The tiny kingdom of Swaziland in southern Africa with a population of approximately one million people has one of the highest infection rates of HIV/AIDS in the world. Recent census reports have shown that the population has dropped by approximately 17,000 people since the last census when half the population was aged 16 or under. If the results are accurate this places the country at nearly 30% of the population below what was projected as the likely rate of growth 20 years ago. Stephanie Nolen of the Globe and Mail interviews several persons who point out the arguments and implications surrounding the results.
WHERE HAVE ALL THE SWAZIS GONE
From Saturday's Globe and Mail
December 22, 2007
MATSAPHA, SWAZILAND - Solomon Dlamini decided the numbers were wrong: When the Swazi government released preliminary results of the national census a few weeks back, Prof. Dlamini, head of the national university's department of demography and statistics, took one look and concluded that the bureaucrats had made a terrible error. The census, a door-to-door count through all of this country's hilly villages, found 912,229 Swazis. That is 17,489 fewer than were counted at the last census, a decade before. Shrinking populations aren't news in developed countries such as Germany or Japan, but in developing countries such as Swaziland, in African countries where half the population was aged 16 or under the last time they counted, populations do not shrink until now. "My training says that in demographic history, this is unprecedented and it cannot be accurate," Prof. Dlamini said in an interview this month in a small office crammed with books. "But my reality says otherwise."
The Swazi government, reacting in consternation to what the statistics imply, is refusing to call the census figures final. But if they are accurate - and most experts believe they are - they mean that Swaziland's population has not only dropped by 17,000 people; it is 300,000 people, or nearly 30 per cent of the population, below what was projected as the likely rate of growth 20 years ago. While the figures are debated, no one here is uncertain about the cause of the drop. "If I were to sit down to count the people I have lost to
HIV, I wouldn't get up again," Prof. Dlamini said with a sigh. He found himself recently arguing with colleagues; some said that it simply isn't conceivable that the population has actually shrunk. No developing nation has ever shrunk. But others pointed out that no country has ever, in recorded history, had an epidemic like that of HIV in Swaziland. And nobody knows what it will mean for the country's future.
A toxic mix of factors combined to make the epidemic so bad here: a highly infectious and virulent strain of the virus circulating in the population; a culture that condones promiscuity and polygamy in men, while denying women the right to refuse to have sex or insist on the use of condoms; a limited economy that relies on sending workers away from home for long stretches to work in highly infected South Africa; and a government, led by a playboy king with an ever-expanding stable of wives, that has denied the scale of the problem, and, while people were dying, poured funds into luxury-car purchases and highway expansions. Once the epidemic had dug in, in the late 1990s, Swaziland's grossly understaffed and underfunded health system collapsed under the pressure. As a consequence, only achingly slow progress has been made in getting anti-AIDS drugs to the sick. So Swaziland has experienced a rate of death that would appear to have outstripped even the high birth rate and the growth of the young population.
This country has the world's highest rate of infection: 26 per cent of adults, 43 per cent of pregnant women, a staggering 49 per cent of young women between the ages 25 and 29. Life expectancy has fallen from 60 in 1997 to 31 in 2004, the lowest rate in the world. Only one in four people will live to the age of 40. "From 1986 to 1997, we grew by more than 200,000 people," Prof. Dlamini said. "But it's the period between these two censuses [1997 and 2007] when the epidemic reached its apex." He had anticipated that the new census would show a growth in population of between 50,000 and 80,000 people, "which would still be 120,000 below the projection," but now thinks he was being naive. "Maybe as a demographer sitting here, you underestimate the degree of the devastation."
"I don't think anybody quite realized what the depth of HIV would be in Swaziland," said Derek von Wissell, director of the National Emergency Response Council on HIV/AIDS. Normally energetic and determinedly positive, Dr. von Wissell slumped in his chair and stared at the ground while he discussed the census. "We knew people were going to die, but we started to do treatment and we stayed upbeat and then ... you see this." Nationwide, he said, it's been a wake-up call. "Even if they undercounted by 10 per cent, we're down 25 per cent from where we should be."
Development experts comment that the population has dropped both because HIV infection lowers fecundity, the number of children women have, and, of course, because of death. "And the people dying in the greatest numbers are your young, productive people, the ones who make the largest contribution to the economy," explained Alan Whiteside, who heads the Health Economics and HIV/AIDS Research Division at the University of KwaZulu-Natal, and who has made extensive study of Swaziland. Nearly half of all deaths today are of people between 16 and 35. (On a normal population curve, of course, it is the elderly who die in the greatest numbers.) "We don't have any idea what this kind of a loss will mean for Swaziland, there's never been anything like it. Where, for God's sake, do they go from here?"
There are two African countries with HIV infection rates that approach Swaziland's: Lesotho and Botswana, both of which also have small populations and a migrant-labour population that travels often to heavily infected South Africa. But Botswana has not seen a dramatic decline in population, while the 2006 Lesotho census found an increase of about 20,000 people to 1.8 million, compared with 10 years earlier. Dr. Whiteside predicted that Botswana, which is far wealthier than Swaziland, has good health care and a much more comprehensive ARV treatment program, will not shrink. But if Lesotho's census were redone in 2009, he predicted, it would likely show a population contracting like Swaziland's. "They are a bit behind Swaziland on the epidemic curve," he explained - that is, the highest levels of mortality had yet to kick in when the last count was done. South Africa, which has the world's highest number of people with HIV/AIDS, might also be shrinking or close to it, but the process is masked by the huge influx of refugees, including at least three million from Zimbabwe.
It is not clear what will happen with Swaziland: The rates of new infection have begun to drop slightly among young people, but they remain stable or are increasing among people in their 30s. It is not clear whether the rate of death has yet reached its highest point. Only a third of people who are sick enough to need ARVs are getting the drugs; if that figure does not accelerate significantly in the coming months, then the population may well continue to shrink, Dr. Whiteside said.
Not everyone is persuaded of the Swazi figures: "Experience has taught me to be skeptical of census data in general," said Rob Dorrington, a professor of actuarial science at the University of Cape Town, who said he finds it hard to believe that AIDS-related mortality could be so high, and suspects other causes for the drop. "It is not unusual for there to be an undercount of children and of men [in a census], and deaths would have to have been implausibly high, given the estimated level of prevalence, for one to be able to detect this through the change in the numbers counted by the census."
Amos Zwane, Swaziland's senior statistician, wrote in his preliminary report on the census that "a population decline or stagnation was not expected and this result is most surprising." His office is going to find a logical explanation, he said, and will not speculate on the cause until they produce final numbers in the middle of 2008. The census takers did not do a post enumeration check, which most censuses do to assess the impact of undercounting. Prof. Dlamini called that a shocking oversight, but then, he said, he is not sure how much it would help. If they were to somehow "find" an additional 30,000 people, the country would only be breaking even, replacing young people as fast as they die. "And it's impossible that they would find 300,000 more people," he said, head in hands - and that's who is missing.