Cristina de la Torre
Shane Khan
Erin Eckert
Jennifer Luna
Todd Koppenhaver
Publication Date
May 1, 2009

MEASURE Evaluation, Macro International Inc. (de la Torre/ Khan/ Eckert/Luna) and USAID/Namibia (Koppenhaver)

This report, produced by MEASURE Evaluation and USAID/Namibia for the Ministry of Health and Social Services in Namibia, identifies and describes what current evidence indicates are the main behavioural and contextual factors that are driving the HIV epidemic in Namibia. The report is intended to assist in the development of a national prevention strategy for combating the HIV/AIDS epidemic.

Data from several sources are triangulated to assess which factors are most likely to contribute to the spread of HIV across the population. In the absence of a national seroprevalence survey, the following were examined to identify the main drivers of the Namibian HIV epidemic:

  1. The prevalence, distribution, and trends over time of proximate determinants of HIV infection within Namibia (obtained from an analysis of the Namibia Demographic and Health Surveys [NDHS] and other local surveys).
  2. Socio-demographic factors associated with HIV infection among clients who were tested for HIV in select New Start voluntary counseling and testing (VCT) facilities throughout Namibia (obtained from analysis of these data).
  3. The findings and conclusions of other researchers who have investigated various aspects of HIV/AIDS vulnerability in Namibia.
  4. Factors most associated with HIV infection in neighbouring countries, and in other generalised epidemics for which representative HIV prevalence surveys exist.

According to the report, a number of factors are likely contributing to the high levels of HIV in Namibia. These various factors are often inter-related and operate in unison to create what these researchers consider to be one of the worst HIV epidemics in the world. The factors are outlined below.

Multiple and concurrent partnerships are likely contributing to the rapid spread of HIV throughout the country. In 2006, 16 percent of sexually active men and 3 percent of sexually active women reported more than one partner over the previous 12 months. Several local studies have also recorded high levels of concurrent partnerships throughout Namibia, although nationally representative data are not available. Having multiple partners is not common, nor apparently a major risk factor for HIV for the majority of women. However, the widespread practice among men of maintaining multiple relationships is contributing to the high levels of HIV infection among women, especially young women.

Intergenerational sex exposes adolescents and young adults to partners who, by virtue of their age and longer sexual history, are more likely to be HIV positive. Among women age 15 to 24, 7 percent of single women and 26 percent of married women have a partner 10 or more years older. Intergenerational sex in Namibia is associated with higher levels of sexually transmitted infections (STIs) and with a greater likelihood of having multiple partners. Intergenerational relationships introduce the virus into the younger cohort, where it quickly spreads as a result of rapid partner turnover and common concurrent partnerships (especially among young men).

Pervasive alcohol abuse and low levels of HIV risk-perception serve to foster multiple and concurrent partnerships, and may discourage consistent condom use. Nationally, 78 percent of men and 62 percent of women used a condom at last sex with a non-marital non-cohabiting partner. In Caprivi and Kavango, regions facing the worst of the epidemic, condom use is the lowest in the nation. Furthermore, low levels of male circumcision are reported in some of the areas with the highest HIV prevalence, namely Caprivi, Ohangwena, Omusati and Oshikoto.

Over the years there has been a steady decline in marital or cohabiting relationships. In 2006, approximately 1 in 3 Namibians ages 35 to 39 had never married or cohabitated with anyone. For women, never marrying or cohabiting was associated with having a greater number of sexual partners over one's lifetime. In most African countries one of the strongest predictors of HIV infection is the number of lifetime sexual partners.

Transactional sex appears to be common, and even expected, in many sectors of Namibia, although research that quantifies this practice is lacking. In the context of widespread poverty and limited employment opportunities, sexual intercourse has become a commodity freely traded for goods and services by men and women. Women appear to be particularly vulnerable to transactional sex, possibly because their marital independence has not been matched with new income generating opportunities and many remain economically dependent on men.

High levels of population mobility also accelerate the spread of HIV. Namibia serves as a corridor for much traffic to and from Southern Africa, receiving migrants from the highest prevalence countries in the world. Furthermore, Namibia's reliance on the mining and fishing industry, as well as on seasonal agricultural production, requires regular internal population displacement. Travel away from home is associated with an increase in multiple partnerships in Namibia. Infections are passed on rapidly through a chain of interconnected sexual networks that can be distributed over various sections of the country. With multiple and concurrent partnerships relatively common in both rural and urban areas, the epidemic has spread to all regions of the country.

The evidence strongly suggests that young women are at highest risk of acquiring HIV. Recent projections estimate that nearly half (44%) of new infections over the next 5 years will occur among 15 to 24 year olds; 77 percent of these will occur in young women. These women are most likely infected early in their sexual life by their first or second partner. It appears that the risk for many women stems from their choice of partner rather from their own behaviour. Only 27 percent of women aged 15 to 49 reported more than two partners in their lifetime, and multiple partnerships were not a risk factor for HIV infection among female VCT clients.

Based on these findings the report makes the following recommendations:

Special attention should be to be paid to the vulnerability of young women. In Namibia, two subgroups of young women appear to be particularly vulnerable to HIV infection.

  1. Young, educated, employed, and urban women who are least likely to abstain from sexual relations if not married, and most likely to have multiple partners, and have sexual relations under the influence of alcohol. These women, however, are also most likely to use condoms, although the extent to which this counters their risk is unclear. They also represent a fairly small group of women.
  2. Young married and cohabiting women, particularly the economically poor and uneducated ones who are mainly exposed to risk through their spouses, and who are far less likely to use condoms or be able to negotiate sexual relationships.

Testing should be encouraged among young married couples, and among older individuals who plan on cohabiting, since condom use among married and cohabiting partners is so low. Prevention efforts for young women must not just target them, but their sexual partners as well. This entails working with older men, as well as educated and employed men who report higher levels of multiple partners and may be engaging more frequently in transactional sex.

Programmes that aim at changing social norms, rather than individual behaviours, may be needed to tackle challenging issues such as transactional sex and intergenerational relationships. Further research on the significance of partnerships, and on partner turnover, would be beneficial.

Creating education and employment opportunities for women in urban and rural areas should be a central component of a national prevention strategy. Limited economic opportunities for women, and their continued economic dependence on men, are likely behind the high levels of transactional and intergenerational sex, both of which are key drivers of this epidemic. Both education and employment proved to be important protective factors against infection among women testing at New Start VCT clinics.

Developing realistic perceptions of risk should be a priority, particularly among young men. This is necessary to achieving further gains in condom use and in reducing multiple and concurrent partnerships. The population needs to be better informed of how prevalent HIV is in their community.

The factors that are driving the epidemic do vary by region and location, and strategies and programmes should vary accordingly. The rural population is harder and costlier to access but should not be ignored. These populations have recorded similar prevalence to urban areas in sentinel surveillance, and may have slightly higher risk (as indicated by VCT client data). High multiple partnerships combined with lower levels of knowledge and condom use are making rural populations vulnerable. Greater research should be undertaken in these communities to understand the dynamics of sexual partnerships in these areas.

It will be important to work with migrant communities, particularly agricultural workers and transport workers who appear to be particularly vulnerable to infection. Reaching these populations where they work or in bars and shebeens are options that should be examined in greater depth. Creating entertainment alternatives that do not involve alcohol and prostitution should also be considered.

Finally, further research is needed to understand the epidemic and inform programmatic decisions. A representative seroprevalence survey would help to more precisely determine the factors directly associated with HIV infection in the country. Further research is also needed with regards to transactional sex, concurrent partnerships, the formation and duration of partnerships, and perceptions of risk. The evidence suggesting a link between male circumcision and HIV in Namibia should further be investigated, with attention to the different types of circumcision currently practiced in the country.


Measure Evaluation website on August 18 2009.