When a research report published five years ago made headlines in July this year, public health experts were left scratching their heads.
The “new” research wasn’t so new. It documented levels of HIV medicines in South Africa’s water systems between 2017 and 2018. And, despite some media reports saying otherwise, the amounts in many of the samples were so small that they were hardly traceable and not an immediate health risk.
So why this dated report was turning up on so many news sites as breaking news was a mystery. More concerning was the widespread and unnecessary public panic it created.
It’s frightening to hear that anything but water is in our tap water.
But pharmaceuticals in water are a well-documented global concern that scientists have been talking about for years. And South African researchers working on issues around public health are well aware of the numerous studies that have picked up everything from painkillers to antibiotics — as well as HIV medicines, or antiretrovirals (ARVs) — in South Africa’s water systems. Because we have one of the largest HIV treatment programmes in the world, that’s what they’d expect.
We wanted to find out how this research got misinterpreted, so we teamed up with the fact-checking organisation Africa Check. They looked at the media reporting failures while we unpack what the report actually found, and what it didn’t, to try to allay some of the public health fears.
What was the research about?
Given the high number — just more than 6-million — of people on ARVs in the country, the main aim of the North-West University (NWU) report was to follow the journey of HIV medicines once they leave the body and consider what this might mean for our health and the environment.
The researchers, who were commissioned by the Water Research Commission (WRC), tracked the route of ARVs through the water system, starting from sewage (where everything we flush ends up), wastewater from hospitals that hasn’t been fully treated and illegally dumped industrial waste. That water gets mixed in with river water, which undergoes treatment at wastewater treatment plants, and is released back into rivers and dams that supply our tap water.
The team tested water at 72 sites across Gauteng and North West between 2017 and 2018, targeting HIV medicines which were widely prescribed at the time and frequently reported in other similar studies, namely efavirenz, nevirapine, lopinavir, ritonavir, lamivudine, stavudine, zidovudine and fluconazole. Though not an ARV, the researchers included fluconazole because it is commonly prescribed to HIV patients to prevent and treat fungal infections.
The sampling sites included points upstream — before the water reaches the plant — and downstream, after the treated water leaves it. That allowed researchers to see how well those facilities were able to clean out the drug residues. The plant is like a washing machine: dirty clothes go in and come out cleaner, but sometimes a few stains remain. A few samples were also taken from dams and drinking water. They tested well-known urban rivers like the Jukskei in Joburg, as well as more rural sites like the Mooi River in North West.
What did the study find?
There were detectable amounts of the ARVs efavirenz, nevirapine and lopinavir in rivers and dams, plus tiny traces of fluconazole.
When it came to drinking water samples, the amounts were extremely small — so small that they were right at the edge of what the instruments can reliably measure.
One exception was a drinking water site just downstream of the Sunderland Ridge wastewater treatment works in Centurion, Gauteng. Here, efavirenz was recorded at unusually high levels, showing that even after water had been through the treatment plant, residues were still released into the Crocodile River system. The efavirenz concentration was 6.2 micrograms per litre of water, roughly the equivalent of three teaspoons of medicine in an Olympic-sized swimming pool. The researchers considered this site a flag that points to a need for further study.
Before treatment, upstream, the ARV concentration was stronger, closer to a drop in five pools. This means treatment at this plant lowered the amount of ARVs in the water, but — as expected — some traces remained.
It’s science doing exactly what it’s meant to do, adding insight and helping us solve a problem
So what’s the trouble with the research?
The problem isn’t with the research itself. The study brought the troubling, and growing, issue into the national conversation and added useful information to what scientists already know. It’s science doing exactly what it’s meant to do, adding insight and helping us solve a problem.
The problem started with how the research report was presented on the NWU website, which was posted in April this year. It didn’t mention when the samples were taken, nor when the report was published, and included complex scientific terms that may have added to the confusion. It was picked up three months later by journalists who made the same mistakes and added to them, which our partners at Africa Check have explained.
Part of the confusion among journalists seems to have been that they didn’t know, or didn’t mention, that there isn’t just one document. There is the comprehensive WRC report from 2020, which we’ve been referencing here, with all of its extensive data. There’s also a peer-reviewed spin-off paper, published in 2022, that zoomed in on a portion of the data to look at how a selection of wastewater treatment plants in Gauteng handle ARVs. It found a mixed picture: some plants reduced them, some didn’t. Another analysis is in progress.
The distinction of a peer-reviewed study is important. It means that academic colleagues in the same field of scientific study have had a detailed look at the study and agreed with its findings. This does not negate the main report findings, but peer review is the gold standard for scientific credibility.
So why did public health experts immediately question the headlines?
Not one of the media reports mentioned that the research report was released in 2020 — several articles said it was a “new” study, others simply covered it as if it was — nor did they say that the samples were taken a couple of years before that.
HIV treatment requires a combination of at least three different ARVs, which have evolved as the medicines get more effective and have fewer side effects. So, from a scientific perspective, the timeline matters, as University of KwaZulu-Natal pharmacologist Andy Gray explains.
If you were to test positive with HIV in South Africa today, you would be treated with a three-in-one entry-level pill called TLD, which includes the drugs tenofovir, lamivudine and dolutegravir. But TLD has only been available since 2019.
“One of the most widely used ARVs, tenofovir, wasn’t tested for at all. Efavirenz, which shows up at about a third of the sites [the researchers sampled from], has largely been replaced with a newer drug, dolutegravir,” says Gray.
In addition to efavirenz being replaced in the most commonly used entry-level pill, some of the other drugs the researchers tested — including nevirapine and lopinavir — are now used much less.
As Gray points out, there are even more people on ARVs now than in 2018, so one can assume there are higher concentrations of ARVs in our water right now. But until a study using more current samples is done, it’s a bit like checking the weather forecast in 2018 to see how storms behave: it won’t tell you if you need an umbrella today, but it does show you the patterns to expect.
What about the fish?
In the river systems themselves, the researchers didn’t see any real-world harm they could measure. So they did an experiment in the laboratory. They added higher amounts of ARVs to water to see how this affected water life, human cells and viruses. In those experiments, the HIV medicines did cause small changes in the different organisms tested, including snails, viruses and human tissue. But the researchers make it clear that it’s too soon to claim these changes pose a health risk to people or wildlife. Some of the media stories, however, made it sound as if those impacts on organisms had already been seen in rivers, rather than in lab experiments — and this led to further alarm.
If we created better awareness among medicine users, we could address the problem at the source
— Neelaveni Padayachee
Should I be worried about the remnants of ARVs?
Responding to the media reports, the department of water & sanitation said the small amounts of ARVs in the water aren’t harmful. But the researchers stopped short of saying the water was either safe or unsafe, noting that their study was a signal that the issue needs to be studied more closely over the long term. The department also flagged HIV drug resistance as a common concern; the researchers had acknowledged the possibility, but said this, too, was an area for further research.
Drug resistance happens when a virus or bacteria changes in a way that makes the medicine used to treat it less effective. With HIV, this can happen if someone doesn’t take their ARVs consistently, missing doses or stopping and starting treatment. This makes the infection harder to control and can lead to illness. If left unchecked, HIV can progress to Aids, weakening the immune system so much that it can no longer fight off infections.
Pharmacovigilance expert Neelaveni Padayachee of Wits University explains that drinking water with tiny amounts of HIV medicine in it won’t have an impact right away. But it could be that, if people keep drinking that water over many months or years, those small amounts could build up in the body. Over time, this might cause problems, such as making the virus harder to treat with the normal dose of ARVs.
The most significant impact would be for HIV-positive people who aren’t yet on treatment, who might absorb just enough ARVs from water to trigger drug-resistant strains, she says.
What else is in our water?
A 2024 study that looked at more than 20 pieces of research testing water across South Africa identified over 100 different medicines in rivers and wastewater. The mix includes painkillers, anti-inflammatories, antibiotics, antidepressants, blood pressure medicines, steroid hormones, diabetes medicines, ARVs and illicit drugs.
Researchers call some of these, including ARVs, contaminants of emerging concern. They’re not all new, but more people are using them now, better water testing is picking them up, and their effects aren’t fully understood, says Renée Street, director at the environment & health research unit of the South African Medical Research Council (SAMRC).
So our treatment plants aren’t cleaning the water properly?
Peer-reviewed studies indicate that conventional wastewater plants cannot adequately remove ARVs. This is also what the NWU team found after analysing the concentrations of ARVs in river water, just before and just after treatment. Curiously, in some cases, they found higher ARV concentrations after wastewater treatment. A separate study from 2024 shows that our wastewater plants aren’t equipped to deal with a variety of other pharmaceuticals and personal care products that have become a concern.
Our water treatment failures aren’t limited to pharmaceuticals. The government’s 2023 Blue Drop report on drinking water found that 277 of the country’s 958 water supply systems were in a “critical state”, due to chemical pollutants, but also contamination with harmful bacteria in the water and ageing infrastructure. Wastewater systems were worse off: nearly two-thirds of the sewage treatment plants were rated for high or critical risk of allowing partially treated or untreated water into rivers and the environment.
Researchers like Philiswa Nomngongo, an expert in nanotechnology for water at the University of Johannesburg, are worried.
“The usual cleaning processes don’t fully remove ARVs, which means tiny traces of these HIV medicines can still end up in rivers and other parts of the environment,” says Nomngongo “Some newer technologies, which use super-tiny materials to break down or filter out ARVs more completely offer hope, but they are still expensive and not affordable at the scale at which we need them.”
What can we do about the pharmaceuticals in the water right now?
Padayachee points to a 2022 study that showed almost 32% of people in South Africa flush their unwanted or expired medicines down the toilet. “If we created better awareness among medicine users, we could address the problem at the source,” she says.
Nomngongo says “we urgently need a proper, well-funded programme to test water across the country, both in cities and rural areas”, to show where the biggest problems are, help guide government decisions and improve treatment methods.
The SAMRC recently received funding to do just that, says Street. Its wastewater surveillance collaboration with Stanford University in the US will allow for a wider variety of monitoring sites, including in rural areas where water sources have never been monitored.
Excellent public health research is being done in South Africa all the time. Scientists should be talking about it, and journalists should be talking to them and reporting with care. Read more media tips from our partners at Africa Check.
This story was produced by the Bhekisisa Centre for Health Journalism. Sign up for the newsletter.
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.