Kai Tak river in Hong Kong drugs pollution antibiotics
© Robert Harding/Rex/ShutterstockThe Kai Tak river in Hong Kong had 34 different active pharmaceutical ingredients at a single site, the highest number recorded
Humanity's drugs have polluted rivers across the entire world and pose "a global threat to environmental and human health", according to the most comprehensive study to date.

Pharmaceuticals and other biologically active compounds used by humans are known to harm wildlife and antibiotics in the environment drive up the risk of resistance to the drugs, one of the greatest threats to humanity.

The scientists measured the concentration of 61 active pharmaceutical ingredients (APIs) at more than 1,000 sites along 258 rivers and in 104 countries, covering all continents. Only two places were unpolluted - Iceland and a Venezuelan village where the indigenous people do not use modern medicines.

The most frequently detected APIs were an anti-epileptic drug, carbamazepine, which is hard to break down, the diabetes drug metformin, and caffeine. All three were found in at least half of the sites. Antibiotics were found at dangerous levels in one in five sites and many sites also had at least one API at levels considered harmful for wildlife, with effects such as feminising fish.

The APIs end up in rivers after being taken by people and livestock and then excreted into the sewer system or directly into the environment, though some may also leak from pharmaceutical factories.

Hotspots with very high levels of APIs included Lahore in Pakistan, La Paz in Bolivia, and Addis Ababa in Ethiopia. Madrid in Spain was in the top 10% of places with highest cumulative concentrations, and Glasgow, UK, and Dallas, US, were in the top 20%.

"The World Health Organization and UN and other organisations say antimicrobial resistance is the single greatest threat to humanity - it's a next pandemic," said John Wilkinson, at the University of York, in the UK, and who led the study, which involved 127 researchers from 86 institutions. "In 19% of all of the sites we monitored, the concentrations of [antibiotics] exceeded the levels that we'd expect to encourage bacteria to develop resistance."

Research published in January estimated that 5 million people died in 2019 from bacterial infections that were resistant to antibiotics. The regions suffering the highest impact from antibiotic resistance in that study closely align with those in the study with the worst drug pollution, suggesting the contamination of rivers may be playing a part in driving up resistance. One site in Bangladesh had levels of the antibiotic metronidazole more than 300 times higher than the safe target, possibly due to leaks from pharmaceutical manufacturing.

Drug pollution was already known to be harming wildlife, from antidepressants causing starlings to feed less and contraceptive drugs reducing fish populations. "If I were a fish living in some of these rivers, I'd be worried right now," said Wilkinson. However, the levels in most rivers would not deliver high doses to people swimming, he said.

The study, published in the journal Proceedings of the National Academy of Sciences, is by far the biggest to date and represents the impact on river pollution of 470 million people. The researchers concluded: "Pharmaceutical pollution poses a global threat to environmental and human health."

Previously, almost all the measurements had been taken in western Europe and North America but the latest research showed API pollution is often much higher elsewhere. The work included 36 countries in which APIs were measured for the first time, particularly in Africa and South America.

Among the drugs detected on all continents except Antarctica were the antidepressants citalopram and venlafaxine, antihistamines cetirizine and fexofenadine, the antibiotic trimethoprim and lidocaine, an anaesthetic. The Kai Tak River in Hong Kong had 34 different APIs at a single site, the highest number recorded.

"Ecological risks could well be greater than predicted for the single APIs due to toxicological interactions of these mixtures," the researchers said. There are more than 2,500 pharmaceuticals in use, but current technology allows the analysis of only 50-100 from a single sample, so researchers focused on the most commonly used.

The highest drug concentrations were found in low-to-middle income countries, including India and Nigeria. The researchers think this may be because people in these nations have enough income to buy pharmaceuticals, but live in places without good sewerage systems, which can remove drugs but are expensive.

The study did not include measurements of illegal drugs such as cocaine and MDMA, which have been detected in rivers at levels harmful to wildlife, although future analysis of the samples may do this.

The scientists hope the research will help focus clean-up efforts on the pharmaceuticals and regions at greatest risk. "We know good sewage connectivity and wastewater treatment is the key to minimising, though not necessarily eliminating, pharmaceutical concentrations," said Wilkinson. "However, that is extremely expensive as there's a lot of infrastructure involved."

Using medicines more carefully is another way to reduce the pollution, he said, particularly antibiotics, which are cheaply available in many countries without prescriptions, and widely taken unnecessarily, for example to treat colds.

"Pharmaceuticals are almost omnipresent in rivers across the world," said Prof Joakim Larsson, of the University of Gothenburg, Sweden, who was not part of the study team.

"The study shows that a fairly large set of pharmaceuticals exceed 'safe levels', and often at a very large number of sites. Bacteria do not respect national borders, so if a new resistant bacterium develops on one side of our planet, it soon becomes a risk for everyone."

The researchers are looking to extend the number of countries covered, as the Covid-19 pandemic halted their surveys. They are also increasing the number of drugs measured and hope to assess levels in rivers across the year in order to examine seasonal trends.