Evidence on the impact of partner reduction

Generally it is difficult to tease out a reduction in the number of partners as a contributor to lower HIV incidence and prevalence. One place where there are some indicators is in Uganda. HIV incidence fell there between 1989 and 1995¹ and prevalence fell between 1992 and 2002 (from between 15 and 30% to between 5 and 12% in pregnant women).² These reductions are clear, but the evidence as to the contribution of partner reduction and increased monogamy to the decline in the figures is much less so.

The second study cited in the last paragraph, for instance, found that abstinence (i.e. delay of sexual debut), partner reduction, and condom use all had their parts to play in the reduction in incidence:

• Abstinence: Median age at sexual debut increased from 16.5 in 1989 to 17.3 in 2000 for women, and from 17.6 in 1995 to 18.3 in 2000 for men.

• Being faithful: Premarital sex among women and multiple partnerships decreased between 1995 and 2000, though there were no significant changes in reporting of extramarital sex among men.

• Condoms: The proportion ‘ever’ having used condoms increased from 1% among women in 1989 and 16% among men in 1995 to 16% and 40% in 2000, respectively.

Moreover, Uganda was the only country in the region to report such declines in casual sex. Figure 4 shows 50 to 70% declines in the number of people reporting it, between 1989 and 1995.

Around the same time, surveys in Kenya, Zambia and Malawi were reporting undiminished high rates of extra-primary sex:

A USAID report on Uganda³ points out that in the mid 1990s, two large randomised trials at Rakai and Masaka in Uganda attempted to look at the impact of STI treatment on reducing HIV incidence, but found that though treatment certainly reduced the amount of STIs, it had no effect on HIV incidence. The main reason for this lack of effect on HIV from STI treatment was the large decrease in risky sex/multiple partner trends that had occurred in Uganda by the time the STI trials began. Most HIV transmission was therefore now occurring within monogamous regular partnerships, where one partner had undiagnosed HIV, but where bacterial STIs would tend to be rare.

More recently, steep (and apparently real) declines in HIV prevalence have been observed in Zimbabwe.

The study⁴ that found the declines pointed to the multifactorial nature of the apparent reasons for the decline:

“We report a decline in HIV prevalence in eastern Zimbabwe between 1998 and 2003 associated with sexual behavior change in four distinct socioeconomic strata. HIV prevalence fell most steeply at young ages - by 23% and 49%, respectively, among men aged 17 to 29 years and women aged 15 to 24 years - and in more educated groups. Sexually experienced men and women reported reductions in casual sex of 49 and 22%, respectively, whereas recent cohorts reported delayed sexual debut. Selective AIDS-induced mortality contributed to the decline in HIV prevalence.”

At the 2006 PEPFAR Implementers’ Meeting in Durban, Dr Owen Murungi from Zimbabwe’s Ministry of Health and Child Welfare teased out some of the reasons for the decline.⁵ This research is also available in a UNAIDS report published in November 2005.⁶

Dr Murungi said that, after the dramatic decline in HIV prevalence during 2004: “The big question to all of us was, is this real?” A review was therefore conducted to determine whether other available data corroborated the finding.

The UNAIDS report cites data from a number of studies of HIV incidence in Zimbabwe. These found falls from around 5% a year in 1993 to 3.6% in 2000 in pregnant women and from 3.5% in 1994 to less than 2% in 2001 in male factory workers.

Reductions in HIV incidence could be the result of natural dynamics of the HIV epidemic. Over time, any epidemic is somewhat self-limiting. Mortality plays more than one part in this. It decreases prevalence directly because of deaths due to AIDS, and can decrease incidence as well, by decreasing the pool of infectious individuals who can spread the infection.

But reductions in HIV incidence can be due to behaviour change too. The UNAIDS report did not find statistically significant evidence of increase in the age of sexual debut. But over the last few years, there does appear to have been a clear and substantial fall in the percentage of young men who reported having had sex during the last 12 months with non-regular partners. In Manicaland Province, statistically significant changes in reported sexual behaviour were observed for both males and females in:

• The age of sexual debut
• New partners in the last year/month
• The number of current partners.

The Manicaland report found substantial evidence of partner reduction, especially in men under 30, where the proportion of men saying they had had ‘non-regular’ sexual partners in the previous 12 months declined from 58% in 1999 to 21% in 2003. Among women of the same age the trend was less significant but the proportion reporting non-regular partners declined from 17 to 8% in the same period (see chart below).

Reported condom use with non-regular partners had also increased in the last five years. Dr Murungi noted that there has also been a steady increase in the number of condoms in circulation, particularly socially marketed condoms (rather than public sector condoms). A chart similar to the one above detailing increases in condom use in Zimbabwe is in Condoms and lubricants.

In Zambia there was a dramatic fall in HIV prevalence in young pregnant women (15 to 19) between 1993 and 1998 - the proportion living with the virus halved, from 28.4 to 14.8%. One study⁷ found “a dominant declining trend in HIV prevalence that corresponds to declines in incidence since the early 1990s attributable to behavioural changes”, which predominantly means a decline in casual sex.

In Ethiopia, a country with a more recent HIV crisis, the proportion of men who reported casual sex at two centres fell in just two years (1997 to 1999) from 17.5 to 3.5%, and the proportion reporting visiting sex workers from 11.2 to just 0.75%.⁸

In Cambodia,⁹ HIV prevalence halved between 1997 and 2002, as did the proportion of men who reported visiting a sex worker over the year, while condom use, already high, increased less dramatically (see chart).

Swaziland, currently the country with the world’s highest HIV prevalence, is also the first country to try to address specifically the phenomenon of long-term extramarital partnerships.

At the Toronto International AIDS Conference in 2006¹⁰ Derek von Wissell of the Swazi National Emergency Response Council on HIV/AIDS presented the results of their campaign.

The first phase of the campaign, featuring poster images of smartly-dressed, ‘aspirational’ men and women saying “I will not share my partner” achieved a 44% reduction in the number of people who had had two or more partners in the last four weeks (from 16 to 9%), a 25% reduction in people who intended to have more than one partner (from 12 to 9%), and a slight increase in the number of people who had been sexually abstinent during the previous six months. This was accompanied by a 36% decline in the proportion of people who said “I always use condoms with a new partner” (from 25 to 16%), but this may reflect the decline in the number of extramarital partners.

The second phase of the campaign was harder-hitting, juxtaposing plain-text mock-ups of phone text messages with invitations to second partners such as ”He’s working late, cum work on me” with other ‘texts’ such as “Your secret lover will kill you” and “Why kill your family?”. While these examples are in English, most campaign materials were in the local language siSwati.

The campaign caused huge controversy in Swaziland, including one of the largest demonstrations the small country has ever seen. This was called by the Swaziland National Network of People Living with HIV/AIDS, which described the campaign as “insulting” to the one-third of the population that already has HIV.

The campaign, which due to the protests only lasted ten days, probably benefited from the publicity, reaching 100% recognition in Swaziland.

An evaluation of the second phase of the campaign¹¹ found that comparison of the survey to 2005 baseline data revealed some positive and apparently immediate changes in behaviour among men. The numbers who reported having two or more partners in the preceding four weeks dropped by nearly half. Approximately 16.7% of men reported having two or more partners in the last four weeks, compared to 30.6%  in 2005. There was a similar trend in multiple sexual partners in the preceding six months, with men in 2006 significantly less likely to say they had multiple partners over that period.