Recruitment and retention of women in fishing communities in HIV prevention research

Introduction Women in fishing communities in Uganda are more at risk and have higher rates of HIV infection. Socio-cultural gender norms, limited access to health information and services, economic disempowerment, sexual abuse and their biological susceptibility make women more at risk of infection. There is need to design interventions that cater for women's vulnerability. We explore factors affecting recruitment and retention of women from fishing communities in HIV prevention research. Methods An HIV incidence cohort screened 2074 volunteers (1057 men and 1017 women) aged 13-49 years from 5 fishing communities along Lake Victoria using demographic, medical history, risk behaviour assessment questionnaires.1000 HIV negative high risk volunteers were enrolled and followed every 6 months for 18 months. Factors associated with completion of study visits among women were analyzed using multivariable logistic regression. Results Women constituted 1,017(49%) of those screened, and 449(45%) of those enrolled with a median (IQR) age of 27 (22-33) years. Main reasons for non-enrolment were HIV infection (33.9%) and reported low risk behaviour (37.5%). A total of 382 (74%) women and 332 (69%) men completed all follow up visits. Older women (>24 yrs) and those unemployed, who had lived in the community for 5 years or more, were more likely to complete all study visits. Conclusion Women had better retention rates than men at 18 months. Strategies for recruiting and retaining younger women and those who have stayed for less than 5 years need to be developed for improved retention of women in fishing communities in HIV prevention and research Programs.


Introduction
Fishing communities (FCs) in Uganda are still among key populations for HIV/AIDS spread despite first noticing HIV more than two decades ago in a fishing community [1]. FCs along Lake Victoria have an HIV prevalence of 22-29%, with an incidence of 3.4 -5 per 100 person years at risk (PYAR) [2][3][4][5].
Women in fishing communities are disproportionately affected by HIV/AIDS with a prevalence of 33-34 % and incidence of 3.4-4.5/100PYAR, more than four times the national averages [2][3][4][5][6] . It is important for women from fishing communities to participate in HIV prevention research for the design of prevention interventions that suit them. Social, economic, cultural, biological factors that facilitate or hinder recruitment and retention should be taken into account when considering women's participation. Factors that may affect women's recruitment in HIV prevention research and vulnerability to HIV/AIDS include; Socio-cultural norms regarding sex, with women being expected to be chaste, subservient partners, making it hard for women in fishing communities to participate in HIV research and be fully informed of safer sex and reproductive health, let alone negotiating safer sex [7,8].
Motherhood, a social expectation of women which makes contraceptive use and protected sex challenging for women without children even those with fewer children [9]. HIV prevention research that requires use of contraception may be hard for women in fishing communities for fear of being labelled social misfits in a society that values fertility. Women with children may fail to participate in HIV prevention research that seems to risk their safety as mothers and the safety of their children [10,11].
African culture values the head of household (usually a man) as the final decision maker, including decisions that directly affect the wife's health [12]. This hinders women's participation in research as they have to seek informal consent from the head of the household or even withdraw consent if the head wasn't aware of her participation at study commencement. Fishing and related tasks are most times gender specific, with the more paying roles being dominated by men. This may encourage HIV risk activities like transactional sex for economic survival and diminished ability to negotiate safer sex [13][14][15][16]. Physical violence, sexual, psychological and financial abuse of women may increase HIV vulnerability, limit access to prevention research, health information and care services.
Many a woman may fail to communicate such abuses for fear of losing their spouses/partners from whom they are socioeconomically subordinates [17][18][19][20]

Recruitment and follow up of volunteers
A prospective observational 18 months cohort study recruited and followed 1,000 high risk HIV negative consenting volunteers aged 13-49 years from 5 conveniently selected (population size more than 1000 and distance to research within 50km), fishing communities along the shores of Lake Victoria during 2009-2011.
Recruitment procedures have been previously described [5].
The study aimed at determining HIV prevalence, incidence,   Table 1).

Women recruitment
Recruitment from the five fishing communities was done over 6  Table 2).

Discussion
Less than half (45%) of the cohort participants were women with the main reasons for none enrolment being HIV infection (prevalence was higher among women (33.9% vs 23.9%) [2] and reported low HIV risk behavior. Women were more readily available for follow up visits, as 93% came back for at least one follow up visit and 74% successfully completed all three scheduled visits.
Retention was associated with older age, more than 24 years, staying in the community for at least 5 years and not earning an income. The HIV incidence cohort recruitment was conducted among HIV high risk participants enrolled from a cross sectional survey following mapping and census of the 5 fishing communities [5], which is strength for the study. Our study had some limitations: Use of face to face interviews may have led to under reporting of socially undesirable risk behaviour [23].
Participants were not specifically asked why they decided to participate in the research, why they decided to come for study Page number not for citation purposes 5 visits and reasons for missing study visits or dropping out, which could have provided more information on recruitment and retention.
Fewer women were recruited into the cohort probably due to a higher HIV prevalence among women compared to men [2], which is similar to another study [24], though less women reported HIV was associated with loss to follow up [26].
Cohort retention rates for women in fishing communities were better than those of men, probably because fewer women were engaged in actual fishing which requires more time away from home, hence were readily available within their communities and homes [2]. Older women, more than 30 years were more likely to attend all study visits as they were more settled, spending fewer nights away from their homes in the previous month. These findings are similar to other studies done in rural communities [27]. However younger women (<24 years) who were less likely to return for follow up visits were also found to have a higher HIV incidence in other studies [4,5,28], implying that efforts towards retaining younger women in HIV prevention research and other prevention programs are needed in these fishing communities.
Most women in fishing communities were married with financial dependants, which too explain the higher retention rates and likelihood of attendance of all scheduled study visits as married women tend to have more social support. Being unmarried is a risk factor for STIs including HIV [28], due to increased number of sexual partners, diminished ability to negotiate safer sex and limited health awareness [7,8,14,15]. Women without income were better retained in the cohort because they were more readily available to comply with study visits. Access to health care provided during the study could also have been motivating factor as women without income might have been finding it harder to access better health services due to financial limitations [7,29].
None income earners might have been also interested in the little monetary compensation for time spent at each of the follow up visits. Income earning women were less likely to attend all study visits probably because the reimbursements for time spent were not sufficient for them to leave their work schedules and come for study visits. Similar studies did not find an association between employment and study retention [27,28].
Having stayed in the fishing communities for at least five years was associated with better cohort retention as women in fishing communities who have lived in the communities longer are less likely to migrate and have over time established social support net works and can easily be located by other community members.
Women who were new immigrants (<5 years in the community), who were less likely to return for follow up visits were found to have a higher HIV incidence in another study [5]. This is similar to another study where migrants engaged in more higher risk sexual behaviours [8]. This may imply that HIV prevention research targeting high incidence groups in FCs may need to devise innovative ways of retaining new immigrants in such studies.

Conclusion
Women in fishing can be recruited in HIV prevention research with favourable retention rates. There is need to design strategies to make retention rates better especially for young women. This will ensure optimization of HIV preventive research and care for women in FCs.

Competing interests
The authors declare no competing interests.

Authors' contributions
Ali Ssetaala participated in coordination of field teams, field data collection, cleaning, conceived the initial manuscript idea and