Article Text
PDF
Abstract
Objectives The primary objective was to assess the acceptability of a savings intervention in which female sex workers (FSW) would save part of their earnings and call back (withdraw) when faced with a financial need that could force them into HIV risk practices. The secondary objectives were to assess its feasibility, concerns and design considerations.
Design A cross-sectional survey. Participants were asked for views on the intervention, their earnings, saving and spending practices, and suggestions for the intervention package.
Setting Kisumu and Siaya counties, Kenya.
Participants FSWs aged ≥18 years, self-identifying as sex workers and living in Kisumu or Siaya county.
Outcome measures The primary outcome was the proportion of participants who believed the Jitegemee intervention would be acceptable to FSWs in Kenya. The secondary outcomes were the proportion who: could generate money to save (assessed from income, spending and loaning practices), reported potential challenges with the intervention and suggested components to inform the intervention package.
Results We enrolled 369 FSWs, 88% aged 18–39 years, 78% unmarried, 94% cared for ≥1 child(ren) and 78% were household heads. Over half (52.1%) had been in sex trade for ≤4 years, with 62.3% reporting <10 clients the previous month. Jitegemee was highly acceptable, at 94.8%; however, participants suggested adding: financial literacy, including saving, spending and loans management (74.8%), forming saving groups (37.5%) and goal-setting (24.1%). Those who did not care for children were 4.86 times more likely to save (adjusted OR (aOR)=4.86, p=0.18), non-household heads were less likely to save (aOR=0.57, p=0.28) and those in the sex trade for 1–4 years and 5–9 years were four to five times more likely to save than those <1 year (aOR=4.49, p=0.01 and aOR=5.22, p=0.01, respectively).
Conclusions Jitegemee intervention was highly acceptable; however, several recommendations were suggested to make the design more appealing and potentially effective.
- Public health
- HIV & AIDS
- HEALTH ECONOMICS
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
Female sex workers (FSWs), as end users of the Jitegemee intervention, participated in designing the study, including the development of the questionnaire, which ensured the study was appropriate and responsive to their needs.
Enrolling FSWs from different sex worker typologies (street-based, brothel-based, home-based, entertainment venue-based and beach-based) expands the generalisability of the results to FSWs in other settings in Kenya.
Relying on acceptability of a proposed intervention to predict actual uptake when the intervention becomes available may overstate the true acceptability.
Given that participants were not asked for reasons for their choice of what to include in the intervention, we may not know why they made the choices they did.
Factors such as earnings, savings, loans and expenditures which are important to the intervention were collected through self-report, thus prone to social desirability and recall biases.
Introduction
Despite sex work being illegal in most sub-Saharan African (SSA) countries, including Kenya, some women resort to the practice as an alternative source of income when they cannot find other opportunities.1–5 Sex work gives them financial independence and the ability to improve their economic status.6 7 Besides this, social factors such as dysfunctional families, lack of education, peer pressure, seeking sexual pleasure and homelessness compel young women to join sex work.2 3 These economic and social factors do not only drive women to sex work, but make them financially dependent on their male clients and less able to negotiate condom use.8 This underscores their vulnerability to sexually transmitted infections including HIV, unintended pregnancies and complications from unsafe abortions, stigma and discrimination, violence and drug and alcohol addiction.9 10
In Kenya, the first phase of a size estimation activity in 2018 estimated the population of female sex workers (FSWs) at 167 940.11 The report also estimated the HIV prevalence among FSWs at 29.3%, compared with 6.6% among women in the general population.12 FSW’s risk of HIV infection is greatly influenced by social, legal and structural factors.13 14 Multiple sexual partnerships, gender-based violence and rape, no/low capacity to insist on condom use, sex while intoxicated and justice systems that criminalise sex work contribute to the elevated risk of HIV among FSWs.13 15–17 These risk factors have been associated with economic disempowerment of FSWs, limiting their ability to say no to unsafe sex or to higher pay that comes with it18–20 or to exit sex work even when they want to.21
Multiple interventions have been implemented to lower the risk of HIV among FSWs while they are still engaged in sex work, but most of them have narrowly focused on sexual risk behaviours despite the recognised importance of economic factors as drivers of HIV risk among this subpopulation.4 A systematic review of interventions to reduce the risk of HIV among FSWs globally showed that none of the 26 studies selected addressed economic security as an intervention to reduce HIV risk.22 Another systematic review on sex work interventions in SSA also found no economic empowerment component among the interventions assessed in the 25 selected studies.23 The few interventions that focus on economic empowerment are often geared towards ‘rehabilitation’ of sex workers,2 3 24 and are premised on the assumption that economic hardship drives women into sex work, therefore providing an alternative source of income would draw them away from the sex trade25 rather than keep them safe within sex work.
Promoting FSWs economic empowerment may provide structural protection from HIV.8 26 A study in Uganda found that when FSWs have access to more capital and invest to start earning additional income outside of sex work, they are likely to be empowered and improve their economic status, thereby reducing their STI (sexually transmitted infections)/HIV risk.27 Another study in Tanzania on savings among FSWs showed that savings provided a financial safety net for FSWs because it accorded financial security, improved their ability to negotiate safe sex and enabled them to be selective about clients.28 In Kenya, a study on precautionary savings intervention among vulnerable women including FSWs found that reductions in reported transactional sex and symptoms of sexually transmitted infections were associated with improved savings.29 For women who may wish to quit sex work, financial insecurity is often a major deterrent.9 30 31 A study among FSWs in Thailand showed that just 1 in 42 sex workers interviewed had never quit sex work, 60% had gone through one or more quit-re-entry-quit cycles, while 38% quit and never returned.21 Therefore, interventions aimed at empowering FSWs to reduce their HIV risk after leaving sex work need to address their economic insecurity.9 21 28 32 Strategies for increasing economic security of FSWs can include microfinance, vocational training and income-generating activities, cooperative banking and savings and money management.3 28
Savings-led have been shown to promote FSW’s financial security and reduce their likelihood of having unprotected sex.3 26–28 Between February and April 2022, we conducted a mixed methods study comprising of quantitative and qualitative data collection with FSWs in Kisumu and Siaya counties, in western Kenya. Our key objectives were to assess the acceptability, feasibility, concerns and design considerations of a savings intervention known as Jitegemee (rely on yourself) in which FSWs would be encouraged to save part of their earnings through a mobile banking platform known in Kenya as M-Pesa to fall back on when faced with an emergency financial need that may compel them to engage in unsafe sex during sex work. To support the FSWs in saving, they would be given instructions on saving and how to call back their savings. Additionally, the intervention would support the FSWs to save and reach a certain level of economic security that would allow them to say no to unsafe sex without fear of losing income for basic needs. Those who would wish to exit sex work would be guided to identify their preferred path to economic independence during or after sex work and supported to set realistic goals and timelines, and to work towards achieving them.
Even though there are alternative platforms for saving including saving through banks and insurance, and Rotating and Savings Credit Association (ROSCA) or self-help groups, saving through M-Pesa would be acceptable perhaps due to its comparative advantage.33 It presents a savings platform that has fewer restrictions or minimal fees for withdrawals compared with: insurance which tend to be risky and uncertain34 and requires a substantial fixed monthly premium; ROSCA or self-help groups which enables member to get soft loans with easy and flexible repayment terms that could be convenient and sensitive to FSW’s needs but involves opportunity cost of time spent attending meetings and the risk of default by members which has led to many such associations breaking35; or saving money in the bank which is safe and has loaning products but less attractive due to the perceived bureaucratic processes and higher interests on loans and service fees.33
Methods
Study design
We conducted a cross-sectional study on acceptability, feasibility and design considerations of the Jitegemee (rely on yourself) intervention to support FSWs to improve their financial security and consequently reduce their risk to HIV during or post-sex work. While under development, the protocol and questionnaire were reviewed by five FSW peer educators (PEs) who gave comments that were incorporated into the final versions. We have used the Strengthening the Reporting of Observational Studies in Epidemiology cross-sectional reporting guidelines36 to prepare this paper.
Eligibility and sample size considerations
To be eligible, women had to be ≥18 years; report exchanging sex for money, services, goods or favours in the previous 30 days; resident of or receiving HIV prevention or treatment services in Kisumu or Siaya county; and willing and competent to provide written informed consent for study participation. Since there was no known study on estimated acceptance rate for economic empowerment programme among FSWs at the start of the study, we assumed the statistically optimal option of a 50% acceptance rate to arrive at a sample size of 370 participants for the survey.
To obtain views from different typologies of FSWs (brothel-based, street-based, home-based, venue-based and beach-based) in the two counties and minimise bias while improving on generalisability of the results, we allocated approximate slots as follows: 55 home-based (30 in Kisumu and 25 in Siaya), 50 brothel-based (30 in Kisumu and 20 in Siaya), 115 entertainment venue-based (75 in Kisumu and 40 in Siaya), 90 street-based (50 in Kisumu and 40 in Siaya) and 65 beach-based (25 in Kisumu and 40 in Siaya). The allocations were roughly based on the proportion of sex workers in each typology in each county, estimated from our programme data of more than 10 years’ of experience working with FSWs in the two counties.
Data collection
We trained PEs, who were themselves FSWs, to recruit participants. During recruitment, potential participants chose if they preferred to be interviewed at the research site, at their venue (eg, brothel) or some other safe place. Trilingual (English, Kiswahili and Dholuo) research assistants trained on the protocol, data collection tools and ethics explained the study, administered consent in the preferred language, screened for eligibility and conducted the interview with consented and eligible FSWs. The questions explored whether the intervention would be acceptable to FSWs, where FSWs typically met sex partners, their risk-taking behaviours, their earnings, savings, loaning and spending behaviours, their investment goals and assets owned, sources of income, health-seeking behaviours, including HIV testing history, views on and possible concerns over the Jitegemee intervention, preferred intervention components and alternative economic activities. Participants were also asked for activities FSWs would engage in to generate income towards saving. Data were collected between February and April 2022. The primary outcome was on acceptability (the proportion accepting the intervention) while secondary outcomes were on feasibility (the proportion who demonstrated the ability to generate money to save, assessed from their income, spending and loaning practices), concerns (the proportion who reported potential challenges with the intervention) and design considerations (the proportion who mentioned different components to include when designing the intervention package to make it more attractive to FSW).
The Jitegemee intervention was described to participants (box 1) before they were asked questions about their views on it. It was explained to them that this intervention will be anchored on the belief that FSWs are capable of saving part of their earnings to reach a certain level of economic security that allows them to say no to unsafe sex. For those who wish to quit sex work, the savings would accord them a stable alternative livelihood post-sex work so that they do not return to the sex trade. The intervention would involve asking FSWs their preferred path to economic independence during or after sex work, how they can save towards their goals and how long it would take to reach those goals, then support them to set realistic goals and timelines and to work towards achieving them. A key feature of the Jitegemee intervention that ensures sustainability is that participants would be supported to use their own earnings to finance their saving goals.
A brief description of the Jitegemee intervention
Interviewer to read out loud: in the next section I want to get your views on an intervention we are thinking about that may help you or your peers to stay safe from HIV or to plan for life after sex work for those who may be thinking about leaving sex work in the near future. The intervention will primarily prepare sex workers to reduce their risk of getting infected with HIV through saving. The intervention is known as Jitegemee, which means rely or depend on yourself. The reason we are calling it Jitegemee is that sex workers who take part in it will use part of their own income to save towards some level of economic independence. They will save through M-Pesa directly into an account opened by the study. The saved amount by each sex worker will be available to her to call back in part or in full any time she needs it so that she does not have to engage in unsafe sex because she needs money urgently. With the savings, sex workers can say ‘No’ to unprotected sex or to certain clients if they want to, or to take a short break from sex work if they need to rest. This is because they have savings and cannot go hungry, for example, because they said ‘No’ to unsafe sex or took a break. Some sex workers may also want the savings to go towards long-term goals such as investing, educating children or even quitting sex work in the future. The intervention staff will help those enrolled in Jitegemee to set saving goals and timelines, work with them to plan their savings while being able to support their other needs, and help them to monitor the achievement of the goals and address challenges that come along the way. The questions that follow will ask you what you think about such an intervention, whether it would work, what the intervention package should comprise of (ie, components), how we can implement it so it works well and the challenges we may face and how to address them.
The question on acceptability (Is it—Jitegemee intervention—something that FSWs in Kenya can accept?) had Yes/No/Maybe/Do not Know response options, with a follow-up question (What should such an intervention comprise of to be acceptable to FSW?) asked to those who responded Yes or May to the acceptability question. For questions on potential challenges and components of the intervention package, a list of response options was prepared with information obtained from the literature review and PEs of FSWs. However, the research assistants did not read out the options for participants to select from; rather, participants were asked, unprompted, what they would like to see included in the Jitegemee intervention and any ethical concerns they and other sex workers might have about the intervention. The research assistants matched the responses given to the options provided in the list; responses that did not match the listed options were recorded under ‘Other, Specify’ and later post-coded by the study team (see online supplemental material 1 for the questionnaire used).
Supplemental material
Data analysis
Data were collected manually through paper-based forms, entered by trained data staff in a password-protected, excel-based database and 30% were randomly selected and reviewed for accuracy and completeness of entry by a senior data officer for purposes of external quality assurance. The officer also ran statistical scripts to check out-of-range values and performed data inconsistency checks.
Data were analysed descriptively to describe the study population in terms of their earning, spending, saving and loaning practices. We also used descriptive statistics to evaluate the acceptability of the Jitegemee intervention, χ2 statistics to determine the association between socio-demographic characteristics of participants and both their saving and spending levels and logistic regression to determine how various participant characteristics influenced saving ability.
The study was conducted between February and April 2022.
Patient and public involvement statement
We involved PEs of FSWs to review the questionnaire and the proposed intervention design components to be presented to participants. A PE for FSW is herself a sex worker who is recognised as a leader and a role model by her peers who elect her to lead them. Once elected, a PE is trained on sexually transmitted infections and HIV to equip her with knowledge and skills to support her peers on behaviour change; PE also delivers condoms and lubricants to their peers at their places of work or residence.37
Results
We screened 373 FSWs and enrolled 369 in the study. Our results indicate that a majority of FSWs (40.8%, n=151) reported the need for a steady source of income as the main reason for joining sex work. Nearly one-quarter (23.8%, n=88) and about one-fifth (19.7%, n=78) cited increased family responsibility (assuming a breadwinner’s role due to being single or lacking financial support from spouse) and being widowed or separated from spouse (19.7%, n=78), as circumstances that drove them to sex work. Peer pressure (6.8%, n=25) and poverty (4.9%, n=18) were the least mentioned factors.
A majority of FSWs enrolled (88%) were aged 18–39 years, 78% were unmarried, 94% cared for ≥1 child(ren), 47% and 32.9% lived with 2–3 and 4–5 persons, respectively, and 78% were the head of their households (table 1). The highest level of schooling was relatively low, with slightly over half (54.8%) reporting primary level. Over half (52.1%) had been in the sex trade for ≤4 years, one-third for 5–9 years and 15% for over 10 years. A majority (86.2%) reported sex work as the main source of income with about two-thirds (62.3%) reporting <10 different male clients the previous month, 20.9% reported 10–30 different clients and 16.8% reported >30 different clients.
Socioeconomic and demographic characteristics of participants
Reported earning was varied, with a little over one-fifth reporting under Kenya shillings (KES) 5000 in the previous month (US$1≈KES 118), 27.4% earned KES 5000–10 000, 26.1% earned KES 10 001–20 000 and 24.5% earned KES >20 000. Monthly expenditure was categorised into four levels: lower level (below KES 10 000 per month), lower middle level (KES 10 000–29 999), upper middle level (KES 30 000–99 999) and high level (≥KES 100 000).
Most of the participants (83.2%) were in lower middle and upper middle expenditure brackets while 48.1% had over KES 10 000 in savings and one-third had saved below KES 5000. Participants reported saving their money mostly in banks (62.5%) and mobile money systems (27.7%); table banking (Chama—a group-based saving strategy) and Saving and Credit Cooperative Societies were minimally preferred, at 5.7% and 1.1%, respectively.
Figure 1 shows very high acceptability of the Jitegemee intervention by FSWs, at 94.8%; however, participants made several recommendations for improvements to make Jitegemee more acceptable and feasible, with the top three being: adding a component of financial literacy, including loans management (74.8%); forming saving buddies and cross-learning from each other about challenges, best practices and success stories (37.5%); and goal-setting (24.1%). A few participants (8.4%, n=31) expressed ethical concerns over the intervention, viewing it as a veiled attempt to force sex workers out of their trade (n=13), that the intervention is an indirect disapproval of sex work (n=11) and that there is a hidden intention to use sex workers as cash cow or to take their saved money and disappear (n=6).
Acceptability of the Jitegemee intervention: ethical concerns and suggestions for improvement. FSWs, female sex workers; SRH, Sexual and Reproductive Health.
Participants cited various activities to generate additional cash to save if they participate in Jitegemee, which included (multiple responses allowed): starting other income activities besides sex work (60.7%); soliciting for more customers (37.1%); reducing current spending (28.2%); working longer hours (20.6%); charging their clients more (15.5%); practicing higher-paying sex, specifically unprotected sex (4.3%) and anal sex (0.3%). We post-classified these activities into three levels of risk (high, medium and low; multiple responses were allowed): unprotected sex and anal sex were classified as high risk (17 responses); seeking more clients, working long hours and charging more were classified as medium risk, because we did not ask whether the sex would be protected or not (270 responses); and seeking alternative income sources and reducing spending were classified as low risk (328 responses).
Other findings show that FSWs in Siaya which is mostly rural, earned significantly less (χ2=30.88, p<0.001) compared with those in Kisumu, a more urban therefore better economically endowed setting. Specifically, 65.2% of FSWs in Siaya and 37.2% in Kisumu earned <KES 10 000, and 34.8% in Siaya and 62.8% in Kisumu earned >KES 10 000. Similarly, participants in Kisumu saved significantly more than those in Siaya (χ2=14.66, p=0.002), with 38.8% in Siaya and 56.1% in Kisumu saving above KES 10 000. When we explored whether participants spent more than they earned, we found no significant difference between the two counties (χ2=2.80, p=0.10), with only 15.7% in Kisumu and 9.7% in Siaya spending more than they earned.
To determine if there would be a need to segment the audience when implementing the intervention and/or to tweak the content to address the factors that would inhibit saving, we determined if participant characteristics influenced saving ability. There was a significant association between the selected participants’ background and their saving and spending behaviours (table 2). For example, saving and being head of household (χ2=5.97, p=0.015), a number of different sexual partners in the last 1 month (χ2=10.03, p=0.007) and the total income earned in the last 1 month (χ2=13.93, p=0.001) were all significantly associated. Additionally, there was a significant association between spending more than earning and having children under one’s care (χ2=4.14, p=0.042), number of different sexual partners (χ2=16.70, p<0.001) and the total income earned in the last 1 month (χ2=68.83, p<0.001). Age at enrolment (p=0.723), marital status (p=0.094), having children under their care (p=0.228), duration of sex work (p=0.148) and main income source (p=0.741) were not significantly associated with saving. Similarly, age at enrolment (p=0.278), marital status (p=0110), being head of household (p=0.122) and duration of sex work (p=0.059) were not significantly associated with spending more than earning.
Association of participants’ socio-demographic characteristics with savings ability and spending
Using logistic regression, we estimated the effect of various socioeconomic statuses of participants on their ability to save more than KES 5000. Based on unadjusted OR (uOR), various age groups had different saving behaviours, with age groups 25–29 (uOR=1.09, p=0.86) and above 40+years (uOR=1.60, p=0.47) having better saving trend than the reference age 18–24 years. Similarly, being unmarried, having no child under their care, living in a house with >6 persons, having above secondary education, being in the sex trade for more than a year, reporting more than 10 clients in a month and reporting income higher than KES 5000 were predictors of better savers with odds greater than the reference category (table 3). In the adjusted model, those who did not care for children were almost 4.9 times more likely to save than those caring for children (adjusted OR (aOR)=4.86, p=0.18), those who were not head of household were 0.57 times less likely to save than household heads (aOR=0.57, p=0.28), those in the sex trade for 1–4 years and 5–9 years were 4.5–5 times likely to save than those <1 year (aOR=4.49, p=0.01 and aOR=5.22, p=0.01, respectively). Finally, those with over 30 clients in the last month were 1.7 times more likely to save than those with less than 10 clients (aOR=1.74, p=0.51), and those earning over KES 30 000 were 28.37 times more likely to save than those with less than KES 5000 (aOR=28.37, p<0.01).
Logistic regression of socio-demographic characteristics with participants’ saving ability
Examining the typologies of sex work in the two counties, brothel-based and street-based FSWs had the highest proportion of savers above KES 5000 while 46.7% of beach-based and 42.9% in other (undefined) locations saved above KES 5000. Those operating in brothels were 2.95 times more likely to save than those operating at entertainment venues (aOR=2.95, p=0.28). Home-based FSWs were equally likely to save as those based at entertainment venues (aOR=1.00, p=0.99) while street-based were 1.2 times more likely to save than those based at entertainment venues (aOR=1.20, p=0.83); all associations with p value of ≥0.05 were insignificant.
Most of the FSWs (89.8%) said they would consider quitting sex work after they have educated their children and other dependents, acquired some assets, especially land and house or have started a viable business for sustenance. The majority (74%) said quitting sex work in the foreseeable future would be difficult due to increased financial burden against lack of stable alternative source of income while others (11%) reported being comfortable with sex work since it is an easy and fast way of making money that did not require financial capital, or that they were addicted to sex work so quitting was not an option (9%). A total of 275 FSWs (75%) have thought of leaving sex work at some point but felt they were not ready, and 56% know someone who had quit sex work and returned.
We also asked for spending lines and classified them into two categories—essential or core needs and non-essential or non-core needs—to identify areas from where savings can be obtained by adjusting the amounts spent. Essential or core expenditures included: (1) food, rent and utilities, medical, transport, house maintenance, cleaning supplies and communication (44.6% of total expenditure), and (2) school expenses were tuition, uniforms, stationery and other general expenses (21.2%). Non-essential or non-core expenditures were: (1) personal and home beautification (furniture, beddings, kitchen equipment, shoes, clothing, beauty products) and job-related alcohol, accounting for 26.5%, and (2) social support (weddings, funerals, donations, gifts), accounting for 7.7% of earnings.
Discussion
Our study assessed the acceptability, feasibility and design considerations of a savings intervention to reduce FSWs’ risk of HIV during and post-sex work. The findings show a very high acceptability of the intervention, at 94.8%. All the FSWs reported to be earning majorly from sex work with about half already reporting some savings, mostly in banks and mobile money platforms, table banking (Chama in Kiswahili, a group-based ‘banking’ strategy in which group members save money during regularly scheduled meetings from which they can take either short or long-term loans at a small interest), and saving and credit societies. Similar studies with FSWs have proved feasible in Asia; findings from pilot studies in Chennai, India, and Ulaanbaatar, Mongolia, have showed the feasibility of combining a savings-led or microfinance intervention with HIV sexual risk reduction programmes for FSWs.38 39 A study in Tanzania on table banking or rotating pay-out (known as Michezo, equivalent to Chama in Kenya) proved to be acceptable and fostered a sense of empowerment among FSWs.28 In Kenya, the Global Network of Sex Work Projects demonstrated the success of combining saving-led or micro-financing and HIV prevention intervention where FSWs contribute monthly and receive a share of profits from invested income on an annual basis.2 Although about 90% of FSWs are open to quitting sex work, 74% acknowledged that due to high financial burden and the lack of a stable alternative source of income, it would be difficult for them to quit sex work in the foreseeable future. These findings corroborate the results from a study in Harare, Zimbabwe, among adolescent FSWs who were hesitant to quit sex work due to the limited survival options and difficulty in getting a job with a stable income.40 41
While 2.3% of those who reported that the intervention would be acceptable to FSWs had no suggestion for modification, the majority recommended several components to include in the intervention design, such as literacy on saving and loans management, forming saving buddy groups for cross-learning of challenges and success stories among participants, goal setting, flexibility in amounts to save and integrity in managing the savings. Similar recommendations have been made in studies that have reported financial literacy as a critical component of successful savings-led or micro-lending interventions targeting FSWs.3 42 It has been suggested that more attention should be paid to financial literacy and business development training,3 incorporating more regular goal-setting activities39 and continuous education on the importance of savings, banking services, budgeting (especially household budgeting) and debt management.27
Guided by Jun and colleagues43 who recommend the need to explicitly seek views and voices of stakeholders involved in, and impacted by, an intervention at the key moments of the intervention design process to assure that not only are the outcomes of the intervention effective, but the processes to achieve the results are not ethically objectionable, we asked participants if FSWs would have concerns over the intervention. Only 8% of the participants expressed concerns over such interventions, perceiving it as a tacit strategy to force women out of sex work, a veiled disapproval of sex work, or a scheme for economic abuse by the study team. Trust issues have also undermined interest in savings managed by peers and may have prompted the recommendation of ensuring integrity in savings management. Therefore, meaningfully engaging sex workers to elect their leaders, instituting a trusted oversight body and supporting them to develop a robust financial governance system can encourage FSWs to save.
About one-quarter of our participants were saving on mobile money platforms which makes it easier to withdraw cash in emergency situations. The majority of the participants (62.5%) reported saving in banks, which is contrary to what other studies in Kenya and other parts of Africa have reported. In Kenya, studies show minimal saving in banks44 while outside Kenya, only 5% were saving in banks in Cote dÍvoire4 and 8% in Tanzania.28 We believe that our findings may reflect social desirability bias where participants may have told us what they assumed was the more ‘respected’ way of saving. While a savings intervention would need to take advantage of the two saving channels that are already popular with FSWs, it would be important to confirm in other studies the reported preference for bank-based saving. The findings however remain relevant for an intervention such as Jitegemee where funds that would be needed more urgently would be saved in mobile money platforms for ease of callback while those that are being saved for long-term plans such as starting a business, buying property or building a home kept in banks.
Our findings show that about two-thirds of the FSW’s expenditure was on household essentials and school-related expenses while one-third was on non-essentials including personal and home beautification, and support to social events. Further exploration of participants’ expenditure against earning revealed about 16% of participants in Kisumu and 10% in Siaya were spending more than their earning and relied on loans from table banking groups (commonly known as Chama), friends and family to bridge the gaps, putting them at more debt. This finding is not unique to our study. Results from a study in Abidjan, Cote d’Ivoire, showed that about 30% of FSWs reported expenditures that exceeded their income and that nearly all FSWs who were observed reported borrowing money regularly to manage their expenses, especially during slow periods.4 45 Other studies have shown that FSWs who report having debt or other economic hardships have been more likely to indicate greater exposure to sources of sexuallly transmited infections (STI).30 46 On the other hand, in as much as saving-led interventions can promote FSW’s financial security, they may also have a negative impact, especially during hard economic times or shocks. As a shock-coping mechanism, some FSWs may risk to have unprotected sex to earn more money20 47 to be able to save. This may increase their risk of HIV infection, especially when the offer comes from HIV-positive clients who are willing to pay more for unprotected sex. For an intervention like Jitegemee, examining what FSWs spend on provides an opportunity to explore what could be cut, postponed, adjusted down or cheaper options sought in order to release some earnings to dedicate towards savings. Therefore, supporting FSWs to manage their spending on non-essentials is important for an effective savings-led HIV intervention. As noted in our study, the push to increase savings is likely to be counter-productive, as it may lead to risky sexual practices that pay more. Some of our participants cited various ways of generating additional cash to bridge their income gaps in order to save, including soliciting for more clients, working longer hours and charging their clients more for unprotected sex. Risk reduction education should therefore be embedded within FSWs economic empowerment interventions.
While the immediate goal of the Jitegemee intervention is to make FSWs have savings for instant cash callback whenever needed instead of engaging in risky sex, a long-term aim is to prepare and enable those who want to quit sex work to fulfil their financial obligations such as educating their children, owning a house or property or investing in a business. For long-term outcome, entrepreneurship or business skills training has been shown to help FSWs achieve financial goals that can eventually replace sex work.4 48
Our results showed that background factors such as low-income levels, being head of household and caring for children were associated with lower savings. While low income and the burden of supporting one’s family may make it difficult for an FSW to save, studies have shown that some FSWs spend beyond their income4 49 and are not cognisant of their impulse spending, which likely contributes to high debt and low savings.4 In their study, Igonya and colleagues,50 found that FSWs were unable to save effectively because they held the belief that ‘quick money does not stay’, a mindset that makes it difficult for FSWs to save. We believe that financial literacy and management training have the potential to empower FSWs to strategise and prioritise their ways of earning, spending and saving.
This study had some limitations. We relied on the acceptability of a proposed Jitegemee intervention to predict the actual uptake when the intervention becomes available; therefore, due to social desirability bias51 52 the proportion who responded affirmatively may overstate the true acceptability. Second, we relied on self-reported data on factors such as earnings, savings, loans and expenditure which are important to the Jitegemee intervention; however, these may be inaccurate as most participants do not keep records therefore based their answers on recall and general estimation.53 Despite these limitations, to our knowledge, this is one of the very few studies that attempted to explore the acceptability and feasibility of a savings intervention for FSWs that will encourage them to save part of their earnings for use when faced with an immediate financial need that could otherwise compel them to engage in unsafe sex.29 44 In addition, an intervention grounded on using own income has a higher chance of being sustainable compared with those relying on external support.25 Finally, while previous studies have explored various interventions to lower the HIV risk among FSWs while they are still engaged in sex work, most have only focused on the sexual risk behaviours without concomitantly addressing economic factors that drive their HIV risk.27 54 55
In conclusion, our study has demonstrated that an intervention to support FSWs save part of their income for use in emergent needs instead of resorting to risky sex is highly acceptable and feasible. The findings suggest that saving may be harder among FSWs with school-going children and dependents, or who are household heads; therefore, it is important for the intervention to take into account these challenges and help FSWs plan on how to balance their income, expenditure and loan-taking and—repayment in order to obtain money to save. This can help in mitigate negative impact of saving such as situations where some FSWs may engage in risky unprotected sex to earn more money to be able to save. Participants also made valuable recommendations on components that need to be added to make the intervention more appealing. This informed the subsequent phases of a multiphase process of co-creating a savings-led intervention with FSWs to reduce their HIV risk. The positive finding of the acceptability study led to the implementation of a pilot randomised-control feasibility study (phase 2) and helped focus the in-depth review of literature on FSWs economic empowerment studies (phase 3). The list of intervention components was then presented to FSWs who prioritised those that could be tested on a larger scale by ranking (phase 4). Finally, a selected number of FSWs PEs participated in reviewing and modifying the draft intervention package developed in a workshop session (phase 5).
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by Maseno University Ethics Review Committee (MUERC 1033/21). Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The authors would like to thank the study participants, and the Ministry of Health leadership in Kisumu and Siaya counties who allowed the study to be conducted. We would also like to acknowledge the invaluable contribution of KA, who authored this paper. Sadly, KA passed away during the review process of this manuscript. Her dedication to this work and her significant impact on the field will be greatly missed. We are deeply grateful for her guidance, expertise and commitment.
References
Footnotes
Deceased The lead author passed on August 2024
Contributors KA and HT designed the study. JA, OO and SC collected the data. NO and TO contributed to the study design and analysed the data. KA, JA, G-NW, JO, MO and BA substantially contributed to the writing of the paper. All authors critically reviewed and approved the final manuscript. KA is the guarantor responsible for the overall content of the paper.
Funding Funding for the study was obtained from Impact Research and Development Organization (IRDO), Kisumu, Kenya, from its research capacity building reserve. Staff engaged in data collection, data analysis and interpretation, and in manuscript writing were all employees of IRDO. The decision to submit the paper for publication was granted by IRDO’s Management Committee.
Competing interests None declared.
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.