HIV incidence among non-migrating persons following a household migration event: a population-based, longitudinal study in Uganda.

Background: The impact of migration on HIV risk among non-migrating household members is poorly understood. We measured HIV incidence among non-migrants living in households with and without migrants in Uganda. Methods: We used four survey rounds of data collected from July 2011-May 2018 from non-migrant participants aged 15-49 years in the Rakai Community Cohort Study, an open, population-based cohort. Non-migrants were individuals with no evidence of migration between surveys or at the prior survey. The primary exposure, household migration, was assessed using census data and defined as ≥1 household member migrating in or out of the house from another community between surveys (∼18 months). Incident HIV cases tested positive following a negative result at the preceding visit. Incidence rate ratios (IRR) with 95% confidence intervals were estimated using Poisson regression with generalized estimating equations and robust standard errors. Analyses were stratified by gender, migration into or out of the household, and the relationship between non-migrants and migrants (i.e., any household migration, spouse, child). Findings: Overall, 11,318 non-migrants (5,674 women) were followed for 37,320 person-years. 28% (6,059/21,370) of non-migrant person-visits had recent migration into or out of the household, and 240 HIV incident cases were identified in non-migrating household members. Overall, non-migrants in migrant households were not at greater risk of acquiring HIV. However, HIV incidence among men was significantly higher when the spouse had recently migrated in (adjIRR:2·12;95%CI:1·05-4·27) or out (adjIRR:4·01;95%CI:2·16-7·44) compared to men with no spousal migration. Women with in- and out-migrant spouses also had higher HIV incidence, but results were not statistically significant. Interpretation: HIV incidence is higher among non-migrating persons with migrant spouses, especially men. Targeted HIV testing and prevention interventions such as pre-exposure prophylaxis could be considered for those with migrant spouses. Funding: National Institutes of Health, US Centers for Disease Control and Prevention. Research in context: We searched PubMed for studies focused on HIV acquisition, prevalence or sexual behaviors among non-migrants who lived with migrants in sub-Saharan Africa (SSA) using search terms such as "HIV", "Emigration and Immigration", "family", "spouses", "household", "parents", and "children". Despite high levels of migration and an established association with HIV risk in SSA, there is limited data on the broader societal impacts of migration on HIV acquisition risk among non-migrant populations directly impacted by it.There has been only one published study that has previously evaluated impact of migration on HIV incidence among non-migrating persons in sub-Saharan Africa. This study, which exclusively assessed spousal migration, was conducted in Tanzania more than two decades earlier prior to HIV treatment availability and found that non-migrant men with long-term mobile partners were more than four times as likely to acquire HIV compared to men who had partners that were residents. To the best of our knowledge, this is the first study to examine the effect of non-spousal migration, including any household migration and child migration, on HIV incidence among non-migrants. Added value of this study: In this study, we used data from the Rakai Community Cohort Study (RCCS), a population-based HIV surveillance cohort to measure the impact of migration on HIV incidence for non-migrant household members. The RCCS captures HIV incident events through regular, repeat HIV testing of participants and migration events through household censuses. Our study adds to the current literature by examining the general effect of migration in the household on HIV incidence in addition to child, and spousal migration. Using data from over 11,000 non-migrant individuals, we found that spousal, but not other types of household migration, substantially increased HIV risk among non-migrants, especially among men. Taken together, our results suggest that spousal migration may be associated with an increased risk of HIV acquisition in the period surrounding and immediately after spousal migration. Implications of all the available evidence: Our findings suggest that spousal migration in or out of the household is associated with greater HIV incidence. Targeted HIV testing and prevention interventions such as pre-exposure prophylaxis could be considered for men with migrant spouses.

Evaluation of multiplex real-time PCR for detection of Haemophilus ducreyi, Treponema pallidum, herpes simplex virus type 1 and 2 in the diagnosis of genital ulcer disease in the Rakai District, Uganda.

OBJECTIVE: To develop a real-time PCR assay that reliably and accurately detects the predominant sexually transmitted aetiological agents of genital ulcer disease (GUD) (Haemophilus ducreyi, Treponema pallidum and herpes simplex virus type 1 (HSV-1) and type 2 (HSV-2)) and to assess the use of real-time PCR diagnostic testing in a rural African field site. METHODS: Two multiplex real-time PCR reactions were used to detect H ducreyi/and HSV-1/HSV-2 in ulcer swabs from 100 people with symptomatic genital ulcers in rural Rakai, Uganda. Results were compared with syphilis, HSV-1 and HSV-2 serology. RESULTS: Of 100 GUD samples analysed from 43 HIV positive and 57 HIV negative individuals, 71% were positive for one or more sexually transmitted infection (STI) pathogens by real-time PCR (61% for HSV-2, 5% for T pallidum, 3% for HSV-1, 1% for H ducreyi and 1% for dual H ducreyi/HSV-2). The frequency of HSV in genital ulcers was 56% (32/57) in HIV negative individuals and 77% (33/43) in HIV positive individuals (p = 0.037). Assay reproducibility was evaluated by repeat PCR testing in the USA with 96% agreement (kappa = 0.85). CONCLUSIONS: STI pathogens were detected in the majority of GUD swab samples from symptomatic patients in Rakai, Uganda, by real-time PCR. HSV-2 was the predominant cause of genital ulcers. Real-time PCR technology can provide sensitive, rapid and reproducible evaluation of GUD aetiology in a resource-limited setting.

Aetiology of genital ulcer disease in female partners of male participants in a circumcision trial in Uganda.

HIV acquisition is associated with herpes simplex virus type 2 (HSV-2) infection and genital ulcer disease (GUD). Three randomized control trials demonstrated that male circumcision significantly decreases HIV, HSV-2, human papillomavirus and self-reported GUD among men. GUD is also decreased among female partners of circumcised men, but it is unknown whether male circumcision status affects GUD pathogens in female partners. For the evaluation of GUD aetiology, two separate multiplex assays were performed to detect Haemophilus ducreyi, Treponema pallidum, HSV-1 and HSV-2. Of all the female GUD swabs evaluated, 67.5% had an aetiology identified, and HSV-2 was the primary pathogen detected (96.3%). However, there was no difference in the proportion of ulcers due to HSV-2 or other pathogens between female partners of circumcised men (11/15, 73.3%) compared with uncircumcised men (15/25, 60.0%, P = 0.39). The seroprevalence of HSV-2 is high in this population and therefore most of the detected HSV-2 infections represent reactivation. Since GUD is associated with HIV acquisition and one-third of GUD in this study did not have an aetiological agent identified, further research is needed to better understand the aetiology of GUD in Africa, and its relationship to circumcision and HIV infection.

Prevalence and risk factors for carcinogenic human papillomavirus infections in rural Rakai, Uganda.

OBJECTIVE: To investigate self-administered vaginal swabs for assessing prevalence and correlates of carcinogenic human papillomavirus (HPV) infection in rural Rakai, Uganda. METHODS: 1003 sexually experienced women enrolled in a community cohort provided self-administered vaginal swabs collected at annual, home-based surveys. Carcinogenic HPV prevalence, adjusted odds ratios (AOR), 95% confidence intervals (CI) and associated risk factors were determined. RESULTS: Carcinogenic HPV prevalence was 19.2%: 46.6% among HIV positive and 14.8% among HIV negative women (p<0.001). Type-specific prevalence ranged from 2.0% (HPV 16 and 52) to 0.2% (HPV 31). Age-specific HPV prevalence decreased significantly (p<0.001) among HIV negative women; however, the decrease among HIV positive women was not as pronounced (p = 0.1). Factors independently associated with carcinogenic HPV infection were HIV (AOR 4.82, CI 3.10 to 7.53), age (AOR 4.97, 95% CI 2.19 to 11.26 for 15-19 year olds compared to 40+ years), more than two sex partners in the past year (AOR 2.21, CI 1.10 to 4.43) and self-reported herpes zoster, candidiasis or tuberculosis (AOR 4.52, CI 1.01 to 20.31). Married women were less likely to have prevalent carcinogenic HPV (AOR 0.46, CI 0.30 to 0.70). CONCLUSIONS: HPV prevalence and correlates measured using self-administered vaginal swabs were similar to studies that use cervical samples. Thus, self-collection can be used as a substitute for cervical specimens and provide an important tool for research in populations unwilling to undergo pelvic exam.

The relational determinants of condom use with commercial sex partners in Thailand.

OBJECTIVE: To analyze the extent and determinants of condom use with commercial sex partners among lower socioeconomic status groups in the Thai population. DESIGN: Respondents were sampled in Udon Thani, Saraburi and Bangkok in 1992. Completed sample size was 678 women in brothels, 330 male truck drivers and 1,075 men aged 17-45 years. Behavioral data and local sexual network information were collected using structured questionnaires (face-to-face interviews), focus groups and in-depth unstructured interviews. METHODS: Data were analyzed using univariate and multivariate logistic regression. RESULTS: Condom use with commercial partners remains inconsistent. Consistent use was reported by 61% of women in brothels, 25% of truck drivers, and 29% of men in the low-income population. The single strongest predictor of consistent condom use for all groups is type of partnership. Consistent use drops significantly with regular (multivisit) commercial sex partners compared with casual (single visit) commercial partners; adjusted odds of consistent use are 0.22 for women and 0.25 for men. Brothel women report that one in five of their commercial partners is a 'regular', and 20% of the young men who report a commercial partner report a 'regular'. DISCUSSION: The strongest determinant of consistent condom use is the nature of the relational bond between the partners, rather than their individual characteristics, knowledge or attitudes. To raise condom use further, programs will have to move beyond the standard knowledge-attitudes-practices paradigm focus on individual attributes to address the contextual determinants of behavior.

PIP: 678 women in brothels, 330 male truck drivers, and 1075 men aged 17-45 years participated in focus groups and were interviewed in Udon Thani, Saraburi, and Bangkok in 1992 to analyze the extent and determinants of condom use with commercial sex partners among lower socioeconomic status groups in the Thai population. Consistent condom use was reported by 61% of women in brothels, 25% of truck drivers, and 29% of men in the low-income population. The single strongest predictor of consistent condom use for all groups is the nature of the relational bond between the partners, rather than their individual characteristics, knowledge or attitudes. Consistent use drops significantly with multivisit commercial sex partners compared with single visit commercial partners. Brothel women report that 20% of their commercial partners are regulars, while 20% of the young men who report a commercial partner also report visiting a "regular" commercial sex partner. To raise condom use further, programs will have to move beyond the standard knowledge, attitudes, and practices approach to address the contextual determinants of behavior.

Bridge populations in the spread of HIV/AIDS in Thailand.

OBJECTIVE: To determine the extent to which men provide a bridge population between commercial sex workers (CSW) and the general female population in Thailand. DESIGN: Sexual network and serological data were collected from a systematic quota sample of low income men and truckers during 1992 in three Thailand provinces. Completed sample size was 1075 men aged 17-45 years and 330 truckers. METHODS: Sexual network information was used to identify those men who have sex with both female CSW and non-CSW partners (the 'bridge population'). A new method was used for calculating the partner acquisition rate and to establish the potential number of women exposed to HIV via inconsistent condom use among the bridge population. RESULTS: Approximately 17% of men and 25% of truckers can be included in the bridge population. These men are more likely to be HIV-positive and to have had at least one other sexually transmitted infection in the past year (odds ratio, 2.2 and 3.4, respectively). Consistent condom use with CSW is less than 30%, and is less than 1% with non-CSW partners. As a result, 30 women in the general population were potentially exposed to HIV per 100 sexually active men in the last year:nine women each additional year. Younger men and truckers expose almost twice as many women to HIV; more female peers than wives are exposed. CONCLUSION: Bridge populations may be as important as "core groups' for the spread of HIV into the general Thai population. Young men and women are strategic intervention targets because they have more partners, are more likely to be in bridging networks, and are more receptive to condom use.

Condom acceptance is higher among travelers in Uganda.

OBJECTIVE: To examine the role played by mobile residents in the spread of HIV through rural Uganda. DESIGN: Travel history and sexual network data were collected from a random sample of 1627 residents aged 15-49 years in Rakai District, Uganda during 1994. METHODS: Travelers and non-travelers are compared with respect to socio-demographic attributes, risk exposure, knowledge, attitudes and use of condoms using descriptive statistics and multivariate logistic regression. A demographic profile of travelers' partners is developed using information from a local network survey module. RESULTS: The population is highly mobile, with over 70% reporting travel to a potentially higher risk destination in the past year. Travelers are somewhat more likely to have higher levels of sexual risk behavior, but the risk appears to be offset by significantly greater knowledge, acceptance, and use of condoms. In multivariate analysis, the sexual risk differential for travelers is explained by occupational exposure and higher socio-economic status. The differential in condom acceptance, by contrast, appears to be associated with travel itself. Condom use with non-spousal partners is three times higher among travelers than non-travelers (P< 0.001), and travel remains a significant predictor after controlling for age, education, residence, occupation and multiple partners. Travelers are more likely to use condoms with both their local and non-local partners. Partners of male travelers are likely to be younger and better educated than those of male non-travelers. CONCLUSIONS: The mobile population in this rural region appears willing to adopt risk reduction measures appropriate to their exposure. This suggests that targeting condom promotion programs to travelers and their partners is likely to be effective in reducing the spatial diffusion of HIV, and may be an efficient method for spreading behavioral change into rural areas.

AIDS and HIV infection in Uganda--are more women infected than men?

In countries in sub-Saharan Africa, HIV is transmitted primarily heterosexually. HIV infection and AIDS in women not only affects women's health but also has implications for the other members of society. Maternal infection is the source of most childhood HIV infection in Africa and maternal health is a strong predictor of child survival. In Uganda, a review of passive AIDS surveillance has shown almost equal numbers of clinical cases reported in men and women. However, in three population-based HIV serosurveys, women were consistently found to have a higher infection rate (approximately 1.4 times) than men. In addition, both AIDS case surveillance and seroprevalence studies demonstrate an earlier age of presentation and mean age of infection in women. The higher rate of HIV infection in women suggests either differential rates of transmission between women and men, higher rates of female sexual exposure to infected men, or longer survival among HIV-infected women compared with men. Although further studies are required to illuminate both the biology and the epidemiology of heterosexual HIV transmission in Africa, these findings of earlier and higher infection rates in women have important implications for women's health and child survival in Uganda and indicate the need for specially targeted interventions to reduce transmission in this group.

HIV risk factors in three geographic strata of rural Rakai District, Uganda.

OBJECTIVES: To examine risk factors for HIV-1 infection in three geographic strata (main road trading centers that service local and international traffic, small trading villages on secondary dirt roads that serve as foci for local communications, and agricultural villages off main and secondary roads) in Rakai District, Uganda. DESIGN AND METHODS: Serological, sociodemographic, knowledge/behaviors and health survey conducted in 21 randomly selected community clusters; complete data were collected for 1292 consenting adults. RESULTS: Fifteen per cent of the men and 24% of the women were HIV-1-positive. On univariate analysis, several sociodemographic and behavioral factors were significantly associated with risk of HIV infection, including age, place of residence, travel, occupation, marital status, number of sex partners, sex for money or gifts, history of sexually transmitted disease (STD), and history of injections. On multivariate analysis, age, residence and number of sex partners remained significantly associated with HIV infection in both sexes; a history of STD and not having been circumcised were significant in men. There was a significant interaction between place of residence and reported number of sex partners: for any given level of sexual activity, the risk of HIV infection was markedly increased if the background community prevalence was high. CONCLUSION: Sexual transmission appears to be the primary behavioral risk factor for infection, but the risks associated with this factor vary substantially between the three geographic strata. These data can be used to design targeted interventions.

HIV incidence and sexually transmitted disease prevalence associated with condom use: a population study in Rakai, Uganda.

OBJECTIVE: Evidence of condom effectiveness for HIV and sexually transmitted disease (STD) prevention is based primarily on high-risk populations. We examined condom effectiveness in a general population with high HIV prevalence in rural Africa. METHODS: Data were from a randomized community trial in Rakai, Uganda. Condom usage information was obtained prospectively from 17,264 sexually active individuals aged 15-59 years over a period of 30 months. HIV incidence and STD prevalence was determined for consistent and irregular condom users, compared to non-users. Adjusted rate ratios (RR) of HIV acquisition were estimated by Poisson multivariate regression, and odds ratios of STDs estimated by logistic regression. RESULTS: Only 4.4% reported consistent condom use and 16.5% reported inconsistent use during the prior year. Condom use was higher among males, and younger, unmarried and better educated individuals, and those reporting multiple sex partners or extramarital relationships. Consistent condom use significantly reduced HIV incidence [RR, 0.37; 95% confidence interval (CI), 0.15-0.88], syphilis [odds ratio (OR), 0.71; 95% CI, 0.53-0.94] and gonorrhea/Chlamydia (OR, 0.50; 95% CI, 0.25-0.97) after adjustment for socio-demographic and behavioral characteristics. Irregular condom use was not protective against HIV or STD and was associated with increased gonorrhea/Chlamydia risk (OR, 1.44; 95% CI, 1.06-1.99). The population attributable fraction of consistent use for prevention of HIV was -4.5% (95% CI, -8.3 to 0.0), due to the low prevalence of consistent use in the population. CONCLUSIONS: Consistent condom use provides protection from HIV and STDs, whereas inconsistent use is not protective. Programs must emphasize consistent condom use for HIV and STD prevention.

Trends in HIV-1 prevalence may not reflect trends in incidence in mature epidemics: data from the Rakai population-based cohort, Uganda.

OBJECTIVES: To assess whether trends in serial HIV-1 prevalence reflect trend in HIV incidence, and to decompose the effects of HIV-1 incidence, mortality, mobility and compliance on HIV-1 prevalence in a population-based cohort. DESIGN: Two-year follow up (1990-1992) of an open cohort of all adults aged 15-59 years, resident in a sample of 31 representative community clusters in rural Rakai District, Uganda. METHODS: A detailed household enumeration was concluded at baseline and in each subsequent year. All household residents were listed, and all deaths and in- and out-migrations that occurred in the intersurvey year wee recorded. In each year, all consenting adults were interviewed and provided a serological sample; 2591 adults aged 15-59 years were enrolled at baseline. RESULTS: HIV prevalence among adults declined significantly 1990 and 1992 (23.4% at baseline, 21.8% in 1991, 20.9% in 1992; P < 0.05). Declining prevalence was also observed in subgroups, including young adults aged 15-24 years (from 20.6 to 16.2% over 3 years; P < 0.02), women of reproductive age (from 27.1 to 23.5%; P < 0.05), and pregnant women (from 25.4 to 20.0%; not significant), However, HIV incidence did not change significantly among all adults aged 15-59 years (2.1 +/- 0.4 per 100 person-years of observation (PYO) in 1990-1991 and 2.0 +/- 0.3 per 100 PYO in 1991-1992], nor in population subgroups. HIV-related mortality was high (13.5 per 100 PYO among the HIV-positive), removing more infected persons that were added by seroconversion. Net out-migration also removed substantial numbers of HIV-positive individuals. CONCLUSIONS: In this mature HIV epidemic, HIV prevalence declined in the presence of stable and incidence. HIV-related mortality contributed most to the prevalence decline. Prevalence was not an adequate surrogate measure of incidence, limiting the utility or serial prevalence measures in assessing the dynamics of the HIV epidemic and in evaluating the impact of current preventive strategies.

PIP: Findings are reported from a 2-year follow-up study of an open cohort of people aged 15-59 years living in a sample of 31 representative community clusters in rural Rakai district, Uganda, to measure whether trends in serial HIV-1 prevalence reflect trends in HIV incidence, and to gain insight into the effects of HIV-1 incidence, mortality, mobility, and compliance upon HIV-1 prevalence. In each year of study, all consenting adults provided a serological sample and were interviewed; 2591 adults were enrolled at baseline. HIV prevalence among adults declined significantly between 1990 and 1992; from 23.4% in 1990, to 21.8% in 1991, and 20.9% in 1992. Declining prevalence was also observed in subgroups, including young adults aged 15-24 years from 20.6% to 16.2%, reproductive-age women from 27.1% to 23.5%, and pregnant women from 25.4% to 20.0%. The decline in HIV prevalence among pregnant women, however, is not significant. HIV incidence did not change significantly among all adults aged 15-59 years, nor in population subgroups. HIV-related mortality was 13.5/person-year of observation among those who were HIV-positive. Substantial numbers of HIV-infected individuals were also loss to emigration.

Adolescents, sexual behaviour and HIV-1 in rural Rakai district, Uganda.

OBJECTIVES: To describe the epidemiology of HIV-1 infection among adolescents aged 13-19 years, in rural Rakai district, Uganda. STUDY DESIGN: Baseline survey and 2-year follow-up (1990-1992) of adolescents in a population-based, open rural cohort. METHODS: Annual enumeration and behavioral/serological survey of all consenting adolescents aged 13-19 years at recruitment, residing in 31 randomly selected community clusters. RESULTS: At baseline, of 909 adolescents present in study clusters, 824 (90.6%) provided interview data and serological samples. No adolescents aged 13-14 years were HIV-infected. Among those aged 15-19 years, 1.8% of men and 19.0% of women were HIV-positive. Among young women aged 15-19 years in marital/consensual union, 21.3% were HIV-positive; this rate did not differ significantly from the 29.1% prevalence in those reporting non-permanent relationships; prevalence was significantly lower in women reporting no current relationship (4.3%). After multivariate adjustment, female sex, age 17-19 years, residence in trading centers/trading villages and a history of sexually transmitted disease symptoms remained significantly associated with HIV infection. Seventy-nine per cent of adolescents provided a follow-up serological sample. No young men aged 13-14 years seroconverted during the study; in young women aged 13-14 years, HIV seroincidence was 0.6 per 100 person-years (PY) of observation. Among young men aged 15-19 years, there were 1.1 +/- 0.6 seroconversions per 100 PY of observation prior to age 21 years; among women 15-19 years, the incidence rate was 3.9 +/- 1.0 per 100 PY of observation prior to age 21 years. The mortality rate among HIV-positive adolescents aged 15-19 years, at 3.9 per 100 PY of observation, was 13-fold higher than that among the HIV-uninfected. By 1992, knowledge of sexual transmission was almost universal, the proportions reporting multiple partners had decreased and condom use had increased over baseline. CONCLUSIONS: Adolescents, and young women in particular, are vulnerable to HIV infection. Despite reported behavioral changes, HIV incidence rates remain substantial, and there is a need for innovative HIV preventive measures.

Demographic impact of HIV infection in rural Rakai district, Uganda: results of a population-based cohort study.

OBJECTIVES: To examine HIV-1-related mortality and demographic impact in a high HIV prevalence rural district of Uganda. DESIGN: One-year follow-up (1990-1991) in a population-based rural cohort. SETTING AND PARTICIPANTS: Annual enumeration of all consenting residents of 1945 households in 31 randomly selected community clusters in Rakai District. Subjects provided yearly HIV serological samples, behavioral and health information. MAIN OUTCOME MEASURE: Mortality in HIV-infected and uninfected persons. RESULTS: Mortality among HIV-seropositive adults aged > or = 15 years of 118.4 per 1000 person-years (PY) was substantially higher than in HIV-seronegative adults [12.4 per 1000 PY; relative risk (RR), 9.5; 95% confidence interval (CI), 6.0-14.9]. Infant mortality among offspring of HIV-infected mothers was almost double that for uninfected women (210 compared with 111 per 1000 live births; RR, 1.9; 95% CI, 1.0-3.5). Adult HIV-related mortality was associated with HIV prevalence and, in this cohort, with higher education, non-agricultural occupation and residence in roadside trading centers. We estimate that adult HIV prevalence in the district is 13% and adult HIV attributable mortality 52%. For all ages combined, district HIV attributable mortality is 28%. CONCLUSION: HIV is the leading cause of adult death in Rakai. Its effects on mortality are particularly marked in the most economically active sectors. However, the overall crude birth rate in the district (45.7 per 1000 population) remains higher than the crude death rate (28.1 per 1000 population), resulting in continued rapid population growth.

The social dynamics of HIV transmission as reflected through discordant couples in rural Uganda.

OBJECTIVE: To describe the role of men and women as sources of HIV transmission and to estimate HIV incidence among discordant couples resident in diverse rural communities in Uganda. SETTING: Rakai, a rural district in Uganda, East Africa. METHODS: A population-based cohort study, which has been conducted as annual serological and behavioral surveys since 1989. Community clusters were stratified into trading centers on main roads, intermediate trading villages on secondary roads and agricultural villages off roads. In the 1990 survey round, serological data were available for 79 discordant and 411 concordant HIV-negative couples aged 13-49 years. The present analysis examines sex-specific seropositivity associated with place of residence and the incidence of seroconversion among discordant couples between 1990 and 1991. RESULTS: Seventy-nine discordant couples were followed; the HIV-positive partner was male in 44 couples (57%) and female in 35 couples (43%). There was marked variation in the sex of the seropositive partner by place of residence: women were the HIV-positive partner in 57% of couples from trading centers, 52% from intermediate villages, and 20% from agricultural communities (P < 0.008). Condom use was higher in discordant couples in which the man was the uninfected partner (17.1%) rather than the woman (9.5%). HIV-positive women, but not HIV-positive men, reported significantly more sexual partners and more genital ulcers than seronegative individuals of the same sex. Seroincidence rates among men and women in discordant relationship were 8.7 and 9.2 per 100 person-years (PY), respectively, which was much higher than in concordant seronegative couples (men, 0.82; women, 0.87 per 100 PY). CONCLUSIONS: In this Ugandan population, men are the predominant source of new infections in rural villages. Risk factors and preventive behaviors vary with the sex of the infected partner, and seroconversion rates are similar in both sexes.

PIP: The analysis of marital relationships in rural Uganda in which one partner is infected with human immunodeficiency virus (HIV) revealed important information about the social dynamics of HIV transmission. As part of an ongoing cohort study of over 3000 adults in southwestern Uganda's Rakai district, data were collected on 79 discordant couples. The HIV-infected partner was male in 44 couples (57%) and female in 35 couples (43%). In the trading centers and intermediate trading villages, women were more likely to be the HIV-infected partner (57% and 52%, respectively); in rural areas, however, the male partner was the HIV carrier in 80% of discordant couples. 14.3% of women in female HIV-positive compared to only 2.1% of those in male HIV-positive discordant relationships reported two or more sexual partners in the year preceding the interview; there was no significant difference on this factor among HIV-positive and HIV-negative men (48.8% and 45,7%, respectively). 17.1% of seronegative men with HIV-infected wives compared to only 9.5% of seropositive men with seronegative wives used condoms with their partner. During the one-year study period, the rate of seroconversion was 8.7/100 years of observation for men and 9.2 for women. Extrapolation of the data from this study to the district as a whole suggests that 18.7% of couples have at least one HIV-infected member. Needed are campaigns to promote monogamy and condom use, particularly among men from rural areas who travel to high prevalence trading areas.

Male circumcision and HIV acquisition and transmission: cohort studies in Rakai, Uganda. Rakai Project Team.

BACKGROUND: Male circumcision is associated with reduced HIV acquisition. METHODS: HIV acquisition was determined in a cohort of 5507 HIV-negative Ugandan men, and in 187 HIV-negative men in discordant relationships. Transmission was determined in 223 HIV-positive men with HIV-negative partners. HIV incidence per 100 person years (py) and adjusted rate ratios (RR) and 95% confidence intervals (CI) were estimated by Poisson regression. HIV-1 serum viral load was determined for the seropositive partners in HIV-discordant couples. RESULTS: The prevalence of circumcision was 16.5% for all men; 99.1% in Muslims and 3.7% in non-Muslims. Circumcision was significantly associated with reduced HIV acquisition in the cohort as a whole (RR 0.53, CI 0.33-0.87), but not among non-Muslim men. Prepubertal circumcision significantly reduced HIV acquisition (RR 0.49, CI 0.26-0.82), but postpubertal circumcision did not. In discordant couples with HIV-negative men, no serconversions occurred in 50 circumcised men, whereas HIV acquisition was 16.7 per 100 py in uncircumcised men (P = 0.004). In couples with HIV-positive men, HIV transmission was significantly reduced in circumcised men with HIV viral loads less than 50000 copies/ml (P = 0.02). INTERPRETATION: Prepubertal circumcision may reduce male HIV acquisition in a general population, but the protective effects are confounded by cultural and behavioral factors in Muslims. In discordant couples, circumcision reduces HIV acquisition and transmission. The assessment of circumcision for HIV prevention is complex and requires randomized trials.

Mortality associated with HIV infection in rural Rakai District, Uganda.

OBJECTIVE: To assess mortality impact of HIV in rural Uganda. METHODS: An open cohort of 19983 adults aged 15-59 years, in Rakai district was followed at 10 month intervals for four surveys. Sociodemographic characteristics and symptomatology/disease conditions were assessed by interview. Deaths among residents and out-migrants were identified household census. Mortality rates were computed per 1000 person years (py) and the rate ratio (RR) of death in HIV-positive/HIV-negative subjects, and the population attributable fraction (PAF) of death were estimated according to sociodemographic characteristics. Mortality associated with potential AIDS defining symptoms and signs was assessed. RESULTS: HIV prevalence was 16.1%. Mortality was 132.6 per 1000 py in HIV-infected versus 6.7 per 1000 py in uninfected subjects, and 73.5% of adult deaths were attributable to HIV infection. Mortality increased with age, but the highest attributable risk of HIV associated deaths were observed in persons aged 20-39 years (PAF > 80%) and in women. HIV associated mortality was highest in the better educated (PAF > or = 75%) and among government employees (PAF > or = 82%). Of the HIV-positive subjects 40.5% reported no illness < 10 months preceding death, symptoms were poor predictors of death (sensitivity 1.6-38.8%), and only 9.1% met the World Health Organization clinical definition of AIDS. Infant mortality rates in babies of HIV-infected and uninfected mothers were 209.4 and 97.7 per 1000, respectively. CONCLUSION: HIV is taking substantial toll in this population, particularly among the younger better educated adults, and infants. Symptomatology or the World Health Organization definition of AIDS are poor predictors of death.

Age of male circumcision and risk of prevalent HIV infection in rural Uganda.

OBJECTIVE: To assess whether circumcision performed on postpubertal men affords the same level of protection from HIV-1 acquisition as circumcisions earlier in childhood. DESIGN: Cross-sectional study of a population-based cohort. SETTING: Rakai district, rural Uganda. METHODS: A total of 6821 men aged 15-59 years were surveyed and venous blood samples were tested for HIV-1 and syphilis. Age at circumcision was dichotomized into men who were circumcised before or at age 12 years (prepubertal) and men circumcised after age 12 years (postpubertal). Postpubertal circumcised men were also subdivided into those reporting circumcision at ages 13-20 years and > or = 21 years. RESULTS: HIV-1 prevalence was 14.1% in uncircumcised men, compared with 16.2% for men circumcised at age > or = 21 years, 10.0% for men circumcised at age 13-20 years, and 6.9% in men circumcised at age < or = 12 years. On bivariate analysis, lower prevalence of HIV-1 associated with prepubertal circumcision was observed in all age, education, ethnic and religious groups. Multivariate adjusted odds ratio of prevalent HIV-1 infection associated with prepubertal circumcision was 0.39 [95% confidence interval (CI), 0.29-0.53]. In the postpubertal group, the adjusted odds ratio for men circumcised at ages 13-20 years was 0.46 (95% CI, 0.28-0.77), and 0.78 (95% CI, 0.43-1.43) for men circumcised after age 20 years. CONCLUSIONS: Prepubertal circumcision is associated with reduced HIV risk, whereas circumcision after age 20 years is not significantly protective against HIV-1 infection. Age at circumcision and reasons for circumcision need to be considered in future studies of circumcision and HIV risk.

A randomized, community trial of intensive sexually transmitted disease control for AIDS prevention, Rakai, Uganda.

OBJECTIVE: To describe the design and first-round survey results of a trial of intensive sexually transmitted disease (STD) control to reduce HIV-1 incidence. STUDY DESIGN: Randomized, controlled, community-based trial in Rakai District, Uganda. METHODS: In this ongoing study, 56 communities were grouped into 10 clusters designed to encompass social/sexual networks; clusters within blocks were randomly assigned to the intervention or control arm. Every 10 months, all consenting resident adults aged 15-59 years are visited in the home for interview and sample collection (serological sample, urine, and, in the case of women, self-administered vaginal swabs). Sera are tested for HIV-1, syphilis, gonorrhea, chlamydia, trichomonas and bacterial vaginosis. Following interview, all consenting adults are offered directly observed, single oral dose treatment (STD treatment in the intervention arm, anthelminthic and iron-folate in the control arm). Treatment is administered irrespective of symptoms or laboratory testing (mass treatment strategy). Both arms receive identical health education, condom and serological counseling services. RESULTS: In the first home visit round, the study enrolled 5834 intervention and 5784 control arm subjects. Compliance with interview, sample collection and treatment was high in both arms (over 90%). Study arm populations were comparable with respect to sociodemographic and behavioral characteristics, and baseline HIV and STD rates. The latter were high: 16.9% of all subjects were HIV-positive, 10.0% had syphilis, and 23.8% of women had trichomonas and 50.9% had bacterial vaginosis. CONCLUSIONS: Testing the effects of STD control on AIDS prevention is feasible in this Ugandan setting.

PIP: An ongoing (1994-98) randomized, community-based trial in Uganda's Rakai District is assessing the assumption that intensive sexually transmitted disease (STD) control efforts result in marked declines in HIV/AIDS prevalence. Described, in this article, are the project design and findings of the first-round baseline survey. 56 communities were grouped into 10 clusters designed to encompass social/sexual networks and clusters within blocks were randomly assigned to the intervention or control arm. All consenting permanent residents of the district are visited in their homes at 10-month intervals where they are administered extensive questionnaires, provide urine and vaginal swab samples, and are offered mass treatment regardless of symptoms or laboratory testing (single oral dose STD treatment in the intervention arm and anthelmintics and iron folate in the control arm). Both groups receive identical health education, condom promotion, and serologic counseling services. In the first round of home visits, 5834 intervention and 5784 control arm subjects were enrolled, representing about 90% of eligible adults. The groups were comparable in terms of sociodemographic and behavioral characteristics and baseline rates of HIV and STDs. 16.9% of subjects were HIV-positive, 10.0% had syphilis, 23.8% of women had trichomonas, and 50.9% had bacterial vaginosis. Detailed STD assessment is expected not only to document the relationship between STD control and HIV, but also to identify which STDs confer the greatest population attributable risk for HIV transmission, facilitating targeted control efforts in the future.

Relative risks and population attributable fraction of incident HIV associated with symptoms of sexually transmitted diseases and treatable symptomatic sexually transmitted diseases in Rakai District, Uganda. Rakai Project Team.

OBJECTIVES: To assess the linkage of sexually transmitted disease (STD) symptoms and treatable STD to HIV incidence. DESIGN: Analysis of a randomized trial of STD control for HIV prevention, Rakai, Uganda. METHODS: Consenting adults 15-59 years of age were seen at 10-monthly home visits, interviewed regarding STD symptoms, and asked to provide samples for HIV and STD diagnoses. HIV incidence was determined in 8089 HIV-negative subjects over 10 457 person years. Adjusted rate ratios (RR) and 95% confidence intervals (CI) of HIV acquisition associated with genital ulcer disease (GUD) and discharge/dysuria were used to estimate the population attributable fraction (PAF) of HIV acquisition. HIV transmission risks associated with STD symptoms in HIV-positive partners of 167 HIV discordant couples and the numbers of sexual partners reported by HIV-positive subjects were used to estimate the PAF of HIV transmission attributable to STD. RESULTS: HIV prevalence was 16%. The risk of HIV acquisition was increased with GUD (RR 3.14; CI 1.98-4.98) and in males with discharge/dysuria (RR 2.44; CI 1.17-5.12), but not in females with discharge/dysuria. The PAF of HIV acquisition was 9.5% (CI 2.8-15.8%) with any of the three STD symptoms. The PAF for GUD was 8.8% (CI 3.7-13.8), but only 8.2% of reported GUD was caused by treatable syphilis or chancroid . The PAF for discharge/dysuria in males was 6.7% (CI 1.1-13.8), but only 25% of symptomatic males had concurrent gonorrhea or chlamydial infection. No significant differences were seen in PAF between study treatment arms. The PAF of HIV transmission associated with STD symptoms in HIV-positive persons was indirectly estimated to be 10.4%. CONCLUSION: In this mature, generalized HIV epidemic setting, most HIV seroconversion occurs without recognized STD symptoms or curable STD detected by screening. Therefore, syndromic management or other strategies of STD treatment are unlikely to substantially reduce HIV incidence in this population. However, STD is associated with significant HIV risk at the individual level, and STD management is needed to protect individuals.