In this paper we have described the structure and organisation of commercial sex work in a Kampala suburb. Our particular focus has been on the different places the women work, and how those locations influence risk, including safety from violence as well as STI infection, and also the potential income of the women in each place. These findings are corroborated by research in other settings which highlight similar experiences for women [6, 9, 14, 17].
Mobility was an important part of the lives of the women in our study. Not only did women vary their place of work, moving between the street, bars and lodges (depending on opportunity as well as their need for cash) but also by moving considerable distances to earn money. This travel may mean moving to a fishing site where the fish catch has been relatively good and the men have cash available or going to a workplace (construction site or oil palm plantation) on pay day. The risks involved in such mobility, because of high rates of STI infection, including HIV, at some of these sites highlighted in this study have been documented elsewhere [12, 18–20].
Other factors are also important. The regular and sometimes excessive consumption of alcohol by both the women themselves during their work, as well as by their clients, was mentioned often in our interviews. The impact of alcohol consumption on sexual behaviour and risk taking, particularly the failure to use condoms, has been documented by a number of different commentators [21–25], as well as in our study population . Because alcohol is both a means of coping with the provision of sexual services and an integral part of the entertainment industry which the sex workers also inhabit, interventions to reduce alcohol consumption are difficult to put in place in the absence of structural changes that may influence the availability and pricing of alcoholic drink.
The structural drivers of HIV infection, such as gender inequity, insecurity and poverty, pose a challenge for interventions with sex workers. These structural factors not only make negotiations over safer sex difficult for many women but also keep many women in sex work because of their need for cash to pay children’s school fees or simply to feed themselves and their children. Income generating opportunities may provide valuable alternatives for the day to day needs of a family, but when large expenses come, particularly when they are unexpected, selling sex may still be the most available way of raising cash. Structural changes often demand transformation in the wider society, but small steps to improving the conditions in which women work can be, and are, being made. There are well-documented examples of the success of approaches to address both the organisational and the legal status of sex workers in other places, most notably in Sonagachi, in India . The informal organisations that some groups of women in our study population had put in place to support each other, as well as manage their relationships with the police and other authorities, may provide structures on which to build initiatives that support women’s rights as well as promote their safety and security.
Countering the violence and risk women face in their work, especially among street-based sex workers, may require confronting the legal issues around sex work [1, 11]. However, such initiatives may not have the full impact desired. Some studies have shown that decriminalizing sex work may not result in improved conditions for all women engaged in sex work because of the many different situations in which women work, which are not easily regulated . In addition, the mobility of the women in our study population, as we have shown in this paper, presents challenges for approaches that are tailored to address the needs of stationary populations.
Another challenge is the label ‘sex worker’. Van den Borne  describes in her work in Malawi the reasons why many women rejected the label ‘prostitute’; they were women in search of a man to provide support, they were not selling sex. Some, as in our study population, may have had multiple partners but these were often transactional relationships with regular partners, men who were not viewed as ‘clients’. Such women may face many of the same risks as sex workers in bars or lodges in terms of risks of infection, but also violence and abuse, but may not be reached by interventions which use labels or descriptions that they feel do not fit what they do [15, 29]. It is very likely that some of the thirty four women in our study population who did not self-identify as sex workers fall into this category.
This study had some limitations which should be considered when interpreting the results. This was an urban population and may not give a picture of sex work in small towns or rural areas. Secondly, our sample was mainly from low socio-economic locations which may differ from high class sex work locations within the city centre.