In PNG, the epidemic that wasn't

See allHide authors and affiliations

Science  11 Jul 2014:
Vol. 345, Issue 6193, pp. 158-161
DOI: 10.1126/science.345.6193.158

Data vacuum despite improved surveillance.

A decade ago, the forecast for the island of New Guinea was dire. A heterosexual epidemic was set to explode, a team of leading international epidemiologists predicted in a report called AIDS in Asia: Face the Facts. They warned that “there is every indication that the island is facing an epidemic which resembles those seen in parts of sub-Saharan Africa.” Papua New Guinea (PNG), which shares the island with Indonesia, structured its response to its epidemic based on those early predictions. But the bomb never went off.

To this day, PNG struggles to describe not only why those early predictions were so wide off the mark, but also the exact contours of the smaller, but still serious, epidemic the country is experiencing today. That confusion has come at a steep price, particularly when it comes to decisions about where to target prevention efforts. “It gives rise to a very unfocused program,” says Stuart Watson, country coordinator in PNG for the Joint United Nations Programme on HIV/AIDS (UNAIDS). “We don't know exactly what we're responding to, and, as a result, we respond to everything.”

Epidemiologists have long had good reason to fear a serious heterosexual epidemic in PNG. Reported condom use is low. Medically supervised circumcision, which protects heterosexual men from HIV, is rare. Sexual violence, in contrast, is rampant; a survey published in the October 2013 issue of The Lancet Global Health found that 40% of PNG men interviewed reported having raped a female “non-partner.” One recent study in a highlands community and Port Moresby, the capital, found that more than 20% of the 154 people tested at sexual health clinics were infected with chlamydia or gonorrhea.

Early predictions suggested that by 2014, 10% of the adults in PNG would be infected with HIV. According to the PNG National Department of Health, the estimate of the adult prevalence today is 0.65%. Watson thinks that understates the true prevalence. “Ask most people cold-face if they believe those figures: no,” he says. But he believes the actual figure is well short of 10%.

Angela Kelly-Hanku (left) with Jolyn Gane in an AIDS ward at Goroka General Hospital.


“The epidemic in PNG doesn't make a lot of sense in some ways,” says epidemiologist Andrew Vallely, who lives in Milne Bay province and works with both the PNG Institute of Medical Research (IMR) and the University of New South Wales (UNSW) in Sydney. “Why has HIV not taken off?”

In trying to gauge HIV prevalence in PNG, epidemiologists have had far more assumptions than hard data. Joanne Robinson, a strategic information adviser at UNAIDS in PNG, notes that from 1993 to 2001, the only official HIV prevalence data came from one antenatal clinic at the general hospital in Port Moresby. “They did have very high prevalence rates, but it wasn't representative of the rest of the country,” Robinson says. Today, data come from 380 sites, and as surveillance has expanded to ever more remote, rural areas, the overall prevalence has dropped.

But the data still give an incomplete—if not outright misleading—picture. “Our estimates are based on extremely limited samples of the population to this day,” Watson says. He notes that only 60% of pregnant women attend antenatal clinics and a mere 40% receive HIV tests. The government has not done systematic, large-scale studies of several “key populations” that often have high HIV prevalence: men who have sex with men (MSM), transgenders, and sex workers. A small study done in Port Moresby by IMR in 2010 did show extremely high HIV prevalence in people who sold sex, whether they were females (19%), males (8.8%), or transgender women (23.7%).

Moreover, the epidemic is geographically patchy, with high heterosexual prevalence in some regions. As the PNG health department noted that year, just five of the country's 22 provinces, all in the highlands, accounted for 60% of reported HIV cases even though they have just 40% of the population. “PNG is now experiencing an epidemic concentrated in particular geographical locations and population groups,” states an HIV/AIDS response progress report it filed in March with UNAIDS.

Such patchiness complicates surveillance. For one, it is hard to access many isolated communities on this island of steep volcanic mountains. To reach the highland town of Goroka from Port Moresby, for example, requires either an expensive flight or a torturous 7-day hike. And the 800 different languages spoken on the island reflect cultural variation, including sexual practices that can affect transmission, such as polygamy and the age of initiation. “What you say about one place is not what you say about another place,” says social anthropologist Angela Kelly-Hanku, who lives in Goroka and works on Vallely's team.

One surprising cultural factor that may have had a role in curbing the epidemic is traditional penile cutting. Several large-scale studies in Africa have shown that medical circumcision—which removes the entire foreskin under sterile conditions—reduces the risk of sexual transmission of HIV from women to men by about 60%. While medical circumcision is uncommon in Papua New Guinea, up to 70% of men in some communities report having had a traditional penile cut as a boy, a procedure that involves cutting different types of slits into the foreskin but not removing it.


Circumcision practices in PNG, understanding HIV's scope in PNG, and HIV in a PNG picture house,

Following a recommendation from UNAIDS and the World Health Organization in 2007 that prevention programs in high-prevalence countries should incorporate male circumcision, PNG health officials became interested in launching a national program. With funding from the Australian government, Vallely's group collaborated on a 4-year study to gauge the procedure's acceptability and potential impact. What they found surprised them.

When the researchers assessed penile cutting around the country, the places that practiced it the most had the lowest HIV prevalence. “It may help explain why the epidemic looks as it does in PNG,” Vallely says. “To be honest, we never imagined when we heard of these practices that this would be the result.”

In collaboration with colleagues in PNG, Vallely and co-workers are now conducting epidemiologic studies to see if the link holds up. At UNSW's Kirby Institute for Infection and Immunity in Society, another group is doing lab studies with foreskins from PNG to explore fundamental mechanistic questions: How does medical circumcision thwart HIV, and does traditional cutting have any impact (see sidebar)? Vallely says if traditional penile cutting does prove capable of lowering the risk of HIV infection, it may ultimately alter public health campaigns. “We don't want to stop something that prevents HIV from taking off in PNG,” he says.

Watson maintains that confusion about the epidemic's contours and its drivers wastes precious resources. In 2012, most government spending went toward managing the response, not delivering services like prevention, treatment, and care. “We have a very top-heavy national response that gobbles up nearly 80% of the funding,” he says, noting that the National AIDS Council employs more than 100 people.

The painful Catch-22 is that PNG is left with little money to improve surveillance and figure out how best to curtail its epidemic. “We don't have the resources for that because we're still responding to an epidemic we don't have,” Watson says.

View Abstract

Stay Connected to Science

Navigate This Article