Share this article on:

Risk Factors for HIV-1 Infection in a Longitudinal, Prospective Cohort of Adults From the Mbeya Region, Tanzania

Geis, Steffen MD*†; Maboko, Leonard MD, PhD; Saathoff, Elmar PhD*; Hoffmann, Oliver MD, PhD*†; Geldmacher, Christof PhD*†; Mmbando, Donan MD; Samky, Eleuter MD§; Michael, Nelson L MD, PhD‖¶; Birx, Deborah L MD**; Robb, Merlin L MD‖††; Hoelscher, Michael MD, PhD*†

JAIDS Journal of Acquired Immune Deficiency Syndromes: April 15th, 2011 - Volume 56 - Issue 5 - p 453-459
doi: 10.1097/QAI.0b013e3182118fa3
Epidemiology and Prevention

Background: To control the global HIV epidemic, targeted interventions to reduce the incidence of HIV infections are urgently needed until an effective HIV vaccine is available. This study describes HIV-1 incidence and associated risk factors in a general population cohort of adults from Mbeya region, Tanzania, who participated in a vaccine preparedness study.

Methods: We conducted a closed prospective cohort study with 6-monthly follow-up from 2002 to 2006 enrolling adults from the general population. HIV-1 incidence and risk factors for HIV-1 acquisition were analyzed using Cox regression.

Results: We observed 2578 seronegative participants for a mean period of 3.06 person years (PY) (7471 PY in total). Overall HIV-1 incidence was 1.35 per 100 PY (95% confidence interval [CI], 1.10-1.64/100 PY). The highest overall HIV-1 incidence was found in females from Itende village (1.55 per 100 PY; 95% CI, 0.99-2.30/100 PY); the highest age-specific incidence was observed in semiurban males aged 30 to 34 years (2.75 per 100 PY; 95% CI, 0.75-7.04). HIV-1 acquisition was independently associated with female gender (hazard ratio [HR], 1.64; 95% CI, 1.05-2.57), younger age at enrollment (age 18-19 versus 35-39 years: HR, 0.29; 95% CI, 0.11-0.75), alcohol consumption (almost daily versus none: HR, 2.01; 95% CI, 1.00-4.07), education level (secondary school versus none: HR, 0.39; 95% CI, 0.17-0.89), and number of lifetime sex partners (more than five versus one: HR, 2.22; 95% CI, 1.13-4.36).

Conclusions: A high incidence of HIV was observed in this cohort, and incident infection was strongly associated with young age, alcohol consumption, low school education level, and number of sex partners. Targeted interventions are needed to address the elevated risk associated with these factors.

Supplemental Digital Content is available in the text.

From the *Division of Infectious Diseases and Tropical Medicine, Medical Center of the University of Munich (LMU), Munich, Germany; †NIMR-Mbeya Medical Research Programme, Mbeya, Tanzania; ‡Mbeya Regional Medical Office, Mbeya, Tanzania; §Mbeya Referral Hospital, Mbeya, Tanzania; ‖US Military HIV Research Program (MHRP), Rockville, MD; ¶Walter Reed Army Institute of Research, Rockville, MD; **Global AIDS Program, Centers for Disease Control and Prevention, Atlanta, GA; and ††Henry M. Jackson Foundation for the Advancement of Military Medicine, Rockville, MD.

Received for publication August 25, 2010; accepted January 20, 2010.

The CODE study was supported by a cooperative agreement between the Henry M. Jackson Foundation for the Advancement of Military Medicine and the US Department of Defense under DAMD17-98-2-7007 and by the National Institute for Allergy and Infectious Diseases, National Institutes of Health (“HIV Vaccine Research and Development-Project 2” Y1-AI-2642-11).

Presented in part at the 10th Congress on Infectious Diseases and Tropical Medicine, June 2010, Cologne, Germany; Abstract number: TRO 03-5.

S.G. and L.M. contributed equally to this article. M.H., M.R., D.B., N.M., L.M., O.H., D.M., and E. Samky designed the cohort study; O.H., S.G., C.G., and L.M. supervised field and laboratory work; S.G., L.M., and E. Saathoff were responsible for data management and conducted the analysis; S.G. wrote the first draft of the article; all authors commented on drafts of the manuscript and approved the final version; S.G., L.M., and M.H. act as guarantors for the results presented in this article.

The opinions in this article are those of the authors and are not to be construed as official or representing the views of the Walter Reed Army Institute of Research, the Armed Forces Medical Research Institute, the US Army, or the US Department of Defense.

Correspondence to: Steffen Geis, MD, and Michael Hoelscher, MD, PhD, Division of Infectious Diseases and Tropical Medicine, Medical Center of the University of Munich (LMU), Leopoldstr. 5, 80802 Munich, Germany (e-mail:

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF version of this article on the journal's web site (

Back to Top | Article Outline


Despite advances in HIV prevention, treatment, and care, HIV/AIDS continues to be a significant cause of morbidity and mortality worldwide, particularly in Africa.1 To halt the spread of HIV-1, it will require further advances on all achieved efforts and additional biomedical interventions like a safe and effective HIV vaccine.

Understanding the dynamics of HIV infections in a community risk setting is essential both for planning such interventions and to evaluate their impact on the population. Most such data derive from trends in HIV-1 prevalence that should be interpreted only with caution.2 The most reliable method to monitor the pandemic is direct HIV incidence measurement. Characterization of recent seroconverters is critical to the design of better targeted preventive interventions. However, these data are rare for general populations in Africa because prospective cohort studies are costly and difficult to execute.

We report HIV-1 incidence rates and associated risk factors from a vaccine preparedness cohort study in the Mbeya region, Tanzania, determining the ability to recruit and retain people from the general population at risk for HIV infection and identifying the factors important for the design of future HIV vaccine efficacy trials.

Back to Top | Article Outline


Ethical Considerations

Laboratory and field work was done in accordance with the Helsinki Declaration of 1975 as revised in 2000 and was also approved by the ethics committees of the Mbeya Referral Hospital, the Tanzanian National Institute of Medical Research, and the Medical Center of the University of Munich as well as the Institutional Review Board of the Walter Reed Army Institute of Research. All participants provided written informed consent before enrollment.

Back to Top | Article Outline


Between September 2002 and April 2003, we used two different approaches to recruit 3096 volunteers of both sexes from the Mbeya region in southwestern Tanzania. Volunteers from Ghana ward, an urban area in Mbeya town, and from Itende, a small semiurban village close to Mbeya town, were recruited in a door-to-door campaign. A third subgroup of volunteers was recruited by advertisement in all wards of Mbeya town.

During the door-to-door campaign, trained staff visited all households in the designated study areas to inform them about the planned cohort study and to invite all eligible household members to community meetings. The ‘‘advertisement’’ recruitment strategy involved use of Institutional Review Board-approved banners, fliers, posters, and public announcements in Mbeya town to also invite potential participants to our community meetings.

During these meetings, the research team provided more details about participation in the study. At the end of each meeting, individuals aged 18 to 45 years were encouraged to report to our designated research facilities that were fully incorporated into the Mbeya Referral Hospital and the Itende Health Centre. These research facilities served for enrollment and follow-up evaluations throughout the whole study.

Back to Top | Article Outline

Enrollment and Study Procedures

We included residents of Mbeya town and Itende village between 18 and 45 years of age, who were planning to stay in the area for at least 3 years and who were willing and able to provide informed consent and to adhere to study procedures, including blood specimen collection.

During the first study visit, potential participants were consented, pre- and posttest counseled, tested for HIV-1, and also underwent a clinical examination. Data on sociodemographic background, relevant behavior, and knowledge about HIV/AIDS were collected in face-to-face-interviews using standardized questionnaires.

Over a 4-year period, we collected venous blood samples for HIV-1 testing and storage by phlebotomy during each of the 6-monthly follow-up visits. At each visit, participants received voluntary counseling about HIV-1 transmission and prevention and free condoms (male and female). Participants with sexually transmitted infections received free treatment according to Tanzanian national guidelines. Those who were HIV-1-infected at enrollment, or became infected during the study, were referred to the Mbeya Referral Hospital. Provision of antiretroviral treatment by the hospital started in 2005. Throughout the duration of the study, our research clinics provided free and comprehensive outpatient care according to Tanzanian national guidelines to all participants. This included referral to the specialty clinics of the Mbeya Referral Hospital, if needed.

Back to Top | Article Outline

Laboratory Tests

Venous blood samples were tested for HIV-1 with a dual enzyme-linked immunosorbent assay strategy (HIV-Determine; Abbott Laboratories, Abbott Park, IL; and Enzygnost HIV-1/2 plus; Behring, Liederbach, Germany), which was confirmed by HIV-1 Western blot (HIVblot 2.2 Genelabs/Abbott, Wiesbaden, Germany) if discordant. The last seronegative samples of incident HIV-1 cases were re-examined with polymerase chain reaction to close the diagnostic windows.

Back to Top | Article Outline

Statistical Analysis

We summarized demographic and behavioral characteristics at baseline. HIV-1 incidence was calculated as the number of seroconversions per 100 person-years (PY). Seroconversion was assumed to have taken place midway between the dates of the last negative and the first positive serology result. We defined the time at risk for HIV-1 infection as the time elapsed after the date of enrollment. Participants who seroconverted exited at the time of seroconversion; participants who remained seronegative were censored at the date of the last follow-up visit they attended. Univariable and multivariable Cox proportional hazard models were used to assess the association of potential risk factors with HIV-1 seroconversion using a modified conceptual framework approach.3 Initially, single factors adjusted for age, sex, and study location, which we considered as a priori confounders, were examined one by one in univariable analyses. Sociodemographic factors (eg, marital status, household size, or occupation) with a univariable P value <0.2 were included in a multivariable model and retained if their P value in this multivariable core model was <0.1. Baseline behavioral factors (eg, condom use, age at sexual debut, or age of partner) were added to this model one by one and included in a multivariable model if their univariable P value was <0.2. Those with a P value <0.1 were again retained. Biologic factors (eg, circumcision, pregnancy, or self-reported symptoms for sexually transmitted infections) were examined using the same approach. All retained factors were analyzed with a stepwise backward regression model excluding them one at a time until all remaining factors had a P value <0.15. The probability of HIV-1 infection over time was calculated and graphically displayed using Kaplan-Meier failure estimates. Stata Version 10 SE (Stata Corp, College Station, TX) was used for all statistical analyses.

Back to Top | Article Outline


Characteristics of the Seronegative Study Population During Enrollment

In total, we enrolled 2578 seronegative participants, 864 in Ghana ward in Mbeya, 846 in the advertisement group, and 868 in semiurban Itende (Table 1). Forty-four percent were men and 46% were aged younger than 25 years. The mean age at enrollment was 27.2 (standard deviation, 7.9) years; no difference was observed between male and female participants' age (male, 27.3 [standard deviation, 7.9] years; female = 27.2 [standard deviation, 7.8]; P = 0.552 [chi-square test]). There also was no significant difference in age distribution between urban participants from Mbeya and semiurban participants from Itende (P = 0.137). Ninety percent of participants had attended primary school and more than 85% had completed primary school. Education levels differed between the two sites and recruitment strategies. Semiurban participants were mostly farmers, whereas half of the urban participants worked as occasional workers or were unemployed. The higher unemployment rate in Mbeya town can thus be attributed to the fact that subsistence farming is impossible in an urban setting. Table 1 summarizes the demographic and social characteristics of the participants during enrollment.



Back to Top | Article Outline


At the first follow-up visit, 137 of 2578 initially seronegative participants (5.3%) did not return and could therefore not be included in the analysis of HIV-1 incidence. All remaining participants were followed for 7471 PY overall with a mean duration of 3.06 years (range, 0.32-3.83 years). Overall, 1776 participants (68.9% of the enrolled) completed follow-up defined as participating until the date of seroconversion or until the end of the study. Participants who were lost to follow-up were more likely to be male, of younger age, from the urban site, and recruited by advertisement (P value for trend < 0.01 for all mentioned variables), whereas not belonging to a religious group and having a low school education level was positively associated with compliance over the whole study period (P value for trend < 0.01 for both variables).

Back to Top | Article Outline

Incidence of HIV-1 Infection

One hundred one participants seroconverted during the course of the study resulting in an HIV-1 incidence rate of 1.35 per 100 PY (95% confidence interval [CI], 1.10-1.64/100 PY). The highest overall HIV-1 incidence was found in females from Itende (1.55 per 100 PY; 95% CI, 0.99-2.30/100 PY), the highest age- and sex-specific incidence among males aged 30 to 34 years from the semiurban setting (2.75 per 100 PY; 95% CI, 0.75-7.04). Figure 1 shows the incidence of HIV seroconversion (point estimates) for males and females stratified by recruitment site and age. When combining participants in two age groups (18-29 and 30-45 years) of both sexes, young women had the highest incidence rate of 1.86 per 100 PY (95% CI, 1.38-2.46) compared with young men (1.33 per 100 PY; 95% CI, 0.88-1.94), elder women (0.81 per 100 PY; 95% CI, 0.43-1.39), and elderly men (0.94 per 100 PY; 95% CI, 0.47-1.69). The almost even distribution of probability of HIV-1 infection over time of the Kaplan-Meier survival functions indicates that there is almost no cohort effect in this distinct study population (Fig. 2).





Back to Top | Article Outline

Age-, Sex- and Residence-Adjusted Risk Factors for Acquiring HIV-1 Infections

Female sex (hazard ratio [HR], 1.22; 95% CI, 0.82-1.84), current residence in Itende village (HR, 1.10; 95% CI, 0.73-1.66), and younger age (age 40-45 versus 18-19 years: HR, 0.21; 95% CI, 0.06-0.71) showed some association with HIV-1 seroconversion when adjusted for the other two variables, respectively (Table 2). Although only the association with age was statistically significant at the 5% level of confidence, we included age, sex, and residence in all models as potentially important a priori confounders.



Other risk factors for HIV-1 incidence (see Table, Supplemental Digital Content 1, that were significant in the univariable analysis included having no religion (HR, 2.15; 95% CI, 1.03-4.50), low school education level (finished primary school versus none: HR, 0.56; 95% CI, 0.32-0.98), alcohol consumption (almost daily versus none: HR, 3.48; 95% CI, 1.73-7.02), and number of sex partners (more than five versus one: HR, 2.68; 95% CI, 1.41-5.11). There was no significant difference in HIV-1 incidence between the two recruitment strategies that were used in Mbeya town (HR of advertisement versus door-to-door cohort, 1.41; 95% CI, 0.82-2.40).

The final multivariable model included all variables shown in Table 2. Female sex (HR, 1.64; 95% CI, 1.05-2.57) and younger age at enrollment (age 18-9 versus 35-39 years: HR, 0.29; 95% CI, 0.11-0.75) were independently associated with HIV-1 incidence, whereas living in Itende did not show an association (HR, 0.90; 95% CI, 0.53-1.52). Furthermore, alcohol consumption (almost daily versus none: HR, 2.01; 95% CI, 1.00-4.07), school education level (secondary school versus none: HR, 0.39; 95% CI, 0.17-0.89), and number of sex partners (more than five versus one: HR, 2.22; 95% CI, 1.13-4.36) were independently associated with HIV-1 incidence. There was also some (nonsignificant) evidence of an association between HIV incidence and having no religion (HR, 1.78; 95% CI, 0.87-3.66).

Back to Top | Article Outline


In our study population, HIV-1 incidence (1.35 per 100 PY) was four times as high as recent estimates from Tanzanian household-based prevalence data for the whole country (0.34/100 PY).4 Cohort studies from northern Tanzania report an incidence between 0.39 and 1.20 per 100 PY depending on year of data collection and study site.5,6 The difference between our and the national estimates can partially be explained by the differences in HIV-1 prevalence; the HIV-1 prevalence in our study (14.8% for males and 21.2% for females in Mbeya; 11.5% for males and 13.4% for females in Itende)7 was almost twice the national average.8 Although estimating incidence from cross-sectional prevalence data is well established,9 incidence measured in cohort studies remains the gold standard for accuracy, especially for intervention trials. In the Step study,10 which assessed the efficacy of a cell-mediated immunity vaccine, no increased HIV incidence was observed during the trial, although cross-sectional data before the study indicated that women were at high risk.

The main burden of new infections in our study is concentrated in women, both in urban and semiurban settings. This is consistent with the mentioned prevalence survey from Tanzania and other African countries.4 An elevated risk resulting from casual partnerships with older men11,12 during early sexual experience, which could have explained this pattern, was not observed in our study (data not shown). A possible explanation for the higher incidence in women could be the increased HIV-1 susceptibility of women, which has been reported in many previous studies.13

The age distribution differs considerably between our urban and rural populations; although in Mbeya town, most of the infections occurred before the age of 25 years, Itende showed a peak incidence rate in the group of 25- to 29-year-old women and 30- to 34-year-old men. The age distribution of HIV-1-infected participants from Itende is similar to that of a rural cohort study close to Mwanza.6 Differences in behavior and socioeconomic status between urban and rural settings itself could explain this delayed acquisition of HIV-1 in Itende as well as the steady spread of HIV from urban settlements into the rural population.14 Important for the planning of recruitment strategies in future Phase III vaccine trials is that a similar HIV-1 incidence was found in the two groups from Mbeya who were recruited by advertisement and by door-to-door campaign. Our data thus do not provide evidence that advertisement leads to an increase in response bias.

We did not observe a significant decline in seroconversions over time attributed to a closed cohort effect like other studies,15,16 although we performed regular health education sessions, offered counseling, and promoted free condom use.

Most studies on risk factors for HIV infection are based on prevalent rather than incident cases. Hence, their results also take account of the time after HIV-1 acquisition and are biased by determinants of disease duration. Our approach of analyzing incident cases for association excludes this bias.

We identified risk factors that are potentially preventable by public health interventions. We observed a strong association between alcohol consumption and newly acquired HIV infection, which has already been described in other studies from Africa,3,17 and we were able to demonstrate that the incidence rate was positively associated with frequency of alcohol consumption. This is usually explained by the effect of alcohol to reduce inhibition and to diminish perception of exposure to risky sexual behavior, violence, forced sex, and rape.18,19 Tanzanians consume alcohol frequently in restaurants, bars, and local brew establishments where they also encounter new sex partners.20 In a previous cohort study of female bar workers in Mbeya region, 67% of the participants were HIV-1-seropositive at enrollment,21 demonstrating the high-risk potential for HIV acquisition in their clients.

Another notable result is the association between the risk of HIV-1 acquisition and the number of lifetime sexual partners, which, in the absence of data regarding concurrent sexual partnerships and frequency of sexual intercourse, we use as a proxy for current sexual behavior.22 A similar association was found in a cross-sectional study of prevalent HIV infection in Northern Tanzania.23 Indeed, there is no evidence that Africans typically have more sexual partners than elsewhere in the world,24 but still sexual behavior might be an important risk factor because having more than one concurrent partnership seems to increase the risk of HIV acquisition25 and is in addition more frequently found in Africa.24

Although in the early phase of the epidemic, HIV risk was linked to higher educational attainment,26,27 some studies in Africa observed a similar shift toward reduced risk among higher educated people as found in our study.28-30 Staying in school for a longer time plays an important role for being more frequently and intensively exposed to health and basic education. Subsequently, better educated people seem to understand and adapt health education messages faster, leading to behavioral changes like delay in sexual debut, reduction of sexual partners, or increased condom use.28,31 Furthermore, educational status is likely to reflect socioeconomic-status, which is also negatively related to HIV incidence.32

Lastly, we found that study participants without religious denomination were at higher risk of HIV-1 acquisition. This finding could be explained by the role of religious communities as influential social networks that can influence health promotion messages.33 Previous studies addressing correlations among religious beliefs, behavior, and risk for HIV-1 infections often focused on Muslim populations with a particular interest in male circumcision, which is a known protective intervention.34,35 Our results indicate such a correlation as well (23% risk reduction in the adjusted univariable analysis; HR, 0.77; 95% CI, 0.34-1.77), although male circumcision did not reach statistical significance in the final multivariable model.

Our study and this analysis have certain limitations. The main objective of the CODE study was to explore the best recruitment strategy for future HIV vaccine trials; thus, our different recruitment strategies might have caused some sampling bias. However, the results of the two urban strata did not differ significantly; thus, we consider them as representative for the urban population of this age group in Mbeya. However, choosing a semiurban setting in walking distance of the urban center of Mbeya might be another source of selection bias because the population in this setting might closely resemble the urban population. Second, losses to follow-up might have influenced our results, although we attained a very high retention rate and identified characteristics for discontinuation of the study. The exact bias is difficult to assess because we were unable to collect information about reasons for loss to follow-up. It should however be noted that our losses to follow-up (31% over the study period including participants who died) are relatively low for a study in sub-Saharan Africa with a duration of 4 years. Third, our risk analysis only relies on data collected during enrollment because this was the most complete data set. It is therefore possible that some of our covariates changed over time, which is not reflected in this analysis. Fourth, the validity of our data obtained during face-to-face interviews is subject to underreporting bias, especially for socially sensitive behaviors. Our study participants might have felt pressure to please our study staff in addition to the normal pressure to underreport sensitive behaviors.36 A method to overcome this bias, namely audio-computer-assisted self-interviewing, was not considered to be appropriate in this setting. Lastly, like in other closed cohorts, we might have encountered a cohort attrition effect, leading to a reduced incidence rate toward the end of the study. However, the relatively stable HIV-1 incidence over time that we found does not indicate that this applies.

In conclusion, this study enhances knowledge for understanding of the ongoing HIV epidemic in East Africa and for implementing further public health actions in this setting. We observed a relatively high incidence of HIV, indicating an ongoing dynamic, especially among young women who are at higher risk, despite huge preventive efforts in the past decades. Alcohol consumption, low school education level, and number of sex partners represent key risk factors; not belonging to a religious group seems to be a risk factor as well. To reduce HIV infection rates, behavioral interventions toward the reduction of alcohol intake and safer sex practices should be intensified in these communities. The association with a low school education level that we found indicates that health education programs should be strengthened and go beyond formal schooling. Existing structures of social networks (eg, religious groups) should be used and new networks should be identified to increase the catchment population for preventive measures. Finally, prevention programs should empower young women because they are the group with the highest risk of HIV-1 infection in this region.

Back to Top | Article Outline


We thank the communities of Mbeya town and Itende village who volunteered to participate in this study, the CODE Research Team for their invaluable contribution, and the Mbeya District Local Government leaders for their cooperation.

Back to Top | Article Outline


1. UNAIDS, World Health Organization. Report on the Global HIV/AIDS Epidemic 2008. Geneva: UNAIDS; 2008.
2. Wawer MJ, Serwadda D, Gray RH, et al. Trends in HIV-1 prevalence may not reflect trends in incidence in mature epidemics: data from the Rakai population-based cohort, Uganda. AIDS. 1997;11:1023-1030.
3. Watson-Jones D, Baisley K, Weiss HA, et al. Risk factors for HIV incidence in women participating in an HSV suppressive treatment trial in Tanzania. AIDS. 2009;23:415-422.
4. Hallett TB, Stover J, Mishra V, et al. Estimates of HIV incidence from household-based prevalence surveys. AIDS. 2010;24:147-152.
5. Kwesigabo G, Killewo J, Urassa W, et al. HIV-1 infection prevalence and incidence trends in areas of contrasting levels of infection in the Kagera region, Tanzania, 1987-2000. J Acquir Immune Defic Syndr. 2005;40:585-591.
6. Wambura M, Urassa M, Isingo R, et al. HIV prevalence and incidence in rural Tanzania: results from 10 years of follow-up in an open-cohort study. J Acquir Immune Defic Syndr. 2007;46:616-623.
7. Arroyo MA, Hoelscher M, Sateren W, et al. HIV-1 diversity and prevalence differ between urban and rural areas in the Mbeya region of Tanzania. AIDS. 2005;19:1517-1524.
8. Tanzania Commission for AIDS (TACAIDS), National Bureau of Statistics (NBS), ORC Macro. Tanzania HIV/AIDS Indicator Survey 2003-04. Calverton, MD: TACAIDS, NBS, and ORC Macro; 2005.
9. Hallett TB, Zaba B, Todd J, et al. Estimating incidence from prevalence in generalised HIV epidemics: methods and validation. PLoS Med. 2008;5:e80.
10. Buchbinder SP, Mehrotra DV, Duerr A, et al. Efficacy assessment of a cell-mediated immunity HIV-1 vaccine (the Step Study): a double-blind, randomised, placebo-controlled, test-of-concept trial. Lancet. 2008;372:1881-1893.
11. Gregson S, Nyamukapa CA, Garnett GP, et al. Sexual mixing patterns and sex-differentials in teenage exposure to HIV infection in rural Zimbabwe. Lancet. 2002;359:1896-1903.
12. Pettifor AE, Hudgens MG, Levandowski BA, et al. Highly efficient HIV transmission to young women in South Africa. AIDS. 2007;21:861-865.
13. Myer L, Kuhn L, Stein ZA, et al. Intravaginal practices, bacterial vaginosis, and women's susceptibility to HIV infection: epidemiological evidence and biological mechanisms. Lancet Infect Dis. 2005;5:786-794.
14. Mwaluko G, Urassa M, Isingo R, et al. Trends in HIV and sexual behaviour in a longitudinal study in a rural population in Tanzania, 1994-2000. AIDS. 2003;17:2645-2651.
15. Shaffer DN, Ngetich IK, Bautista CT, et al. HIV-1 incidence rates and risk factors in agricultural workers and dependents in rural Kenya: 36-month follow-up of the Kericho HIV cohort study. J Acquir Immune Defic Syndr. 2010;53:514-521.
16. Baeten JM, Richardson BA, Martin HL Jr, et al. Trends in HIV-1 incidence in a cohort of prostitutes in Kenya: implications for HIV-1 vaccine efficacy trials. J Acquir Immune Defic Syndr. 2000;24:458-464.
17. Fisher JC, Bang H, Kapiga SH. The association between HIV infection and alcohol use: a systematic review and meta-analysis of African studies. Sex Transm Dis. 2007;34:856-863.
18. Zablotska IB, Gray RH, Koenig MA, et al. Alcohol use, intimate partner violence, sexual coercion and HIV among women aged 15-24 in Rakai, Uganda. AIDS Behav. 2009;13:225-233.
19. Zablotska IB, Gray RH, Serwadda D, et al. Alcohol use before sex and HIV acquisition: a longitudinal study in Rakai, Uganda. AIDS. 2006;20:1191-1196.
20. Hoffmann O, Zaba B, Wolff B, et al. Methodological lessons from a cohort study of high risk women in Tanzania. Sex Transm Infect. 2004;80(Suppl 2):ii69-ii73.
21. Riedner G, Hoffmann O, Rusizoka M, et al. Decline in sexually transmitted infection prevalence and HIV incidence in female barworkers attending prevention and care services in Mbeya Region, Tanzania. AIDS. 2006;20:609-615.
22. Quigley MA, Morgan D, Malamba SS, et al. Case-control study of risk factors for incident HIV infection in rural Uganda. J Acquir Immune Defic Syndr. 2000;23:418-425.
23. Landman KZ, Ostermann J, Crump JA, et al. Gender differences in the risk of HIV infection among persons reporting abstinence, monogamy, and multiple sexual partners in northern Tanzania. PLoS ONE. 2008;3:e3075.
24. Halperin DT, Epstein H. Concurrent sexual partnerships help to explain Africa's high HIV prevalence: implications for prevention. Lancet. 2004;364:4-6.
25. Morris M, Kretzschmar M. Concurrent partnerships and the spread of HIV. AIDS. 1997;11:641-648.
26. Smith J, Nalagoda F, Wawer MJ, et al. Education attainment as a predictor of HIV risk in rural Uganda: results from a population-based study. Int J STD AIDS. 1999;10:452-459.
27. Hargreaves JR, Glynn JR. Educational attainment and HIV-1 infection in developing countries: a systematic review. Trop Med Int Health. 2002;7:489-498.
28. Kilian AH, Gregson S, Ndyanabangi B, et al. Reductions in risk behaviour provide the most consistent explanation for declining HIV-1 prevalence in Uganda. AIDS. 1999;13:391-398.
29. Bradley H, Bedada A, Brahmbhatt H, et al. Educational attainment and HIV status among Ethiopian voluntary counseling and testing clients. AIDS Behav. 2007;11:736-742.
30. Michelo C, Sandoy IF, Fylkesnes K. Marked HIV prevalence declines in higher educated young people: evidence from population-based surveys (1995-2003) in Zambia. AIDS. 2006;20:1031-1038.
31. Fylkesnes K, Musonda RM, Kasumba K, et al. The HIV epidemic in Zambia: socio-demographic prevalence patterns and indications of trends among childbearing women. AIDS. 1997;11:339-345.
32. Seeley JA, Malamba SS, Nunn AJ, et al. Socioeconomic status, gender, and risk of HIV-1 infection in a rural community in south west Uganda. Med Anthropol Q. 1994;8:78-89.
33. Garner RC. Safe sects? Dynamic religion and AIDS in South Africa. J Mod Afr Stud. 2000;38:41-69.
34. Gray RH, Kigozi G, Serwadda D, et al. Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet. 2007;369:657-666.
35. Siegfried N, Muller M, Deeks J, et al. HIV and male circumcision--a systematic review with assessment of the quality of studies. Lancet Infect Dis. 2005;5:165-173.
36. Ghanem KG, Hutton HE, Zenilman JM, et al. Audio computer assisted self interview and face to face interview modes in assessing response bias among STD clinic patients. Sex Transm Infect. 2005;81:421-425.

HIV-1; incidence; risk factors; cohort study; Tanzania; Africa

© 2011 Lippincott Williams & Wilkins, Inc.