Comparison of STD prevalences in the Mwanza, Rakai, and Masaka trial populations: the role of selection bias and diagnostic errors.

OBJECTIVES: To assess bias in estimates of STD prevalence in population based surveys resulting from diagnostic error and selection bias. To evaluate the effects of such biases on STD prevalence estimates from three community randomised trials of STD treatment for HIV prevention in Masaka and Rakai, Uganda and Mwanza, Tanzania. METHODS: Age and sex stratified prevalences of gonorrhoea, chlamydia, syphilis, HSV-2 infection, and trichomoniasis observed at baseline in the three trials were adjusted for sensitivity and specificity of diagnostic tests and for sample selection criteria. RESULTS: STD prevalences were underestimated in all three populations because of diagnostic errors and selection bias. After adjustment, gonorrhoea prevalence was higher in men and women in Mwanza (2.8% and 2.3%) compared to Rakai (1.1% and 1.9%) and Masaka (0.9% and 1.8%). Chlamydia prevalence was higher in women in Mwanza (13.0%) compared to Rakai (3.2%) and Masaka (1.6%) but similar in men (2.3% in Mwanza, 2.7% in Rakai, and 2.2% in Masaka). Prevalence of trichomoniasis was higher in women in Mwanza compared to women in Rakai (41.9% versus 30.8%). Herpes simplex virus type 2 (HSV-2) seroprevalence and prevalence of serological syphilis (TPHA+/RPR+) were similar in the three populations but the prevalence of high titre syphilis (TPHA+/RPR >/=1:8) in men and women was higher in Mwanza (5.6% and 6.3%) than in Rakai (2.3% and 1.4%) and Masaka (1.2% and 0.7%). CONCLUSIONS: Limited sensitivity of diagnostic and screening tests led to underestimation of STD prevalence in all three trials but especially in Mwanza. Adjusted prevalences of curable STD were higher in Mwanza than in Rakai and Masaka.

Syndromic management of sexually-transmitted infections and behaviour change interventions on transmission of HIV-1 in rural Uganda: a community randomised trial.

BACKGROUND: Treatment of sexually-transmitted infections (STIs) and behavioural interventions are the main methods to prevent HIV in developing countries. We aimed to assess the effect of these interventions on incidence of HIV-1 and other sexually-transmitted infections. METHODS: We randomly allocated all adults living in 18 communities in rural Uganda to receive behavioural interventions alone (group A), behavioural and STI interventions (group B), or routine government health services and community development activities (group C). The primary outcome was HIV-1 incidence. Secondary outcomes were incidence of herpes simplex virus type 2 (HSV2) and active syphilis and prevalence of gonorrhoea, chlamydia, reported genital ulcers, reported genital discharge, and markers of behavioural change. Analysis was per protocol. FINDINGS: Compared with group C, the incidence rate ratio of HIV-1 was 0.94 (0.60-1.45, p=0.72) in group A and 1.00 (0.63-1.58, p=0.98) in group B, and the prevalence ratio of use of condoms with last casual partner was 1.12 (95% CI 0.99-1.25) in group A and 1.27 (1.02-1.56) in group B. Incidence of HSV2 was lower in group A than in group C (incidence rate ratio 0.65, 0.53-0.80) and incidence of active syphilis for high rapid plasma reagent test titre and prevalence of gonorrhoea were both lower in group B than in group C (active syphilis incidence rate ratio, 0.52, 0.27-0.98; gonorrhoea prevalence ratio, 0.25, 0.10-0.64). INTERPRETATION: The interventions we used were insufficient to reduce HIV-1 incidence in rural Uganda, where secular changes are occurring. More effective STI and behavioural interventions need to be developed for HIV control in mature epidemics.

HIV-1 epidemic trends in rural south-west Uganda over a 10-year period.

The objective of this study was to examine the epidemic trends of HIV-1 infection in a rural population cohort in Uganda followed for 10 years. The methods used were to assess incidence and prevalence trends in adults in this longitudinal cohort study. The results showed that incidence of infection has fallen significantly in all adults, and separately in males, females, young adults and older adults over the course of the study period. There was also a reduction in prevalence, especially in young men and women. There was some evidence of a cohort effect in women. The conclusions are that this study provides the first evidence of a falling incidence in a rural general population in Africa. This was an observational cohort exposed to national health education messages, giving hope that similar campaigns elsewhere in Africa could be used effectively in efforts to control the HIV epidemic.

Declining HIV-1 incidence and associated prevalence over 10 years in a rural population in south-west Uganda: a cohort study.

BACKGROUND: In Uganda, there have been encouraging reports of reductions in HIV-1 prevalence but not in incidence, which is the most reliable measure of epidemic trends. We describe HIV-1 incidence and prevalence trends in a rural population-based cohort between 1989 and 1999. METHODS: We surveyed the adult population of 15 neighbouring villages for HIV-1 infection using annual censuses, questionnaires, and serological surveys. We report crude annual incidence rates by calendar year and prevalence by survey round. FINDINGS: 6566 HIV-1 seronegative adults were bled two or more times between January, 1990, and December, 1999, contributing 31984 person years at risk (PYAR) and 190 seroconversions. HIV-1 incidence fell from 8.0 to 5.2 per 1000 PYAR between 1990 and 1999 (p=0.002, chi(2) for trend). Significant sex-specific and age-group-specific reductions in incidence were evident. Incidence was 37% lower for 1995-99 than for 1990-94 (p=0.002, t-test). On average, 4642 adult residents had a definite HIV-1 serostatus at each yearly survey round. HIV-1 prevalence fell significantly between the first and tenth annual survey rounds (p=0.03, chi(2) for trend), especially among men aged 20-24 years (6.5% to 2.2%) and 25-29 years (15.2% to 10.9%) and women aged 13-19 years (2.8% to 0.9%) and 20-24 years (19.3% to 10.1%) (all p<0.001, chi(2) for trend). INTERPRETATION: Our findings of a significant drop in adult HIV-1 incidence in rural Ugandans give hope to AIDS control programmes elsewhere in sub-Saharan Africa where rates of HIV-1 infection remain high.

Effect of HIV-1 and increasing immunosuppression on malaria parasitaemia and clinical episodes in adults in rural Uganda: a cohort study.

BACKGROUND: An association between HIV-1 and malaria is expected in theory, but has not been convincingly shown in practice. We studied the effects of HIV-1 infection and advancing immunosuppression on falciparum parasitaemia and clinical malaria. METHODS: HIV-1-positive and HIV-1-negative adults selected from a population-based cohort in rural Uganda were invited to attend a clinic every 3 months (routine visits) and whenever they were sick (interim visits). At each visit, information was collected on recent fever, body temperature, and malaria parasites. Participants were assigned a clinical stage at each routine visit and had regular CD4-cell measurements. FINDINGS: 484 participants made 7220 routine clinic visits between 1990 and 1998. Parasitaemia was more common at visits by HIV-1-positive individuals (328 of 2788 [11.8%] vs 231 of 3688 [6.3%], p<0.0001). At HIV-1-positive visits, lower CD4-cell counts were associated with higher parasite densities, compared with HIV-1-negative visits (p=0.0076). Clinical malaria was significantly more common at HIV-1-positive visits (55 of 2788 [2.0%] vs 26 of 3688 [0.7%], p=0.0003) and the odds of having clinical malaria increased with falling CD4-cell count (p=0.0002) and advancing clinical stage (p=0.0024). Participants made 3377 interim visits. The risk of clinical malaria was significantly higher at visits by HIV-1-positive individuals than HIV-1-negative individuals (4.0% vs 1.9%, p=0.009). The risk of clinical malaria tended to increase with falling CD4-cell counts (p=0.052). INTERPRETATION: HIV-1 infection is associated with an increased frequency of clinical malaria and parasitaemia. This association tends to become more pronounced with advancing immunosuppression, and could have important public-health implications for sub-Saharan Africa.

Seven-year trends in HIV-1 infection rates, and changes in sexual behaviour, among adults in rural Uganda.

OBJECTIVE: To assess trends in HIV-1 infection rates and changes in sexual behaviour over 7 years in rural Uganda. METHODS: An adult cohort followed through eight medical-serological annual surveys since 1989-1990. All consenting participants gave a blood sample and were interviewed on sexual behaviour. RESULTS: On average, 65% of residents gave a blood sample at each round. Overall HIV-1 prevalence declined from 8.2% at round 1 to 6.9% at round 8 (P = 0.008). Decline was most evident among men aged 20-24 years (11.7 to 3.6%; P < 0.001) and women aged 13-19 (4.4% to 1.4%; P = 0.003) and 20-24 (20.9% to 13.8%; P = 0.003). However, prevalence increased significantly among women aged 25-34 (13.1% to 16.6%; P = 0.04). Although overall incidence declined from 7.7/1000 person-years (PY) in 1990 to 4.6/1000 PY in 1996, neither this nor the age-sex specific rates changed significantly (P > 0.2). Age-standardized death rates for HIV-negative individuals were 6.5/1000 PY in 1990 and 8.2/1000 PY in 1996; corresponding rates for HIV-positive individuals were 129.7 and 102.7/1000 PY, respectively. There were no significant trends in age-adjusted death rates during follow-up for either group. There was evidence of behaviour change towards increase in condom use in males and females, marriage at later age for girls, later sexual debut for boys and a fall in fertility especially among unmarried teenagers. CONCLUSIONS: This is the first general population cohort study showing overall long-term significant reduction in HIV prevalence and parallel evidence of sexual behaviour change. There are however no significant reductions in either HIV incidence or mortality.

Sexual networks in Uganda: mixing patterns between a trading town, its rural hinterland and a nearby fishing village.

The study was based in south-west Uganda where significant differences in HIV prevalence have been found between urban and rural areas. Longitudinal data collected in a diary format was used to determine the extent to which high-risk men and women living in a truck stop/trading town had sexual contact with people from surrounding rural areas and a nearby fishing village. Study participants were 143 men, 75 of whom were resident in the town, 40 in a fishing village and 28 in rural areas, and 81 women, of whom 47 were resident in the town, 25 in the fishing village and 9 in a rural area. During 1687 man weeks the 143 men made 3149 trips and had 5189 sexual contacts. Ninety-two per cent of these sexual contacts occurred in the man's current place of residence and 21% were with a new partner. The 81 women participated for 1280 women weeks during which they recorded 6378 sexual contacts. Women who lived in the fishing village and the rural area had around 90% of their contacts with local men while those who lived in the town fell into 3 categories: women who charged a relatively high price for commercial sex had only 11% of contacts with men living in the town, while those who charged a tenth of the price had 71% of contacts with town men. The small number of women who fell into an intermediate category, in terms of price, had sexual contact with a wide variety of men. These findings show that there is little scope for HIV infection to spread between different residential or occupational groups. This may help to explain how large differences in HIV seropositivity between neighbouring localities can be maintained for long periods, despite considerable social and economic mixing between groups and high levels of sexual partner change within groups.

PIP: In southwest Uganda, there is concern that sexual mixing between rural areas with low HIV prevalence and urban areas with high HIV prevalence will produce uniformly high rates of HIV. To determine the extent of such mixing, a prospective study was conducted in a trading town on the trans-African highway with a steady flow of male truckers, an agricultural hinterland to the west, and a fishing village to the east. A total of 143 men recruited largely from bars and 81 women reported to be sexually promiscuous kept records of their sexual contacts over a 6-month period. During 1687 man-weeks of observation, the men made 3149 road trips and had 5189 sexual contacts; 92% of these contacts occurred in the men's place of residence and 21% involved a new partner. 59% of town men's sexual contacts, 61% reported by men in the fishing village, and 52% of those in rural areas involved casual partners. An additional 6378 sexual encounters were recorded during 1280 woman-weeks. Close to 90% of women's sexual contacts in the 2 rural areas involved local men. Among the highest-paid town commercial sex workers, only 11% of sexual contacts involved men from the town; in contrast, 71% of sexual encounters among town women who charged only 10% the price of their more expensive counterparts were with local men. Serologic testing of a subset of 75 men and 52 women yielded HIV prevalences of 28% and 52%, respectively. The distinct sexual networks identified in this study suggest it is unlikely that rates of HIV infection in the rural areas will reach those in nearby trading towns.

Computer-assisted quality assurance in an HIV serology laboratory.

A computer-based quality assurance programme for an HIV-1 serology laboratory is described. The programme was designed to minimise transcription errors and to provide rapid feedback on laboratory performance. Similar systems could readily be introduced to any laboratory with access to simple computing facilities.

Algorithms for detecting antibodies to HIV-1: results from a rural Ugandan cohort.

OBJECTIVE: To evaluate an algorithm using two enzyme immunoassays (EIA) for anti-HIV-1 antibodies in a rural African population and to assess alternative simplified algorithms. METHODS: Sera obtained from 7895 individuals in a rural population survey were tested using an algorithm based on two different EIA systems: Recombigen HIV-1 EIA and Wellcozyme HIV-1 Recombinant. Alternative algorithms were assessed using negative or confirmed positive sera. RESULTS: None of the 227 sera classified as unequivocably negative by the two assays were positive by Western blot. Of 192 sera unequivocably positive by both assays, four were seronegative by Western blot. The possibility of technical error cannot be ruled out in three of these. One of the alternative algorithms assessed classified all borderline or discordant assay results as negative had a specificity of 100% and a sensitivity of 98.4%. The cost of this algorithm is one-third that of the conventional algorithm. CONCLUSIONS: Our evaluation suggests that high specificity and sensitivity can be obtained without using Western blot and at a considerable reduction in cost.

PIP: Although the Western blot test is widely used to confirm HIV-1 serostatus, concerns over its additional cost have prompted review of the need for supplementary testing and the evaluation of alternative test algorithms. Serostatus tends to be confirmed with this additional test especially when tested individuals will be informed of their serostatus or when results will be used for research purposes. The confirmation procedure has been adopted as a means of securing suitably high levels of specificity and sensitivity. With the goal of exploring potential alternatives to Western blot confirmation, the authors describe the use of parallel testing with a competitive and an indirect enzyme immunoassay with and without supplementary Western blots. Sera were obtained from 7895 people in the rural population survey and tested with an algorithm based on the Recombigen HIV-1 EIA and Wellcozyme HIV-1 Recombinant; alternative algorithms were assessed on negative or confirmed positive sera. None of the 227 sera classified as negative by the 2 assays were positive by Western blot. Of the 192 identified ass positive by both assays, 4 were found to be seronegative with Western blot. The possibility of technical error does, however, exist for 3 of these latter cases. One of the alternative algorithms assessed classified all borderline or discordant assay results as negative with 100% specificity and 98.4% sensitivity. This particular algorithm costs only one-third the price of the conventional algorithm. These results therefore suggest that high specificity and sensitivity may be obtained without using Western blot and at a considerable reduction in cost.