English Opinion

Downstream

Dear reader, in the 1990s, the vultures of South Asia began to disappear. Not gradually: three species of Gyps vulture, among the most abundant large raptors on earth, lost more than ninety-five percent of their population within a decade; for the white-rumped vulture, the fall was above ninety-nine percent. Nobody could explain it. The birds were dying of visceral gout, their organs coated in uric acid, and the cause was not a virus.

It was a painkiller. Indian farmers had begun treating sick cattle with diclofenac, an ordinary anti-inflammatory available in any pharmacy. When a treated animal died, the vultures ate the carcass, and diclofenac destroys vulture kidneys at doses that are trivial for a cow. A veterinary drug, correctly prescribed for the animal in front of the vet, had emptied the sky. India banned the veterinary formulation in 2006, though 30 ml multi-dose vials of the human formulation remained in cattle sheds until those too were restricted to single doses in 2015. The consequences are still being counted. Feral dog populations rose, and rabies with them, where the carcasses went uneaten; an analysis published in 2024 attributed roughly a hundred thousand additional human deaths a year in India between 2000 and 2005 to the collapse, a figure other researchers have since disputed.

I begin with the vultures because they are the clearest demonstration of a principle we rarely think about: a medicine does not stop working when it leaves the patient. It is excreted, often unchanged or as an active metabolite, into the sewer, where bacterial enzymes strip off the conjugates the liver attached and hand the active parent compound back, and from there into the environment, where it goes on being pharmacologically active in whatever encounters it.

How widespread is this? In 2022, a study led by the University of York attempted the first global answer built on a single consistent method, sampling 1,052 sites along 258 rivers in 104 countries for 61 active pharmaceutical ingredients. They found contamination almost everywhere they looked, and at more than a quarter of sites, concentrations exceeded levels considered safe for aquatic organisms. The worst-affected regions were in sub-Saharan Africa, South Asia, and South America, and the pattern was not about how much medicine a population takes. It tracked wastewater infrastructure and the presence of drug manufacturing.

The other classic case is hormonal. The synthetic estrogen in oral contraceptives, ethinylestradiol, survives sewage treatment at low concentrations, and in a whole lake experiment at the Experimental Lakes Area in Ontario, researchers added it over three summers at five to six nanograms per liter and watched the fathead minnow population collapse: males were feminized, and reproduction failed. Five parts per trillion did that.

Then there are antibiotics, which are the reason this subject should worry us most. A river carrying a low concentration of antibiotics is not merely polluted. It is a selection chamber that applies steady pressure favoring resistant organisms, in an environment where bacteria trade genes freely and those genes eventually find their way back to us. The highest concentrations recorded anywhere have been downstream of manufacturing sites, in the very countries that supply much of the world's generic medicine.

Why do treatment plants not solve this? Because they were never designed to. Treatment is built to remove organic load, solids, and pathogens. It is not built to strip small, polar, deliberately stable molecules present at nanograms per liter. Removal does happen, but incidentally and very unevenly: paracetamol and ibuprofen are largely degraded, while carbamazepine and diclofenac pass through more or less untouched. And what a drug company optimizes for, resistance to metabolic breakdown, is precisely what makes a compound survive the sewage works.

Our position in the Gulf is unusual and largely unstudied: high and rising pharmaceutical consumption, a young population moving early onto chronic therapy for diabetes and cardiovascular disease, and almost no perennial rivers. That last fact sounds protective, but it may be the opposite: with little natural dilution, our residues concentrate in treated wastewater, which we use extensively for landscape irrigation and increasingly for agriculture. Add decades of generous antibiotic dispensing, restricted on paper since Saudi Arabia began enforcing its prescription-only rule in 2018, but still working through the stock in household cupboards, and the habit of pouring leftovers down the drain, and you have a system recirculating its own pharmacology with little monitoring of what is in it.

The remedies are not exotic. Take back schemes at pharmacies so unused medicines do not go down the sink. Effluent standards for manufacturing sites, which the industry resists and which cost less than the resistance they would prevent. Advanced treatment, such as ozonation or activated carbon, at plants whose output is reused. Prescribing volume itself, which is the cheapest intervention of all. And above everything, measurement: a national survey of what our treated wastewater contains, which, as far as I know, has never been done at scale.

Dear reader, we have built a civilization that treats the pill as ending at the patient. It does not. It ends in the water, in the soil, in the fish, and occasionally in a vulture. The vultures were the warning, and they were expensive to ignore.