Millions of Americans take herbal supplements with a liver risk, like turmeric

A turmeric capsule taken for stiff joints looks like the safest item in the medicine cabinet. A survey study in JAMA Network Open estimated that about 15.6 million U.S. adults had taken at least one of six botanicals tied to liver injury in the previous 30 days, and turmeric was the most common of them. The study counted exposure, and it did not count damaged livers.

Read next on Morning Overview: Winter weather advisories cover I-80, I-70 and the Bighorns with up to 15 inches of snow

Supplement use itself is routine: 57.6% of the adults in the survey had used at least one herbal or dietary supplement in the past 30 days.

Six botanicals and how many adults took each

The work came from Alisa Likhitsup, Vincent Chen and Robert Fontana of the University of Michigan’s gastroenterology and hepatology division, who analyzed nationally representative NHANES survey data collected from January 2017 to March 2020. Of 9,685 adults, 4.7% reported using at least one of turmeric or curcumin, green tea, Garcinia cambogia, black cohosh, red yeast rice or ashwagandha. Scaled to the population, that is the 15.6 million figure, a group about as large as the roughly 14.8 million estimated to use NSAID pain relievers or the roughly 14.0 million on the cholesterol drug simvastatin.

Turmeric dominates the list. By the paper’s weighted estimates, about 11.4 million adults used turmeric or curcumin, against 3.3 million for green tea, 1.25 million for ashwagandha, 1.2 million for black cohosh, 890,000 for Garcinia cambogia and 626,000 for red yeast rice. Turmeric users most often named joint health or arthritis as the reason, at 26.8%. Garcinia users were mostly trying to lose weight, black cohosh users wanted relief from hot flashes, and red yeast rice users were mostly after heart health.

The survey respondents who took any of the six were older and more educated than non-users, and they were more likely to have arthritis, with an adjusted odds ratio of 2.27. Population estimates were scaled with 2020 Census figures. Exposure, in other words, skews toward older adults managing joint pain.

What LiverTox says about turmeric injury

The federal LiverTox database, maintained by the National Institutes of Health, gives turmeric its top likelihood rating, an A, meaning it is a well-documented cause of clinically apparent liver injury. The entry says such injury is probably very rare, on the order of 1 in 10,000 to 1 in 100,000 exposed people, and that it looks idiosyncratic and possibly immune-mediated. A gene variant, HLA-B*35:01, turned up in more than 70% of reported cases against 10% to 15% of controls. Most cases involve highly bioavailable curcumin products, though plain ground turmeric and turmeric tea have also been implicated.

The clinical picture starts quietly. Fatigue, nausea and poor appetite come first, then dark urine and jaundice, with onset typically one to four months after starting the product. Most patients recover within one to three months of stopping it, but some cases progress to acute liver failure, transplantation or death, and LiverTox notes that continuing the product after symptoms appear raises the odds of a severe outcome. The National Center for Complementary and Integrative Health, in its turmeric fact sheet updated in April 2025, says bioavailable curcumin formulations may harm the liver and advises stopping them and calling a health care provider if those symptoms appear.

The case record behind that rating is modest in size but sobering. LiverTox walks through three detailed cases, two of which ended in full recovery and one in death after a 14-month latency, and it cites a series of 10 U.S. cases from the Drug-Induced Liver Injury Network, an 11-case pathology series and an Italian analysis of 37 cases. Turmeric appears to be the most common cause of clinically apparent herbal-related liver injury in the United States.

Supplement injury counts and a lightly regulated market

The survey sits against a worrying background. According to News-Medical’s summary of the paper, the share of drug-induced liver injury cases linked to supplements climbed from 7% in 2005 to 20% in 2014, and in 2014 supplements were associated with an estimated 23,000 emergency department visits and 2,154 hospitalizations. More than 80,000 supplement products are on the market. The FDA states that it does not have the authority to approve dietary supplements for safety and effectiveness, and that companies are responsible for making sure their products meet safety standards.

What a survey of 9,685 adults cannot show

Nothing in the paper establishes that any individual’s liver was harmed. The data are self-reported 30-day use, daily doses were not recorded, product labels were not checked by chemical analysis, and the counts for several botanicals are small: 19 to 28 users each for black cohosh, red yeast rice, Garcinia cambogia and ashwagandha. Data collection in the 2019-2020 NHANES cycle was also cut short by COVID-19.

The authors drew a narrow conclusion from that. Clinicians, they wrote, should be aware of possible adverse events from consumption of these largely unregulated products. The open question is the one the survey cannot answer: against LiverTox’s estimate of one injury per 10,000 to 100,000 people exposed to turmeric, how many of the 15.6 million actually landed in a hepatologist’s office remains unmeasured.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.