Reprinted from Noseweek (South Africa) issue no 52, December 2003
Africa is Not Dying of AIDS
By Rian Malan

Another version of the article “Apocalypse When?” as it ran in the South African journal Noseweek.


As I write, my wife fingers the blade of a carving knife and eyes the back of my neck. She's sick of this. It's been going on forever. We met in Los Angeles in the 1980s, at the very dawn of the AIDS era. All the newspapers were full of prophecies about the heterosexual epidemic that was about to blight our lives. She wouldn't sleep with me until I had HIV clearance.

Later we moved to South Africa, where the anxiety deepened. Every newspaper we opened provided more details about the AIDS bomb that was about to explode over our heads. Premature deaths would double by 2010. Life expectancy would fall to 35. Africa's hunger crisis would persist "for generations" because so many people would be weakened by HIV infections.

I have seen pictures of people dying of AIDS - a terrible sight. I visualized Jo'burg, with such corpses stacked in the streets, as in the days of the black plague. It was scary, very scary. That's why I accepted a contract to assassinate Thabo Mbeki when he started questioning HIV basics. I speak metaphorically, of course. The weapon was words, and the contractor was a US magazine that suspected Mbeki had gone off his rocker. So did I. Everyone knew the facts, and there was one fact that overwhelmed all others: 250,000 South Africans had died of AIDS in 1999, the year preceding. A quarter-million deaths in a single year. This was horrible. Apocalyptic, in fact. By far the worst catastrophe in South Africa's history.

I thought of a novel way to demonstrate Mbeki's folly. My article would open with a scene in a coffin factory where men worked overtime to provide caskets for the dead while the president consorted with mad scientists who believed AIDS was a fantasy. I opened the 1999 Yellow Pages and set to work, only to discover that half of Jo'burg's coffin factories (okay, two out of four) listed there had gone out of business since it was printed. More surprising still, the survivors claimed it was business as usual, no boom anywhere. "How can that be?" I cried, citing official UN estimates of a quarter-million HIV casualties in the year preceding. Abe Schwegman of B&A coffins scratched his head. "I don't know," he said. "You tell me."

By the end of the day I was following Mbeki into AIDS dementia, and the rest was very ugly: ruined dinner parties, broken friendships, ridicule, fights with US editors. And I never even became a proper dissident. I was just interrogating the estimates, but the madam's patience wore thin anyway. Her eyes glazed over. Her fingers thrummed irritably on tables. After a year she put her foot down. Choose, she said. AIDS or me. So I severed all ties with conspiracy theorists, put my papers in the garage, and sat out the rest of the war on the sidelines.

Hostilities ended on November 12, 2003, when the government announced an R12bn, five-year program to provide antiretroviral drugs through state hospitals. The war was over. It was safe to come out again. I sat down at my machine. As I write, the madam is standing behind me, hands on hips. She does not like what she sees. There will be trouble unless I keep this brief.

WE ALL KNOW, THANKS to Twain, that statistics are often the lowest form of lie, but when it comes to HIV/AIDS, we suspend all skepticism. Why? AIDS is the most political disease ever. We have been fighting about it since the day it was identified. The key battleground is public perception, and the most deadly weapon is the estimate. When I lived in LA, HIV incidence was estimated to be doubling every year or so. In l985, a science journal estimated that 1.7 million Americans were already infected, with "three to five million" soon likely to follow suit. Oprah Winfrey told the nation that by 1990, "one in five heterosexuals will be dead of AIDS."

We now know that these estimates were vastly and indeed deliberately exaggerated, but they achieved the desired end: AIDS was catapulted to the top of the West's spending agenda, and the estimators turned their attention elsewhere. India's epidemic was likened to "a volcano waiting to explode." Africa faced "a tidal wave of death." By 1992 they were estimating that all humanity was threatened.

Who were they, these estimators? For the most part, they worked in Geneva for the World Health Organization (WHO) or UNAIDS, using a computer simulator called Epimodel. Every year, all over Africa, blood samples would be taken from a small sample of pregnant women and screened for signs of HIV infection. The results would be programmed into Epimodel, which transmuted them into estimates. If so many women were infected, it followed that a similar proportion of their husbands and lovers must be infected, too. These hypothetical unfortunates would proceed to die in dumbfounding numbers, leaving throngs of hypothetical orphans behind them.

Because Africa is disorganized and in some parts, unknowable, we had little choice other than to accept this. ("We" always expect the worst of Africa, anyway.)

Reporting on AIDS in Africa became a quest for anecdotes to support Geneva's estimates, and the estimates grew ever more terrible - 9.6 million cumulative AIDS deaths by 1997, rising to 17 million three years later.

Or so we were told. After my bad experience with Joburg's Yellow Pages, I visited the swamplands west of Lake Victoria where AIDS first emerged and was said to have wrought ghastly havoc. I gathered reams of anecdote about the epidemic, but statistical corroboration was hard to come by. According to government census bureau, death rates in the worst affected areas had been in decline since world war two. AIDS-era mortality studies yielded some of the lowest "crude" death rates (rate of deaths from all causes) ever measured. Populations seemed to have exploded even as the epidemic was peaking.

Ask AIDS experts about this, and they say the historic data is too uncertain to make valid comparisons. But these same experts will tell you that South Africa is vastly different - "The only country in sub-Saharan Africa where sufficient deaths are routinely registered to attempt to produce national estimates of mortality," says British demographer Ian Timaeus, a titan in his field. According to Timaeus, upwards of 80% of deaths are registered here, which makes us unique: the only corner of Africa where it's possible to judge computer-generated AIDS estimates against objective reality.

In the year 2000, even as I was reaching for the Yellow Pages, Timaeus joined a team of South African researchers bent on eliminating all doubts about the magnitude of our AIDS epidemic. Sponsored by the Medical Research Council, the team's mission was to validate (for the first time ever) the output of AIDS computer models against real-life death registration. Towards this end, the Medical Research Council (MRC) team was granted privileged access to death reports as they streamed into Home Affairs in Pretoria. The first results became available in 2001, and they ran thus: 339,000 adult deaths in 1998, 375,000 in 1999, and 410,000 in 2000.

This was grimly consistent with predictions of rising mortality, but the scale was problematic. Epimodel estimated 250,000 AIDS deaths in 1999, but there were only 375,000 adult deaths in total that year - far too few to accommodate UN claims on behalf of the HI virus. In short, Epimodel had failed its first and only reality check. The MRC quietly shelved it in favor of their own, local model called ASSA 600, which yielded a "more realistic" death toll from AIDS of 143,000 for calendar year 1999.

At this level, AIDS deaths were about 40% of the total, which left only 232,000 deaths from non-HIV causes. This, too, was implausibly low, but the MRC made the problem vanish by stating that deaths from ordinary disease had declined at the cumulatively massive rate of nearly 3% per year since 1985. This seemed very odd. How could deaths decrease in the face of new cholera and malaria epidemics, mounting poverty, the reportedly widespread emergence of drug-resistant killer microbes and a health system said to be in "terminal decline?"

But anyway, these guys were the experts, and their tinkering achieved the desired end: modeled AIDS deaths and real deaths were reconciled, the books balanced, truth was revealed. The fruit of the MRC's ground-breaking labor was published in June 2001, and Mbeki's hash appeared to have been settled, along with mine. To be sure, I carped about curious adjustments, but fell silent in the face of graphs showing massive changes in the pattern of death, with more and more people dying at sexually active ages. "How can you argue with this?" cried my wife, eyes flashing angrily. I couldn't. I put my AIDS papers in the garage and ate my hat.

A few months later, it started coming up again.

And here I must introduce Rodney Richards, whom I met in cyberspace at the height of my HIV obsession. Rodney was a dissident from Colorado, US, but also a scientist with impressive credentials, including a doctorate in biochemistry and a 10-year stint designing advanced methods of HIV diagnosis. Unlike many of his peers, Rodney had the grace to be tormented by doubts from time to time, and believed these would be settled in South Africa, the only African country where vast exaggerations could not pass undetected.

Around October 2001, he contacted me to say that the vaunted ASSA 600 model had been quietly retired and replaced by ASSA 2000, which was producing estimates even lower than its predecessor: for calendar 1999, a "mere" 92,000 AIDS deaths in total. Rodney interpreted this to mean that real-life deaths were not exploding at the predicted rate, hence the need for downward revisions. A year later came another surprise, this time in the form of a Stats SA study of actual death certificates. All deaths caused by HIV or any of its euphemisms were counted as AIDS deaths, and there was evidence for only 40,000 such in 1999.

I daren't pursue any of this, not with a wife sworn to leave if I reverted to old bad habits. Besides, I could barely credit it anyway, even when Census 2001 produced further alarms. Guided by computer-generated estimates, most demographers were forecasting that SA's population growth would slow to around 1.6% per annum as the AIDS epidemic kicked in. Instead, Census 2001 revealed that it was growing at a healthy 2% a year. I tried to ask Stats SA for an explanation, but their phones just rang. Ace demographer Professor Laurie Schlemmer told me that census mortality data is being withheld from publication, apparently because it is "sensitive."

And then this from Rodney a few months back: "ASSA 2000 has also been retired." He directed me to the MRC website, where I found a report noting that modeling was an inexact science, and that "the number of people dying of AIDS has only now started to increase." A new model was in the works, said the MRC, and it would probably produce estimates "about 10% lower than the estimates contained in this booklet."

The revision was necessitated; I was told, by the discovery of a tiny over-estimate of HIV prevalence in 2002. But small things have geometric impact in AIDS models, and lowering the 2002 figure causes the whole epidemic curve behind it to sag. Rodney was reluctant, because such an exercise is not scientifically valid, but I persuaded him to run the revised number on his own simulator and see what he came up with for 1999. The answer, very crudely, was an AIDS death toll somewhere around 65,000 - a far cry indeed from the 250,000 initially put forth by UNAIDS.

The madam has just read this, and she is not impressed. "It's obscene," she says. "You're treating this as if it's just a computer game. People are dying out there."

Well, yes. I concede that. People are dying, but that doesn't spare us from the fact that Africa's AIDS pandemic is something of a computer game. The real question is: is the game doing the dying - or us - any good? Modelers claim omniscience, but in truth, as they themselves admit, all estimates are subject to uncertainties and large margins of error. Much larger than expected, in some cases.

Consider, for instance, all those newspaper stories portraying South African universities as crucibles of rampant HIV infection. A year or so back, modelers produced estimates showing that one in four SA undergraduates were doomed. Prevalence obviously shifted according to racial composition and region, with KwaZulu-Natal institutions worst affected and Rand Afrikaans University (still 70% white) coming in at 9.5%.

In the case of RAU, this exercise predicted that one in 10 students would be infected, but real-life tests on a random sample of 1188 real students yielded a startlingly different conclusion: on-campus prevalence was 1.1%, barely a ninth of the modeled figure. "Doubt is cast on present estimates," said the RAU report, "and further research is strongly advocated."

Grahamstown district surgeon Stuart Dyer has reached a similar conclusion about HIV prevalence in South Africa's prisons, where up to 60% of inmates are said to be infected. "Sexually transmitted diseases are common in the prison where I work," he wrote to The Lancet, "and all prisoners who have any such disease are tested for HIV. Prisoners with any other illnesses that do not resolve rapidly (within one to two weeks) are also tested for HIV. As a result, a large number of HIV tests are done every week. This prison, which holds 550 inmates and is always full or overfull, has an HIV infection rate of 2-4% and has had only two deaths from AIDS in the seven years I have been working there." Dyer goes on to express a dim view of statistics that give the impression that "the whole of South Africa will be depopulated within 24 months," and concludes by stating, "HIV infection in SA prisons is currently 2.3%."

This seemed to imply that the Prisons Department had quietly conducted an HIV survey among inmates, but Dyer declined further comment, so let's turn to another enigma. SA's medical aid schemes have set up special programs to manage an anticipated flood of middle-class HIV cases. The modelers estimated that around 450,000 people are eligible, but only 22,500 have joined up.

In the Sunday Times, experts attributed this staggering shortfall to fear of stigmatization, implying that middle class South Africans are so prudish they would rather die than consult their GPs about a sexually transmitted disease. Do you buy that? I don't. Another argument is that most HIV-infected people show no symptoms and are therefore ignorant of their HIV-positive status. I don't buy that either - not when it comes to the middle class people who belong to medical aid schemes. Middle class people are tested when applying for jobs, bonds or insurance policies, and one assumes that middle class sexual athletes have tests just in case. They see the doctor even for relatively mild illness - which one in two HIV-positives should be experiencing. People are encouraged to join the program immediately they test HIV-positive, so that their condition can be monitored. Even if you accept, against all odds, that only 25% are aware of their condition, 112,500 people should have come forth to join the HIV management programs. Instead, we're stuck with 22,500, a four-to-one shortfall - which would tend to support other indications that the infection rate in this population has been significantly overestimated.

And so we come to the end of this diatribe. The ashtrays are overflowing, the room strewn with scientific papers. One says the number of AIDS orphans in Africa might be 30-50% lower than commonly claimed. Another states that real-life death rates among Africa's teachers are "typically two to six times lower" than estimated. A third - headlined, "Cry the beloved paradigm" - argues that sex can't account for HIV's modeled rampage across parts of Africa unless you assume a probability of transmission per coital act 300 times higher than anything ever measured. I could go on, but the madam is spitting mad, and it's more than my life is worth to continue.

Besides, I'm not sure what it all means. All that is clear at this point is that over the past four years, SA scientists have proved beyond any reasonable doubt that Geneva is producing estimates that seem designed to force a certain outcome. Because Western activists (and scientists, and drug companies) find AIDS more compelling than any other African problem, they say we should spend upwards of $300 dollars a year keeping an individual alive with AIDS drugs. This is a noble proposition - but what if that individual's friends and neighbors are dying in much larger numbers of starvation or politically less interesting, but equally fatal, diseases that could be cured for a few cents if medicines were made available, which often aren't?

It is time to have a debate about this. Such a debate couldn't begin until someone had assessed the accuracy of the estimates emanating from Geneva. The MRC rose to the occasion. We should be proud of them, and they of themselves. And we should share their discoveries with African and Asian states before they are overwhelmed by lobbyists seeking to commandeer a disproportionate amount of pitifully limited health resources to fight a condition that is undoubtedly deadly, but whose rate of occurrence appears to have been hugely overestimated, skewing political, financial and humanitarian agendas.

noseweek issue no 52


AIDS: Glad Tidings

Author and journalist Rian Malan has made a happy discovery: the statisticians at the Medical Research Council that three years ago announced that the AIDS apocalypse was upon us, have quietly been downscaling their estimates of AIDS deaths in South Africa ... to less than half the number of deaths they claimed were AIDS-related in 2000.

In fact, if you didn't know what's been cooked up in the statisticians' kitchen over the past two years, you might have thought the once massive HIV-AIDS epidemic is miraculously on the retreat in South Africa. Indications are that even their latest, reduced estimates - all based on contrived statistical "models", rather than on real, measured infection and death rates - are still too high. (One is tempted to ask: If the anti-retroviral drug program had already been rolled out, might we all mistakenly have attributed the now reduced AIDS death rate to the drugs?)

If you find Malan's piece hard to believe, note these recent developments:

HIV Management Solutions (a commercial spin-off of the Wits medical faculty) recently completed two important surveys of real HIV prevalence in South Africa: the one amongst employees of South Africa's four major banks, the other of all Telkom's employees. The surveys are believed to be the biggest HIV-prevalence surveys ever conducted in SA. Tens of thousands of were tested.

The banks' survey revealed an HIV prevalence rate of 3.4%. The Telkom survey, according to well-informed sources, found an infection rate of 2.7%.

This is less than half the already reduced rate most recently predicted by the MRC's statistical modelers.

Good news! Hurrah! Much less illness; far fewer deaths in the pipeline! you exclaim.

The banks were certainly happy to publish their result because it is not in their shareholders' interest to exaggerate the problem. (Why Telkom has not yet published its survey result we cannot imagine.)

But how come not a peep in the media? And why no mention on HIV Management Solutions' own website of any of their own survey statistics? (All you'll find there are the original horror forecasts of Dorrington et al at the MRC.) Must we deduce that it's obviously not in their shareholder's interest to cut the hype (or even just to present the facts)?

Another indicator of the real incidence of HIV-AIDS might be found at the JSE. All listed companies have to release earnings reports twice a year. In them they are also obliged to mention factors that have a material impact on their business. We've checked every one of them this year for mention of HIV-AIDS, and found only two such mentions - out of 167 companies that are listed.

The one was Discovery, as you would expect - but the only mention of AIDS in their 60-page statement was a line saying they had decided to close their AIDS reserve fund of R120m, no reason given.

The other was Shoprite Holdings, owner of Checkers, the biggest grocery chain in Africa. Their only comment: "The advance of HIV/AIDS has not had a noticeable effect on the Group's sales patterns or employment statistics for the year under review.''

Beyond anecdotes - stories one hears from doctors and friends - Africa's AIDS pandemic has indeed until now been something of a computer game. Which is not to say there are not people dying of this dread condition. The question here is: is the statistical computer game doing them - and us - any good? What we - and especially sufferers from AIDS - surely don't need is to have the subject reduced to a script for a B-grade horror movie.

Let's not forget: just one of the consequences of the dramatic reduction in the AIDS mortality estimates suggests the prognosis for those tested HIV-positive could perhaps be much, much better than was once predicted. And that they can now with greater confidence devote more of their time and energy to making the most of life, and be less pre-occupied with death.

It's clear to most intelligent people that AIDS in Africa, no matter what it's immediate cause, is another symptom of the misdistribution of resources between rich and poor - access to food, education, employment and all forms of medical care. Some of the celebrities at Nelson Mandela's 46664 AIDS concert touched on the point when they said that AIDS in Africa is a political and human rights issue, not just a health issue.

We suspect inviting Western teenagers to listen to pop songs for a minute or two - even if it raises hundreds of millions of dollars for the Mandela Foundation - isn't going to begin to solve the problem either.

- The Editor



to Africa In Perspective Related Articles - Africa - Treating Poverty with Toxic Drugs

to Mothers, Babies & AIDS