Data rules: how numbers turned our HIV plans

Numbers put us in the know. We saw that with Covid-19, and now again, with Rwanda snuffing out the Marburg virus outbreak in about a month.

Why? Because keeping accurate track of epidemiological data — how many people are infected, get treated, recover or die, and where and when all of this happens when a disease breaks out — allows health authorities not only to see how an epidemic unfolds but also to make evidence-based decisions to change its course. 

Rewind to South Africa close to 25 years before, where in 2001, little data was available on HIV prevalence in the country. 

Instead of concrete counts and scientific facts driving authorities’ decisions on how to tackle HIV, the then president and health minister denied that the virus causes Aids — and with it denied scores of people treatment, which experts estimate could have saved 330,000 lives between 2001 and 2005. 

But the Human Sciences Research Council’s (HSRC) first national survey in 2002, in which blood samples were collected from 10,000 people in households across South Africa, was “a turning point”, says Salim Abdool Karim, director of the Centre for the Aids Programme of Research in South Africa (Caprisa), in a video about the survey’s history. “At a time when [the] country had a high level of denial — or more appropriately, denial at high levels — this survey began to make HIV real.” 

Just more than 11% of participants’ blood samples tested positive for the virus, a finding that helped people understand how big the problem truly was — and which Aids denialists could not argue with. 

“We went to the office of the president with our findings,” says Olive Shisana — who was the director-general in the national health department from 1994 to 1998 and became the CEO of the HSRC in 2005 — in the video history (she’s now the special adviser for social policy in President Cyril Ramaphosa’s office). “We told them that HIV is the reality in South Africa and that we need to do something about it — and we did.”

Today, South Africa has the biggest antiretroviral (ARV) programme in the world, with close to 6-million people on treatment and a finely worked out data set that tracks things like HIV testing, new infections, ARV use, viral suppression and Aids deaths among different age groups, genders and risk groups (for example, sex workers and men who have sex with men). 

But numbers alone are just information. 

To turn data into intelligence — something that can be used to drive decisions — we have to understand what the statistics tell us about the past, and what they predict for the future. 

Because, says Abdool Karim, “you cannot advise […] public health leaders or politicians about what to do, if you don’t have good evidence”. 

Here’s what South Africa’s HIV policymaking story looks like.

Big drop, big gains

In 2000, there were roughly 1,460 new HIV infections in South Africa every day. By 2023, that number had fallen to just more than 400 a day — about a quarter of what it was at the turn of the century, at the height of Aids denialism. 

Models show that convincing people to use condoms led to a big drop in infections before ARVs became widely available. But once treatment programmes ramped up — from around 2012, when someone could start treatment at a higher CD4 count than the earlier restrictive 200 cells/µl — it was ARV uptake that really propelled the decline in infections. (The number of CD4 cells in someone’s blood shows how strong their immune system is. A count of 500 cells or more per microlitre of blood is seen as healthy, but a count below 200 is usually a sign of the immune system being weak. Initially, when ARVs were not nearly as readily available as today, only people whose CD4 counts had dropped that low were put on treatment).

For about 15 years early on in the epidemic, the incidence among children (the rate at which new infections are acquired) was a much bigger part of the total than it is today. In 2004, almost 95% of the just more than 81,200 cases among kids up to the age of 15 were from moms transmitting the virus to their babies. 

These numbers started to really drop only from around 2005, three years after the watershed Constitutional Court ruling that forced the government to offer all pregnant HIV-positive women the ARV drug nevirapine to stop the virus from infecting their unborn babies. 

Twenty-two is too many

But despite new infections in kids dropping 10-fold in about a decade, 22 children still got HIV in South Africa in 2023 every day. 

Part of this could be because of a fair portion of children starting to have sex before 15, something researchers call early sexual debut. The HSRC’s latest household survey on HIV shows that one in nine young people were sexually active by the time they turned 15. This could up young people’s chance of getting HIV, in part because young people are more prone to risky behaviour, like having multiple, concurrent partners or not using condoms. 

Biology also plays a role. 

In the case of girls, when tissues in the cervix and vagina are still developing, it’s easier for the virus to get into the cells lining those organs, especially if HIV is delivered through semen with a high viral load ( with lots of virus). On top of that, when girls start having sex young, it’s often with someone at least five years older than them, who has multiple sexual partners at the same time. In fact, in the HSRC survey just more than a third of the teen girls who have sex with men said their partners were at least five years older than them.

In 2023, about 30% of that year’s roughly 149,000 new HIV infections were among teen girls and young women between the ages of 15 and 24. All of this adds to why teen girls and young women are a group of people policymakers pay special attention to in efforts to curb HIV infection. 

Test and treat

The number of people who knew their HIV status jumped almost four times between 2000 and 2010. By then, almost two-thirds of people with the virus knew they were infected — and the gap closed rapidly over the next decade.

By 2020, 94% of people knew their HIV status, and by last year South Africa had already hit this first goal of the UN’s set of three 95s to end Aids as a public health crisis by 2030. (The two other targets are for 95% of people who are diagnosed with HIV to be on ARVs and 95% of those to be virally suppressed).

But despite the proportion of HIV-positive people on treatment almost doubling in five years — from 31% in 2010 to 58% in 2015 — and uptake of ARVs having grown to 76% over the five years since, it’s now levelling off, with a change of only three percentage points between 2020 and 2023.

Mind the gap 

Zooming in on South Africa’s progress towards reaching the 95-95-95 goals shows we’ll miss the second 95 target — getting 95% of people who know they have HIV on treatment — by almost a million (the cut-off date for the world reaching these ambitious milestones is the end of 2025).

Up to 2020 we were on track, and had the trend continued we would have hit the target by the end of next year. 

Instead ARV uptake started to taper off. 

Apart from “taking our foot off the pedal”, Yogan Pillay, director for HIV and TB delivery at the Bill & Melinda Gates Foundation, says Covid also added to this slowdown. Studies show that during the pandemic, HIV testing and starting people on treatment in South Africa dipped (as also elsewhere in the world) — and the numbers suggest it’s been difficult to get back up to speed again. 

UNAids figures show a similar trend worldwide. 

“We’re seeing treatment scale-up go quite quickly, [but] we’re not seeing the level of new infections go down as quickly — and that’s a cause for great concern,” Mitchell Warren, head of the international HIV advocacy organisation Avac, told Bhekisisa earlier this month.  

Shrinking budgets are a problem too, says UNAids, with the world having about $10bn (around R180bn) too little to spend on HIV prevention and treatment. “Cuts in resourcing and a rising anti-rights push are endangering the progress that has been made [towards ending Aids by 2030],” read the agency’s statement accompanying the release of their latest report in July.   

Warren warns that under the new Trump administration in the US it’s probably “safe to say” that there will be cuts in government spending — including for programmes such as the President’s Emergency Plan for Aids Relief (Pepfar), which funds HIV programmes in African countries, including South Africa.  

“The issue is what gets cut, how quickly it gets cut, and what gets cut in a way that isn’t with strategy. If financial cuts mean people fall out of treatment, it would be a disaster. We can’t lose people; in fact, we need to get more people into treatment programmes.” 

Fewer deaths, longer lives

South Africa’s HIV response has added years to South Africans’ lives, analysis shows.

Someone born in 2005 — the year that Aids deaths peaked — could expect to live to 54. Ten years later, life expectancy at birth was 64 years, and today it’s 66. 

A modelling study shows that a big part of the increase up to 2014 is directly because of HIV treatment becoming available, and that there would have been 1.72-million more HIV-linked deaths between 2000 and 2014 if ARVs had not been available.

But the study also highlights — in tragically real numbers — that about 500,000 more HIV-linked deaths could have been prevented up to 2014 “if South Africa had moved swiftly to implement WHO [World Health Organisation] guidelines and had achieved high levels of [treatment] uptake”. In fact, the authors’ calculations show that at least 40% of such deaths in 2014 were in people who had been diagnosed but had not started treatment yet.

South Africa’s treatment guidelines generally seem to have changed two years after new WHO recommendations were released. For example, for someone to start ARVs when their CD4 count was around 350 cells/µl was allowed only from 2012, though the WHO backed this already in 2010. Similarly, the WHO relaxed the CD4 count requirement for treatment to 500 cells/µl in 2013; South Africa followed suit only in 2015.  

‘HIV is still an issue’

What’s become clear now, Warren says, is that the world, and also South Africa, can no longer just continue as before and expect new HIV infections to decline faster. “We have to think differently. One of the things we have to do is reimagine prevention programmes.” 

Take lenacapavir, for example, the twice-a-year HIV-prevention shot that kept 100% of users safe from getting HIV in a large, multi-country trial, results released in July showed. 

The jab hasn’t yet been approved by any regulatory agency, including South Africa’s medicine regulator, Sahpra, though experts expect it could “get approved sometime by the middle of next year”, Warren notes. “It could be a game changer — if the world moves fast enough.” 

In October, the national health department’s head of procurement, Khadija Jamaloodien, told Bhekisisa they’re considering publishing a formal request for manufacturers to say at which price they can provide the product, how much of it they can make and what they will charge for it, before Sahpra’s process has been concluded.

Warren urges: “We need to begin preparing today, because I would argue that we have spent a decade squandering the opportunities of a daily [prevention] pill, a monthly ring and the two-monthly injectable [CAB-LA]. If we want to bring the number of new infections down, we cannot afford to squander this opportunity too.”

Linda Pretorius is Bhekisisa’s content editor. She has a PhD in biosystems from the University of Pretoria has been working as a science writer, editor and proofreader in the book industry and for academic journals over the past 15 years. At Bhekisisa she helps authors to shape and develop their stories to pack a punch

Jacques Verryn is a software developer and independent consultant, with special interest in data analysis and visualisation

This story was produced by the Bhekisisa Centre for Health Journalism. Sign up for the newsletter