Southern African Regional Poverty Network (SARPN) SARPN thematic photo
Regional themes > HIV/AIDS Last update: 2020-11-27  

 Related documents


Preliminary results from a baseline study of HIV risk factors in Swaziland

A. Brody (PhD, UNICEF Representative in Swaziland) and J. Ascroft (PhD, Professor Emeritus, University of Iowa)

Posted with permission of Ms Sasrah Crowe, media officer at Unicef's Johannesburg office.
[Printer friendly version - 22Kb < 1min (3 pages)]     [ Share with a friend  ]

Summary of findings

Background and Methodology

As part of a communication baseline study on HIV risk factors, UNICEF Swaziland sponsored a 2003 survey in two rural districts (tinkhundla) in Swaziland. The main purpose of the study was to look at knowledge and related potential forms of risk behavior that could be influencing rapid spread of HIV, as a basis for further sharpening the design of a community communication and planning intervention that uses a UNICEF-developed communication tool called "The String Game Story." That story seeks to provide in-depth understanding of the range of both biological and social-cultural risk factors which are hypothesized to be facilitating the rapid spread of HIV in Swaziland.

After consultation with national authorities, the opportunity arose to include anonymous testing of HIV status for all respondents, in order to provide a check on possible relationships between specific risk behaviours and HIV status, and to secure a baseline on the ultimate dependent variable of the proposed communication intervention, e.g. HIV status. Using systematic random sampling of households drawn from aerial maps and census lists, the study selected 283 households, interviewed all those aged 12 and older present in those households (872 respondents), and carried out HIV rapid tests on over 1,094 persons (including 222 additional children aged 8-11 who were not interviewed). Through pre-survey provision of information and mobilization of community leaders, careful design of a "one page" questionnaire, and intensive training of interviewers, a negligible refusal rate was achieved (less than 10 percent in first district, which were subsequently randomly replaced, and zero refusals in the second district).

Comparison of Methodology with Faciilty-Based Estimates on HIV Prevalence

All previous data and estimates on HIV prevalence in Swaziland have been based on sentinel surveillance at antenatal sites. UNAIDS and others use models to make projections from that data to estimate prevalence among the overall population including males. Those projections are often criticized for overestimating HIV prevalence, and many non-specialists often draw false conclusions from them regarding overall population prevalence rates and numbers of persons infected. This study provided the first randomly sampled population figure to date for a Swazi population, although it was limited to two rural areas. A significant under-representation from the sample of males aged 20-45 means that findings cannot be projected to the larger national population, especially as young males in the cities may have a different risk profile compared to those they left behind in the villages. However, the sample included larger numbers of women, and a full complement of males and females aged 20 years and below, on which the most important findings below have been based.

Community based Estimates on HIV Prevalence by Age and Sex

The study results found 36 percent HIV prevalence for women aged 19-49, which (within the margin of error) was in line with the previous estimate of 38.6 percent HIV prevalence found among pregnant women in the 2002 national ANC survey. However, the study found a prevalence rate for age 15-49 females was nearly 10 percent lower than previous estimates, at 29 percent. This resulted from a highly significant disconnect in HIV prevalence in 18 year old girls compared to 19 year olds. A similar disconnect among males appears to occur between age 23-25, but because over two-thirds of males are missing in the 20-29 age group, it is difficult to draw conclusions about them. Adjustments for "missing males" who are absent partners of the women in the community suggest an overall HIV prevalence for those aged 8 and above of 17 percent, plus or minus 3 percent.

Evidence of Generational Behaviour Change-and an Hypothesis

Age 12-18 females in the study had HIV prevalence of four percent. Study findings on age of first sexual experience suggest a "generational behaviour change" has occurred, with only 14 percent of girls and 5 percent of boys in the age group 12-18 reporting initiation of sex, compared to over 60 percent in older age groups who reported they had their first sexual experience between ages 12-18. Among the age 8-18 years children measured in the study, boys had a slightly higher HIV prevalence rate than girls, with overall HIV prevalence (male and female) in this group at 5 percent.

It is hypothesized that two factors have contributed to the unexpectedly lower HIV rates among girls aged 18 and below compared to women aged 19 and above:

  1. Major campaigns since 2000 have "broken the silence" on issues of HIV and AIDS and sexual abuse, and promoted values of abstinence and virginity among children (both boys and girls), combined with wide-scale initiatives to "sweep out from under the carpet" the issues of sexual abuse of girls, including a widespread mobilization of religious and traditional leaders behind these efforts.

  2. The rapid spread of HIV from 1992-94 translated into large numbers of illnesses and deaths from AIDS beginning around 2000. Teenagers who in 2000-2001 reached the age of 15-17, which was the previous age of sexual debut for girls, could observe relatives and former role models aged 20-30 becoming ill and dying from AIDS, sending a stronger message than any poster or slogan ever could that AIDS is real, and not to be played with.
Knowledge About HIV and Other STI's

The study found extremely high (99 percent) awareness of HIV and AIDS, but comparatively low understanding of other STIs among teenagers. Condom use in the study population was less than 33 percent of those who are sexually active, with only 20 percent of females and 15 percent of males reporting they always used a condom. Another potential risk area identified by the study was "dry sex," with over 80 percent of women reporting sex to be at times "painfully dry."

Quality of knowledge about transmission of HIV was fairly good. A significant exception was an overestimation of the risks of oral sex, and a potentially very high risk underestimation of the risks of anal sex compared to normal vaginal sex. The latter misconception is a disturbing one in a context where parents or traditional authorities may be introducing "virginity testing" for girls, with the risk that prospective lovers who are ignorant of the higher risks of anal sex may resort to it. The unexpectedly high rates HIV prevalence rates among boys aged 8-18 also highlight the risks they face, and that the campaigns against child sexual abuse must include them as well, and that education for both boys and girls must include information about the very high risks of anal sex.

Sources of Information

Sources of information about HIV and AIDS were largely radio for older people, and teachers and health workers for the youth. Newspapers, television, and NGOs - which have often been viewed as key channels for "AIDS education", were not given by respondents as a major source of information, nor were peers. These finding have implications for where future communication efforts should focus if they are to reach the 70 percent of the Swazi population who reside in rural areas.


A Message of Hope

The study has important implications for action. For five years now, Swazis have heard nothing but bleak news about their future, and statistics, such as projections that "70% of today's 15-year-olds will die of AIDS", that seem to imply they are facing an all-powerful enemy.

These new findings now make it possible to give to young people in Swaziland a new "message of hope" that they are "95 percent HIV free". The communication challenge ahead is to give them the in-depth information, understanding and social protection which they need to stay that way.

Octoplus Information Solutions Top of page | Home | Contact SARPN | Disclaimer