Our research has revealed several characteristics and experiences of women who engage in sex work in rural Namibia which help to understand the limitations of the effectiveness of the BCC program. Our research found that the critical factors related to having protected sex varied in different circumstances. For instance, when a FSW was low on cash, financial factors may be critical. If the FSW was attracted to the man, or if he was rich, or if she hoped to become his girlfriend or have his baby then she was more likely to have unprotected sex. These research findings are thus novel for this setting in that they go beyond thinking about increasing condom use as driven by education and even beyond the newer thinking of the 3Es: education, empowerment and economic independence to think about transactional sex as heterogeneous and thus driven by different factors. In our study FSW were driven by the desire for money or love or both, and this depended on the types of male partner (primary, regular, stranger), the conditions of the exchange (pre-determined, a gift) and on the socio-economic status of the man (local, business man, tourist).
In our study the participants were less commercialized or formalized as sex workers compared to urban settings. These informal FSWs did not consistently define themselves as a sex worker but instead engaged in transactional sex under certain circumstances. Since many HIV prevention programs target more formalized sex workers, the findings from our study have important policy implications as they demonstrate that in rural Namibia sex workers are not always a clear, defined group and transactional sex may be being used by much larger proportions of the population thereby increasing HIV transmission risk in the population. This underscores the need to move preventative interventions typically offered in a venue such as a brothel and targeted toward FSW to become more diffuse and reach out to the broader population of women who may not typically define themselves as a sex worker.
Due to the high rate of HIV/AIDS infection in Namibia, research, risk reduction programs and policy initiatives are required to combat HIV/AIDS in high-risk populations, such as women engaging in sex work, including those involved in this study. Such initiatives, like the BCC Program, commonly employ a knowledge, attitude, behavior (KAB) approach, whereby the acquisition of knowledge (e.g. correct and consistent condom use can prevent HIV transmission) should result in attitude adjustment (e.g. intention to use condoms), which should ultimately produce behavior change (e.g. consistent and correct condom use) . While this approach is likely helping to curb the HIV prevalence rates in many countries , the KAB model does not in itself adequately address the full range of factors influencing behavior change, and thus further comprehensive prevention strategies are warranted, particularly where HIV persists in vulnerable populations .
In particular, the KAB model often oversimplifies or ignores other internal and external forces impacting upon behavior change, such as financial and socio-cultural pressures. As this study demonstrates, knowledge of safe sex practices and the intention to use condoms do not necessarily result in consistent condom use for women engaged in transactional sexual activities. Condom use was also found to be influenced by sex workers' perceived control over their own sexual decision-making processes , which is tied to economic and socio-cultural constraints. This finding is consistent with studies of female sex workers in urban environments in Namibia , in other LMIC, Nigeria , Ghana , the Caribbean  as well as in high income countries, the USA  and Canada  but may, arguably, be even more heightened in rural, traditional environments like northern Namibia. As such safer sex for females engaging in transactional sexual practices-and Namibian women, in general-may be more accurately viewed as a product of the '3 Es' of education, empowerment, and economic independence. Our study revealed further nuances beyond the emerging acknowledgement of the importance of 3 Es. The desire to be loved may encourage FSW to consent to high risk sexual practices. There was a desire by some FSW to become a certain man's girlfriend/wife and this may involve having a child, in part because FSWs say "having his kid shows that you're his woman". Further, FSW talked about the social capital of having a wealthy man's child so much so that even if the man leaves the woman may be considered to be of higher social standing.
From this new knowledge we have gained further insight into the complex factors likely required to prevent HIV transmission in this population of FSW and thus potentially slow the epidemic throughout populations. For women who are similar to the FSW in this study, the KAB model is likely a useful beginning (education, condom distribution, voluntary counselling and testing) but even more important are the 3 Es (education, empowerment and economic independence) and potentially of utmost importance, gender equality and the opportunity to develop equal and loving relationships. To these ends therefore, recommendations for ways forward for individuals similar to the women included in this study include addressing gendered inequalities- economic and otherwise, that are maintained by patriarchal socio-cultural practices. Men and women, including community leaders, businesses and government should be encouraged to better represent women . This supports the previous recognition in Namibia that there is a need "to work toward elevating the perceptions of women in society through education institutions as well as political and traditional leaders" .
One specific strategy suggested by previous research in India and South Africa to empower and educate female sex workers is to continue training and encouraging them to become peer educators themselves . Although on the other hand, a peer educators approach in Namibia or elsewhere may not be helpful since it is unlikely to impact upon the economic needs for sex work-being a peer educator is rarely a paid role. However, it is possible that a peer educators approach could address other features of sex work in this setting, such as engaging in sex work with the aim to develop a loving relationship. Such a peer educator program however, will need to be trialled in future research to assess the program's content and approach and to evaluate its efficacy and acceptability. As well, the women in our study pointed out that community members negatively view programs aimed at empowering women, to the extent that a peer educator stated that men may become 'angry' if they feel a program is doing this. Thus, in contexts where such programs may be found threatening by men, future empowerment programs will need to take into account the safety of participating individuals.
Our research found that sex work did not necessarily provide the financial independence sought by the women. Men in these circumstances still set the rules, the prices and the conditions of the exchange and thus the women remained economically dependent on the male. This underscores the importance of facilitating females' economic independence through vocational training opportunities for interested sex workers, and developing income-generation projects and individual and group micro-credit programs targeting these women. As a result of the research findings reported in this paper which were circulated in 2007 as an internal report within the local Namibian NGO, programs to address these needs have recently been initiated. Coupled with developing and supporting the circumstances in which female sex workers could utilize their new skills and knowledge, this may allow these women to become more financially independent and potentially obtain a viable economic alternative to sex work. This increased economic security may further empower women, adding to their ability to negotiate safer sexual practices, which, in turn, reduces the risk of HIV acquisition and transmission [13, 17]. Since the recognition of the importance of economic security in relation to HIV prevention is relatively new there are few projects that have evaluated this approach in Namibia or elsewhere. For instance a 2004 Cochrane review  and a 2002 Lancet systematic review  of HIV prevention programs do not include evaluations of structural or economic programs. Though a 2009 publication describes an evaluation of the Sonagachi Project for Indian sex workers which indicates that broader strategies like social supports, advocacy, and micro-finance can significantly reduce vulnerability to HIV compared to more narrow approaches of clinical and education services alone .
There were several limitations involved in this study, including language difficulties and translations between English and Oshiwambo. However it is not anticipated that language challenges had a large impact on data acquisition or interpretation since BCC Program staff and other informants proficient in English acted as translators.
A second limitation is the potential that results have been affected by social desirability bias. Participants could have provided answers that they believed the interviewer desired. To avoid this potential problem, the interviewer attempted to remain open-minded and unbiased, avoided leading questions and organized interactions and interviews with participants without the presence of BCC Program representatives.
Finally, this study was restricted to exploring the views of informal female sex workers regarding their transactional sexual activities, including their use of condoms. The new knowledge gained from exploring the views of female sex workers in this study may not apply to all FSW. A range of other critical issues that emerged from the fieldwork were not pursued, including types of male partners (e.g. primary, non-primary/transactional, regular, one-time), conditions of exchange (e.g. money/"gift" given at every encounter, predetermined or variable amount), and socio-economic status of transactional partners (e.g. local man, businessman, tourist). Further analyses of these issues may similarly highlight qualitative information potentially important for HIV prevention strategies, and policy implications.