Home Current Issue Previous Issues Published Ahead-of-Print For Authors Journal Info
Skip Navigation LinksHome > June 2012 - Volume 39 - Issue 6 > High HIV Incidence and Socio-Behavioral Risk Patterns in Fis...
Sexually Transmitted Diseases:
doi: 10.1097/OLQ.0b013e318251555d
Original Study

High HIV Incidence and Socio-Behavioral Risk Patterns in Fishing Communities on the Shores of Lake Victoria, Uganda

Seeley, Janet PhD*,†,‡; Nakiyingi-Miiro, Jessica PhD*,‡; Kamali, Anatoli MSc*,§; Mpendo, Juliet MPH; Asiki, Gershim MSc*; Abaasa, Andrew MSc*; De Bont, Jan PhD; Nielsen, Leslie RN; Kaleebu, Pontiano PhD*,‡,§

Free Access
Article Outline
Collapse Box

Author Information

From the *Medical Research Council/Uganda Virus Research Institute, Uganda Research Unit on AIDS, Entebbe, Uganda; School of International Development, University of East Anglia, Norwich, United Kingdom; London School of Hygiene and Tropical Medicine, London, United Kingdom; §Uganda Virus Research Institute, Entebbe, Uganda; Uganda Virus Research Institute/International Aids Vaccine Initiative HIV Vaccine Program, Entebbe, Uganda; and International AIDS Vaccine Initiative, New York, NY

The authors are grateful for the funding given by the European and Developing countries Clinical Trials Partnership (EDCTP), Irish Aid, the Swedish International Development Cooperation Agency (SIDA), International AIDS Vaccine Initiative (IAVI), and the Medical Research Council of the United Kingdom (MRC, UK) to support this research. They also thank the field study teams, laboratory and data management staff for their contribution to this study, and to the participants for their time, information, enthusiasm, and support for this research. They thank the comments of two anonymous reviewers and Prof Jonathan Levin, Dr. Noah Kiwanuka, Dr. Stan Musgrave, and Dr. Richard White.

Correspondence: Janet Seeley, PhD, MRC/UVRI, P.O. Box 49, Entebbe, Uganda. E-mail: janet.seeley@mrcuganda.org.

Received for publication August 27, 2011, and accepted January 5, 2012.

Collapse Box

Abstract

Background: We report on HIV acquisition and its associated risk factors in 5 fishing communities on the shores of Lake Victoria in Uganda. A cohort of 1000 HIV-uninfected at-risk volunteers aged 13 to 49 years were recruited in 2009 and followed up for 18 months.

Methods: At enrollment and semiannual visits, socio-demographic and risk behavior data were collected through a structured questionnaire and blood samples tested for HIV and syphilis. Detailed life histories were collected from 78 volunteers using in-depth interviews.

Results: Of the 1000 volunteers enrolled, 919 (91.9%) were followed up, with 762 (76.2%) reaching the study end points (either seroconverted or completed 4 visits). There were 59 incident cases in 1205.6 person-years at risk (PYAR), resulting in an incidence rate of 4.9 (95% CI = 3.8 to 6.3) per 100 PYAR. The highest HIV incidence rates were among those working in bars (9.8/100 PYAR [4.7–20.6]), protestants (8.6/100 PYAR [5.8–12.7]), those aged 13 to 24 years (7.5/100 PYAR [5.2–11.0]), and new immigrants (6.6/100 PYAR [4.9–8.9]). HIV infection was independently associated with being young (adjusted hazard ratio (aHR) = 2.5 [95% CI = 1.3–4.9]), reporting genital sores/discharge recently (aHR = 2.8 [1.6–5.0]), regular alcohol consumption (aHR = 3.3 [1.6–6.1]), use of marijuana (aHR = 2.9 [1.0–8.0]), cigarette smoking (aHR = 3.6 [1.4–9.3]), and religion (compared with Catholics, Protestants had aHR = 2.7 [1.4–5.3] and Muslims had aHR = 2.3 [1.1–4.8]).

Conclusions: These fishing communities experienced high HIV infection, which was mainly explained by high-risk behavior. There is an urgent need to target HIV prevention and research efforts to this vulnerable and neglected group.

The UNAIDS epidemic update for 2010 recorded falling incidence rates in 22 sub-Saharan African countries, including in Uganda where HIV incidence has stabilized at 0.74% per year and prevalence has remained between 6.5% and 7.0% since 2001.1 These rates are from general populations and may mask the epidemic trends in higher risk groups, such as sex workers and fishing communities.25 Fishermen and fishmongers have been considered at risk of HIV infection in Uganda, but there have been little data to support this.6,7 In this article, we present HIV incidence and risk behavior data from 5 fishing villages on the shores of Lake Victoria in Uganda, drawn from 2 years of research. We have noted elsewhere that HIV prevalence in this population was reported to be 28.8%, increasing with age from 7.8% for those aged 13 to 17 years to peak at 36.4% among those aged 30 to 34 years, before declining to 14.8% in those aged 45 to 49 years.5

Back to Top | Article Outline
Study Sites

The fishing sites are spread across 3 lakeshore districts (Masaka, Wakiso, and Mukono). Masaka district had 2 sites that are about 120 km from the sites in Wakiso/Mukono and 50 km from each other. Wakiso had 2 sites about 5 km apart, and the Mukono site is an island about 40 km by boat from the Wakiso sites. The 5 sites have a total population of 15,415 people, 10,188 (66%) of whom were 13 to 49 years old. The study setting is described in more detail elsewhere.5

Back to Top | Article Outline

METHODS

Between February and August 2009, as part of an epidemiologic and social science study to determine the suitability of the fishing communities for future intervention studies, we recruited 1000 HIV-uninfected “at-risk” volunteers, aged 13 to 49 years, from a total of 2074 people screened after informed consent/assent across the 5 communities. Approximately half were selected from 2 Masaka District sites, and half from the other 3 sites. Before recruitment, each community was mapped, a census was taken, and identification numbers were assigned to all residents and regular visitors. All community members aged 13 to 49 years were given cards bearing their identification number to invite them for screening at designated study clinics established in the communities.

At screening and each follow-up visit, demographic and risk behavior data were collected through a structured questionnaire, and a blood sample provided for HIV testing, initially to rule out HIV-prevalent cases, and later to identify HIV-incident cases. HIV status was determined by a preset testing algorithm. All blood samples were tested on Determine HIV rapid test (ABBOTT Laboratories, Diagnostic division, Chicago, IL), and if positive, confirmed by 2 independent enzyme-linked immunosorbent assay (ELISA) tests: Vironostika (HIV Uni-Form II plus 0 microelisa system, Biomerieux, Boxtel, The Netherlands); and Murex HIV-1.2.0 (Murex, Biotech Limited, Dartford, United Kingdom), and by HIV-1/2 Western Blot (Calypte Biomedical Corporation, Rockville, MD) if enzyme-linked immunosorbent assays were indeterminate or discordant. All HIV-positive volunteers had CD4 counts done before referral for care.

A representative sample (by age and gender) of the screened population was maintained by frequently comparing screening and census data. Eligible volunteers were required to give their consent to participate (informed written consent was obtained from adults and assent in conjunction with parent/guardian consent for minors) and to be HIV-uninfected but “at-risk” of infection, defined as any one of the following: unprotected sex with more than one partner and/or new partners in the past 3 months, being away from home or having a partner who is away from home for at least 2 nights in the past month, reported or current sexually transmitted infection (STI) in the past 3 months, and reporting being in an HIV serodiscordant relationship.

Enrolled volunteers were followed up every 6 months for a total of 18 months. Syphilis testing was done at enrollment, at the 12-month visit, and at any visit when syphilis infection was suspected. Syphilis serology was done using Rapid Plasma Reagin (RPR; Biotec Laboratories Limited, Ipswich, Suffolk, United Kingdom) test, and if positive, confirmed by the Treponema Pallidum Haema-Agglutination assay (TPHA–BIOTEC Laboratories, United Kingdom). Those diagnosed with syphilis and other STIs were treated as recommended by the Ministry of Health guidelines in Uganda. Risk-reduction counseling and condoms were provided at each visit.

Detailed qualitative data, including life histories, were collected by experienced interviewers from 78 individuals. Interviewees were purposively chosen to represent different occupational groups, ages, and gender. Participants were selected soon after screening for enrollment to the cohort or for the selected incident cases, after the clinic visit when HIV was detected. Thirteen percent of incident cases were included in those selected.

The Uganda Virus Research Institute Science and Ethics Committee and Uganda National Council of Science and Technology provided ethical clearance for this study.

Back to Top | Article Outline
Sample Size Estimation

A random sample of 1000 participants was selected. This was selected with the assumptions that the 1000 participants would be followed up for HIV incidence for 18 months, with 10% dropout and an incidence rate of approximately 5/100 person-years at risk (PYAR). This incidence rate would be estimated with a precision of ±1.2.

Back to Top | Article Outline
Data Handling and Analysis

Data were entered using MS Access 2003 (1992–2003 Microsoft Corporation, Redmond, WA), and analyzed in Stata 11 (1996–2011 StataCorp, College Station, TX). Estimated date of HIV infection was defined as the midpoint between the last HIV-negative and the first HIV-positive result dates. PYAR were calculated as the time from enrollment to the date of the last HIV seronegative result, or to the estimated date of HIV infection for each participant.

To obtain age-specific HIV incidence rates and rate ratios, participants were grouped into 3 age groups (13–24 years, 25–29 years, and 30+ years) at the beginning of each observation period. Alcohol consumption was defined as regular if the participants reported consumption at least once a week, and rare if they drank once a month or less. Consistent condom use was defined as using condoms with every sexual encounter. Duration in the community was divided into 2 groups, with new immigrants defined as those who had been in the community for <5 years. Syphilis infection was defined by rapid plasma reagin test+ with 1:8+ titre and treponema pallidum haema-agglutination assay+.

Data for each participant were divided into observation periods corresponding to the observation time between 2 consecutive visits. Behavioral and risk factor variables obtained at the beginning and end of each observation period were included in the model to find out which ones were significantly associated with HIV incidence.

In the analysis, we used Cox proportional hazards model with time-varying covariates, allowing for intragroup correlation (because individuals had multiple records) by using robust standard errors. After bivariable analysis, multivariable model building was carried out using forward stepwise regression starting with the baseline socio-demographic variables before adding the behavioral variables and other potential risk factors. Age, sex, and factors remaining statistically significant at P < 0.1 were retained in the final multivariable model. Separate analyses by sex were not done because there was no significant interaction between age and sex in this population (P = 0.085).8

The qualitative data were analyzed using a hierarchical thematic framework. Four team members organized the data into key themes and concepts, continually comparing their coding to reach a consensus on the way data were coded. The resulting thematic charts were used to examine the data for patterns related to risk behavior.

Back to Top | Article Outline

RESULTS

Of the 2074 volunteers screened (20.4% of the eligible population), 1000 (48.2%) HIV seronegative at-risk volunteers were enrolled between February and August 2009. The median age of the 919 study volunteers included in the incidence analysis was 28 years (interquartile range: 23–34), 54.4% were male, 68.3% married, 77.5% Christians, and 47.7% had attained primary education. About 28% were involved in fishing, 34% in fishing-related activities, 9% were involved in small-scale businesses, 9% were farmers, and 6% had worked in bars in the fishing communities. Of those involved in fishing and fishing-related activities, 17% had other occupations, with the majority working either in bars (15%), small business (7%), or farming (6%). Of the 919 participants, 58.4% had been residing in their community for <5 years, whereas 53.8% reported being away from home regularly (for 2 or more nights in the month preceding the interview).

The study schema (Fig. 1) shows that 919 of 1000 (91.9%) attended 1 follow-up visit, 821 of 890 (92.2%) attended 2 follow-up visits, and 714 of 802 (89.0%) attended 3 follow-up visits. Overall, 762 (76.2%) reached the study end point (either seroconverted or completed the 3 follow-up visits). Follow-up was independently associated with sex, age, duration in the community, ethnicity, and study site (Table 1). Volunteers less likely to return for follow-up were young (aged 13–24 years), male, of minority (non-Ganda) ethnicity, recruited from Wakiso/Mukono (Entebbe site), or new migrants (<5 years in the community).

Figure 1
Figure 1
Image Tools
Table 1
Table 1
Image Tools

At enrollment, 57.3% of participants reported having only 1 sexual partner, 38.0% reported 2 or more sexual partners, and 4.2% reported no sexual partner in the past 3 months (0.4% responded “don't know” [not shown]). Of the 41.2% who reported having at least 1 new sexual partner in the 3 months preceding the interview, only 26.1% reported consistent condom use, whereas 20.4% and 53.5% reported inconsistent and no-condom use with the new partner, respectively. Women were more likely to receive gifts for sex than men (42% vs. 18%), whereas men were more likely to give gifts (50% vs. 13%). Although 3% of the participants knew their partners were HIV-infected, only 31% consistently used condoms with these HIV-infected partners, and yet 46% knew whether their partner was on anti-retroviral therapy.

Back to Top | Article Outline
HIV Incidence and Risk Factors for HIV Acquisition

During 18 months of follow-up, 59 participants seroconverted in 1205.6 PYAR, an HIV incidence rate of 4.9 [95% CI = 3.8–6.3]) per 100 PYAR, with 5.2/100 [3.7–7.3] and 4.5/100 [3.1–6.7] in men and women, respectively. High HIV incidence was observed among cigarette smokers (12.5/100 [4.7–33.4]), those with HIV-infected partners (12.1/100 [5.0–29.0]), marijuana users (11.1/100 [4.6–26.7]), those working in bars (9.8/100 PYAR [4.6–20.6]), Protestants (8.6/100 [5.8–12.7]), those with genital discharge/sores (8.3/100 [5.8–11.9]), regular alcohol consumers (8.3/100 [5.7–12.1]), those aged 13 to 24 years (7.5/100 [5.2–11.0]), those who had got new partners recently (7.5/100 [4.9–11.3]), those who had sex under the influence of alcohol (7.5/100 [4.9–11.5]), those with multiple sexual partners (7.2/100 [4.7–11.0]), and new immigrants (6.6/100 [4.9–8.8]).

Socio-demographic factors independently associated with risk of HIV acquisition in a multivariable analysis (Table 2) were being younger than 25 years (adjusted hazard ratio (aHR) = 2.5 [1.3–4.9]), religion (compared with Catholics, Protestants had aHR = 2.7 [1.4, 5.3], and Muslims had aHR = 2.3 [1.1, 4.8]), and spending <5 years in the fishing community (HR = 2.1 [1.1–4.9]).

Table 2
Table 2
Image Tools

Socio-demographic factors investigated (data not shown), which were not associated with HIV acquisition, included sex, marital status, education level, ethnicity, and financial dependency. Although occupation was not associated with HIV infection, bar workers were more likely to acquire HIV than those not working there (HR = 4.7 [0.9, 20.6], P = 0.07).

Behavioral factors and STI symptoms independently associated with HIV acquisition (Table 2) were as follows: having either genital discharge or sores recently (aHR = 2.8 [1.6, 5.0]), drinking alcohol regularly (aHR = 3.1 [1.6, 6.1]), smoking marijuana (aHR = 2.9 [1.0, 8.0]), and smoking cigarettes (aHR = 3.5 [1.4, 8.6]).

There was an association between alcohol consumption and sexual behavior. Table 3 shows that as the frequency of alcohol use increased, the percentage of individuals having sex while under the influence of alcohol also increased from 6.1% for those who had never drunk previously to 65.1% for the regular drinkers (trend P value <0.001). Similarly, the percentage of volunteers with 2 or more sex partners or with at least 1 new sex partner also increased as the frequency of alcohol consumption increased.

Table 3
Table 3
Image Tools

Factors such as the number of sexual partners in the previous 3 months, having sex under the influence of alcohol, knowledge of sexual partners with HIV infection, and being away from home frequently were not significantly associated with HIV infection in the multivariable model, but individually showed an association with HIV acquisition when adjusted for age and sex (Table 4). This was not explained by collinearity because multicollinearity was assessed and the variables were found not to be collinear with the other variables in the model. Participants who had sex under the influence of alcohol (aHR = 2.1 [1.2–3.6]), or those with more than one sexual partner (aHR = 3.0 [0.8–10.0], P = 0.1), or those who knew their partners were HIV-infected (aHR = 3.1 [1.3–7.5]), or those who stayed frequently away from home (aHR = 1.8 [1.0–3.1]) were more likely to seroconvert.

Table 4
Table 4
Image Tools

There was weak evidence in the crude analysis that reporting a new sexual partner at a previous (HR = 1.6 [1.0–2.8]) or current visit (HR = 1.9 [1.1–3.3]) was associated with HIV infection. Syphilis infection and frequency of condom use with new partners were not associated with HIV acquisition (Table 4).

Data from the in-depth interviews provide some insights into the quantitative findings. All respondents either reported multiple sexual partners (with a few men referring to their casual partners as being “too many to count”) or of their partner having many partners. Ten of the 78 people interviewed had recently seroconverted. They all assumed that their, or a partner's, having multiple partners was the reason for their infection.

Frequent changes in occupation were reported, with fishing, fish-related tasks, and bar/restaurant work being interchanged with, for example, farming and small business, depending on the different opportunities for work that were available. Marital instability and the search for lucrative occupations were often given as reasons for moving between fish-landing sites or to and from inland areas. Several women reported that finding a partner to support them was important to sustain them financially.

Back to Top | Article Outline

DISCUSSION

Prevalence among those screened for this study, which included all occupational groups (not only fishermen), including men and women, was 29%, and overall incidence was 4.9 per 100 PYAR-substantiating concerns raised in our previous publications that these communities were at high risk for HIV infection.3,9,10 In addition, the higher risk of HIV infection among young people (aged, 13–24 years) differs from the recently observed trend in the general population in Uganda where higher rates of infection are reported among older and married people.1

Sex workers, injection drug users, and men who have sex with men continually receive special attention in UNAIDS global reports as groups at high risk of infection.1,11 These groups are often referred to as being communities, even if as individuals they are scattered in the general population. Our findings show that there is a pocket or “community” of the general population at high risk of infection, a “community” that is geographically well defined. We would suggest that there are distinct patterns of behavior, such as partner change and mobility, which promote HIV transmission and ensure that these sites serve as a reservoir of HIV and other STI.

Recent commentators have observed that an urban culture characterizing prominent landing sites (availability of alcohol, leisure activities such as pool tables, and the prevalence of non-kin ties) continues to attract young people to fishing sites, despite growing uncertainty over the economic viability of fisheries.12,13 However, despite falling incomes, gendered inequities remain, with fishermen's relative wealth fuelling transactional relationships with women. The impact of this social context is shown in our findings, with higher rates of HIV infection or acquisition among regular drinkers of alcohol, probably as a result of sexual disinhibition, an association corroborated by other studies in Africa.1417 Further qualitative data analysis is being undertaken to investigate the surprising finding that Muslims had higher rates of HIV incidence than Catholics. This could be partially explained by factors other than religion that put this group at risk (for example sexual mixing and recent immigration). Unfortunately, we did not confirm whether all Muslim men were circumcised.

Syphilis infection was lower among those enrolled for the study compared with the general population (2.4% vs. 11%), and was lower than genital discharge or genital sores (56.9%), with only 1.4% of participants getting newly infected with syphilis during follow-up. This could partly explain why syphilis was not significantly associated with HIV acquisition. Another reason could be that syphilis is not an ulcerative STI, except in its early stages.

The strength of this study is the use of census data to select a fairly representative set of individuals by age, sex, and occupation, collecting both behavioral and biologic variables in the same setting, and good participation rates giving us the opportunity to examine incidence and risk factors in this context. The socio-behavioral context of the study provided supportive evidence on the vulnerability of new migrants and young people to infections in the fishing communities.

However, the study had at least 2 limitations. We followed a group preselected to be at higher risk. This is likely to have led to overestimation of incidence compared with all residents in these communities. Nevertheless, even if we assume all those individuals who were ineligible because there were “not at-risk” did not seroconvert during the period of the study, HIV incidence would still be of the order of 3.3/100 (59/1790) PYAR, a much higher risk than the general population in Uganda.1 The study was further limited by the lack of laboratory data for other STIs because of logistical and financial challenges.

Our fears that such populations are mobile, difficult to recruit, and follow in longitudinal studies were unfounded, as we managed to achieve an annual follow-up >80%. Many people were often only away for short periods for fishing or trading, as they considered the community their residence. Sustaining people's interest in interventions and/or research can, as we have found, be achieved by adapting working hours for research teams (including staying overnight at the sites) and promoting community engagement through meetings and innovative activities like sports events (football and boat races). These activities have proved effective in reaching many who might otherwise be considered beyond reach.

Although there are some exceptions,18 many fishing sites on Lake Victoria, particularly on islands, have long been neglected in the provision of basic health services and HIV care and education. Although the water acts as a barrier to some service providers who face challenges in finding safe transportation, service provision is beginning to improve. However, tackling the many public health issues in these communities requires concerted effort. Prevention strategies need to take into account the social contact of individual sexual decision making in these communities; solely focusing on individual-level HIV prevention will fail to address the many influences on behavior in these sites.

UNAIDS in their strategy document “Getting to Zero” observed the following: Young women and men in fishing communities on the shores of Lake Victoria, where we know HIV incidence is high, are among these most vulnerable and neglected groups. Given that the number of people involved in fishing-related activities in Uganda is estimated to be in excess of 1 million,19 many people are at risk of HIV infection. There is, therefore, an urgent need to target HIV-prevention efforts of what we know works, as well as increase HIV awareness and scale up health facilities.

The HIV response gives us an opportunity to strengthen the social fabric, improve social justice and reinforce the systems that deliver critical services for the most vulnerable members of our communities. We must achieve a balance between intensifying work in the hardest-hit countries and identifying other settings, such as cities, where the impact of HIV is affecting specific communities—particularly men who have sex with men, sex workers and their clients and people who use drugs.' (2010: 10)

Addressing the high levels of HIV and other STI must be a priority for not only service providers but also for research. More targeted health education needs to be implemented and evaluated; circumcision could reduce transmission to men, and access to an effective microbicide could transform the lives of women. Access to treatment as prevention and an effective vaccine could potentially make a difference for everyone.20 We need to conduct research to evaluate new approaches, including large-scale efficacy trials, which, from our study findings of high incidence and good retention, seem feasible to conduct in these communities.

Back to Top | Article Outline

REFERENCES

1. UNAIDS. UNAIDS Report on the Global AIDS Epidemic 2010. Geneva, Switzerland: UNAIDS, 2010.

2. Vandepitte J, Bukenya J, Weiss HA, et al.. HIV and other sexually transmitted infections in a cohort of women involved in high-risk sexual behavior in Kampala, Uganda. Sex Transm Dis 2011; 38:316–323.

3. Kissling E, Allison EH, Seeley JA, et al.. Fisherfolk are among groups most at risk of HIV: Cross-country analysis of prevalence and numbers infected. AIDS 2005; 19:1939–1946.

4. Kwena ZA, Cohen CR, Sang NM, et al.. Fishermen as a suitable population for HIV intervention trials. AIDS Res Treat. Epub ahead of print July 18, 2010.

5. Asiki G, Mpendo J, Abaasa A, et al.. HIV and syphilis prevalence and associated risk factors among fishing communities of Lake Victoria, Uganda. Sex Transm Infect 2011; 87:511–515.

6. Uganda AIDS Commission. The Modes of Tranmission Study. The Uganda Country Synthesis Report. Kampala, Uganda, 2008.

7. Vandenbroucke JP, von Elm E, Altman DG, et al.. STROBE Initiative. Strengthening the reporting of observational studies in epidemiology (STROBE): Explanation and elaboration. PLoS Med 2007; 4:e297.

8. Uganda AIDS Commission. The Uganda HIV/AIDS status report July 2004-December 2005. Kampala: Uganda AIDS Commission, Government of the Republic of Uganda, 2006.

9. Allison EH, Seeley JA. HIV and AIDS among fisherfolk: A threat to ‘responsible fisheries’? Fish Fisheries 2004; 5:215–234.

10. Seeley JA, Allison EH. HIV/AIDS in fishing communities: Challenges to delivering antiretroviral therapy to vulnerable groups. AIDS care 2005; 17:688–697.

11. UNAIDS. Getting to zero. 2011–2015 strategy Joint United Nations Programme on HIV/AIDS (UNAIDS). Geneva, Switzerland, 2010.

12. Beuving JJ. Playing pool along the shores of Lake Victoria: Fishermen, careers and capital accumulation in the Ugandan Nile perch business. JIAI 2010; 80:224–248.

13. Mojola SA. Fishing in dangerous waters: Ecology, gender and economy in HIV risk. Soc Sci Med 2011; 72:149–156.

14. Mbulaiteye S, Ruberantwari A, Nakiyingi J, et al.. Alcohol and HIV: A study among sexually active adults in rural southwest Uganda. Int J Epidemiol 2000; 29:911–915.

15. Geis S, Maboko L, Saathoff E, et al.. Risk factors for HIV-1 infection in a longitudinal, prospective cohort of adults from the Mbeya Region, Tanzania. J Acquir Immune Defic Syndr 2011; 56:453–459.

16. Mavedzenge SN, Weiss HA, Montgomery ET, et al.. Determinants of differential HIV incidence among women in three southern African locations. J Acquir Immune Defic Syn 2011; 58:89–99.

17. Leigh BC, Temple MT, Trocki KF. The relationship of alcohol use to sexual activity in a US national sample. Soc Sci Med 1994; 39:1527–1535.

18. Parker M, Allen T. Does mass drug administration for the integrated treatment of neglected tropical diseases really work? Assessing evidence for the control of schistosomiasis and soil-transmitted helminths in Uganda. Health Res Policy Syst 2011; 9:3.

19. Fulgencio K. Globalisation of the Nile perch: Assessing the socio-cultural implications of the Lake Victoria fishery in Uganda. Afr J Polit Sci Int Relat 2009; 3:433–442.

20. Granich R, Crowley S, Vitoria M, et al.. Highly active antiretroviral treatment as prevention of HIV transmission: Review of scientific evidence and update. Curr Opin HIV AIDS 2010; 5:298–304.

© Copyright 2012 American Sexually Transmitted Diseases Association

Login