleeding and kidney injury. The furanocoumarins in grapefruit permanently disable an enzyme in the small intestine that breaks drugs down (CYP3A4). The drug is broken down less in the gut, so more of it reaches the blood. One whole grapefruit or 200 ml of juice is enough to raise the blood level to a clinically meaningful degree. With the blood-pressure drug felodipine taken with a serving of grapefruit, the drug in the blood was on average 3 times as high as with water. With the same 250 ml serving of juice, the effect in different people ranged from 0 times to 8 times. With the cholesterol-lowering drug simvastatin taken with 200 ml of juice a day for 3 days, the total amount of drug in the blood (AUC) was 330% of that with water. With atorvastatin and juice it was 180% to 250%. Pravastatin, rosuvastatin and fluvastatin are not affected. Drinking 200 ml of juice within 4 hours before the drug has the greatest effect. After 10 hours half of the effect remains, and after 24 hours a quarter. The review says sour orange (often used to make marmalade), lime and pomelo have the same effect. Sweet oranges such as navel and Valencia oranges do not contain furanocoumarins and do not have this effect. A US FDA public-education page names simvastatin, atorvastatin and nifedipine. It says that when you drink a lot of grapefruit juice, too much of a cholesterol-lowering drug stays in the body, which increases the risk of liver and muscle damage and can lead to kidney failure.