Skip to main content

Table 1 Overview and characteristics of included studies

From: A systematic review of the clinical and social epidemiological research among sex workers in Uganda

Source

Setting

Design

Objective

Sample size and characteristics

Results

Bukenya et al., 2013

Kampala

Longitudinal cohort: cross-sectional baseline analysis

To describe the prevalence and determinants of inconsistent condom use

N = 905 sexually active Ugandan female sex workers

40.2 % inconsistent condom use with paying clients in the last month. Increased risk: sex work not the sole source of income, sexual debut before 14 years, daily consumption of alcohol, fewer paying clients in last month, and currently pregnant. Decreased risk: currently married, higher number of sexual partners.

Erickson et al., 2015

Gulu, northern Uganda

Cross-sectional survey

To describe the prevalence and correlates of dual contraceptive use

N = 400 Ugandan female sex workers

45.0 % had ever used dual contraceptives.

Increased odds of dual contraceptive use: older age, prior unintended pregnancy, HIV testing. Decreased odds of dual contraceptive use: rushing client negotiations because of police presence.

Francis et al., 2013

Multisite – Uganda (Kampala), Tanzania

Longitudinal cohort: cross-sectional analysis

To describe intervaginal cleansing among sex workers

N = 200 female sex workers, n = 100 Ugandan female sex workers

Among Ugandan sex workers: 81.8 % consistent condom use in past 3 months; 52.0 % HIV positive.

Among Ugandan sex workers 100 % reported intravaginal cleansing ever; 4.5 cleansing acts per day; 80.3 % participants reported cleansing after half of their total sex acts. The frequency of cleansing was significant higher following sex, menstruation or vaginal discomfort.

Guédou et al., 2012

Multisite: Uganda (Kampala), South Africa, Benin, India

Double blind RCT: cross-sectional secondary analysis

To examine the association between prevalent intermediate vaginal flora (IVF), bacterial vaginosis (BV) and HIV infection among all sex workers screened for the RCT

N = 1367 female sex workers, n = 516 Ugandan female sex workers

Among total sample, 27.0 % HIV prevalence, 47.6 % BV prevalence, and 19.2 % IVF. BV and IVF were significantly associated with HIV.

Among Ugandan sex workers, HIV prevalence 32.4 %, additional stratified analyses not available.

Guédou et al., 2013

Multisite: Uganda (Kampala), South Africa, Benin, India

Double blind RCT: longitudinal secondary analysis

To examine predictors of recurrent BV

N = 440 female sex workers with >1 episode of BV, n = 167 from Uganda

Among total sample, BV incident rate of 20.8 recurrences/100-person-months; Risk factors: vaginal cleansing increased risk; consistent condom use and vaginal candidiasis decreased risk.

Among Ugandan sex workers, 7.9 recurrences/100 person-months. No additional stratified analyses available.

Matovu et al., 2012

Kampala

Cross-sectional survey

To assess sexual risk behaviours, condom use and STI infection among sex workers

N = 259 Ugandan female sex workers

55.1 % used condoms inconsistently in past month; 77.2 % self-reported STI in past 12 months; 86 % sought treatment 3 days after recognition of symptoms; consistent condom use was 72.1 % with causal partners, 40.8 % with regular partners, 6.3 % with spouses.

Morris et al., 2006

Multisite - Uganda/Kenya on the Mobassa-Kampala highway

Longitudinal study: diaries of sexual activity for 30 days

To exploring the effect of condom use among sex worker on the trans-Africa highway in contributing to HIV epidemic

N = 578 Ugandan/Kenyan female sex workers, n = 175 Ugandan

Total of 14072 sex acts, 77.7 % of sex acts used condoms; Modelling – using HIV prevalence of 30.0–50.0 % it was estimated there are 3200–4148 new HIV infections per year on the Mombasa-Kampala highway.

Morris et al., 2009

Multisite - Uganda/Kenya on the Mobassa-Kampala highway

Longitudinal study: diaries of sexual activity for 30 days

To describe sexual behaviour among sex workers on the Mombasa-Kampala highway, compare risk between Ugandan and Kenyan sex workers

N = 578 Ugandan/Kenyan female sex workers, n = 175 Ugandan

Compared to Ugandan sex workers, Kenyan sex workers had higher consistent condom use (79.2 % vs 73.9 %), more likely to use condom during sex act, higher condom use with regular clients.

Compared to Ugandan bars, bars in Kenya were more likely to: have condom dispensers, (25.0 % vs 1.0 %); distribute or sell condoms, (73.9 % vs 47.6 %); and have more weekly condom distribution.

Muldoon et al., 2014

Gulu, northern Uganda

Cross-sectional survey

To examine the proportion of sex workers with a history of LRA abduction, access to post-abduction reintegration services and relative mental health

N = 400 Ugandan female sex workers, n = 129 with history of abduction

From a sample of 400 sex workers, 32.3 % had been abducted, 43.4 % had accessed a reintegration program. Mental health status was not significantly different between those who did and did not access a reintegration program. HIV prevalence 41.1 %.

Muldoon et al., 2015

Gulu, northern Uganda

Cross-sectional survey

To describe the prevalence and correlates of client violence, assess relationship between policing and client violence

N = 400 Uandan female sex workers

Most common forms of client violence: physical assault (58.7 %), rape (38.3 %), gang rape (15.8 %). HIV prevalence was 33.8 %. Inconsistent condom use was 84.0 %.

Increased odds of client violence: rushing client negotiations because of police presence, servicing clients in a bar, inconsistent condom use with any client, working for a manager/pimp.

Pickering et al., 1997a

Fishing village in south-western Uganda

Longitudinal study: diaries of sexual activity for 6 months

To describe sexual mixing patterns inside and outside town

N = 26 Ugandan female sex workers

Women contributed 421 women-weeks; 15 were married and 42.0 % of sex partner were with commercial partner; 11 were single and 20.0 % of sex acts were with non-commercial partners; 90.0 % of contacts were from men resident in the village.

Pickering et al., 1997b

Trading town in south-western Uganda

Longitudinal study: diaries of sexual activity for 6 months

To describe sexual mixing patterns

N = 48 Ugandan female sex workers

Women contributed 789 women-weeks; average 5.8 clients per week; 10.0 % of clients were non-commercial; condom use was 99.0 % with commercial partners.

Pickering et al., 1997c

Trading town and fishing village in south-western Uganda

Longitudinal study: diaries of sexual activity for 6 months

To describe sexual mixing patterns

N = 81 Ugandan female sex workers

Women contributed 1280 women-weeks; 34 women from fishing villages and rural areas 90.0 % of sex acts with local men; 47 women from town contacts 87.0 % of sex partners were with with truck drivers or outside clients; 52.0 % were HIV positive, no significant difference by location.

Redd et al., 2014

Kampala

Longitudinal cohort: longitudinal clinical analysis

To determine the rates of HIV primary and super-infection among sex workers in Kampala

N = 85 HIV positive Ugandan female sex workers

The prevalence of HIV superinfection was 8.2 % (3.4/100 person-years) and was not significantly different from the rate of primary infection in the same population (3.7/100 person-years).

Schwitters et al., 2015

Kampala

Cross-sectional survey

To estimate the prevalence of client initiated violence in the previous 6 months among sex workers

N = 1467 Ugandan female sex workers

81.8 % had experienced at least one form of client-initiated violence in previous 6 months: 39.1 % physical abuse, 44.7 % verbal abuse, 49.1 % forced sex, 54.9 % not paid.

Increased odds of violence: longer duration in sex work, more frequent client demand for unprotected sex, consumption of 5+ alcoholic drink, soliciting in outdoor spaces (e.g. streets, parks, parking lots etc.).

Ssemwanga et al., 2012a

Kampala

Longitudinal cohort: longitudinal clinical analysis

To identify prevalence of multiple HIV infections and associated features of partnership histories

N = 324 HIV-positive Ugandan female sex workers

9.0 % had multiple infections, sex workers working in same localities had phylogenetically similar viruses.

Ssemwanga et al., 2012b

Kampala

Longitudinal cohort: longitudinal clinical analysis

To classify HIV drug resistance among ART naïve women with new HIV diagnosis

N = 42 ART naïve Ugandan female sex workers with new HIV diagnosis

HIV drug resistance point prevalence estimate of 2.6 % (95 % confidence interval, 0.1 %–13.8 %).

Van Damme et al., 2008

Multisite: Uganda (Kampala), South Africa, Benin, India

Double blind RCT: primary analysis

To investigate efficacy of cellulose sulphate microbicide gel to reduce new HIV infection

N = 1398 HIV-negative female sex workers total; N = 303 Ugandan

Cellulose sulphate gel did not prevent HIV infection and may have increased the risk of HIV acquisition, hazard ratio 1.61 (0.9–3.0). Within Ugandan sub-group, sex workers reported 17–19 (med) sex partners, 19–21 (med) sex acts in previous 7 days. 97.5 % condom use per sex act. Additional stratified analyses not available.

Vandepitte et al., 2011

Kampala

Longitudinal cohort: cross-sectional baseline analysis

To examine baseline prevalence and risk factors of HIV and STIs

N = 1027 Ugandan female sex workers

HIV prevalence 37.0 %, gonorrhoea 13.0 %, chlamydia 8.9%, T. Vaginalis 17.1 %, BV 55.0 %, candida infection 11 %, HSV-2 antibodies 79.9%, active syphilis 10.0 %.

Increased HIV risk: older age, widowed, lack of education, sex work as sole income, street-based sex work, not knowing HIV status, using alcohol and intravaginal cleansing with soap.

Vandepitte et al., 2012a

Kampala

Longitudinal cohort: cross-sectional baseline analysis

To assess the prevalence and determinants of mycoplasma genitalium (MG) among sex workers

N = 1025 endocervical swabs from Ugandan female sex workers

MG prevalence: 14.0 % - more prevalent in HIV+; less prevalent in older women, those who were pregnant but never gave birth. Associated with gonorrhoeae, candida, trichomonas vaginalis.

Vandepitte et al., 2012b

Kampala

Longitudinal cohort: cross-sectional baseline clinical analysis

To describe the symptoms and signs associated with MG among Ugandan sex workers

N = 1027 Ugandan female sex workers

MG prevalence 14.0 %, increased risk: dysuria and mucopurulent vaginal discharge.

Vandepitte et al., 2013

Kampala

Longitudinal cohort: longitudinal clinical analysis

To investigating the patterns of clearance and recurrence of untreated MG

N = 119 Ugandan female sex workers with MG

Overall clearance rate 25.7/100 person years; 55.0 % spontaneously cleared infection within 3 months, 83.0 % within 6 months, 93.0 % within 12 months. Infection recurred in 39.0 % of women.

Vandepitte et al., 2014a

Kampala

Longitudinal cohort: nested case control

To examining the temporal association between MG status prior to HIV infection

N = 168 Ugandan female sex workers, n = 42 cases, n = 126 controls

42 sex workers acquired HIV during the study, incident rate of 3.6/100 person years; Non-significant association between MG infection and HIV acquisition.

Vandepitte et al., 2014b

Kampala

Longitudinal cohort: longitudinal clinical analysis

To assess the prevalence and antimicrobial susceptibility patterns of gonorrhoea among sex workers in Kampala

N = 148 Ugandan female sex workers with diagnosis of gonorrhoea

83.1 % ciproflaxin resistance, 68.2 % penicillin resistance. 97.3 % tetracycline resistance.