The HIV prevalence was lower in men for both viruses, the female:male age-adjusted prevalence ratios being 1.45 (95% CI, 0.99–2.12) for HIV-1 and 2.00 (95% CI, 1.33–3.02) for HIV-2.
Changes in HIV prevalence from 1996 to 2006
We observed an increase in the overall prevalence of HIV-1 from 2.3% (54/2301) in 1996 to 4.6% (118/2548) in 2006 with an age and sex adjusted 2006/1996 prevalence ratio of 1.90 (Table 2). The single HIV-1 infection prevalence increased threefold, whereas the prevalence of dual infections, HIV 1+2, declined (prevalence ratio = 0.46; 95% CI, 0.23–0.93).
HIV-1 prevalence increased more for women than men over this 10-year time period. Adjusted for age the difference was not significant (test for interaction, P = 0.54). Among young adults aged 15–24 years the prevalence was low at 1.5% for men and 1.8% for women. In the older age groups, the largest increase was among women aged 25–34 years and among men aged 35–44 years. The prevalence was around 9% in these groups in 2006.
The HIV-2 prevalence decreased from 7.4% in 1996 to 4.4% in 2006 (Table 3). HIV-2 declined significantly below 45 years for both men and women. No decline was found for people older than 45, the prevalence ratio being 0.91 for men and 0.82 for women in this age group. Measured over the whole study period from 1987 to 2006, the prevalence ratio was 0.46 for men and 0.99 for women in this age group. The prevalence is therefore still very high for older women, 17.6%.
Our study design implied that some subjects were included in both surveys in 1996 and 2006. This could possibly distort the comparison of the prevalence in the two periods. Comparing only newly included subjects in 1996 and 2006 made only small changes to the results. Hence, little bias seemed to be created by including all subjects in the analyses.
HIV incidence in 1996 to 2006
In the 1987 survey there were no cases of HIV-1 infection . In the present study, 650 (278 men, 372 women) of the participants who were HIV-1-seronegative in 1996 survey were reexamined in 2006 contributing a total of 6058 person-years of observation (PYO). Thirty had become HIV-1-seropositive (8 men, 22 women), giving an overall incidence rate of HIV-1 infection of 0.50 per 100 person-years (Table 4). The incidence rate was lower in men than in women, the female/male incidence rate ratio (IRR) being 2.08.
The HIV-1 incidence rate was highest in the 25–34-year age group, followed by the 15–24-year age group. The HIV-1 incidence declined with age in the groups from 25–34 to 55+ years. The HIV-1 incidence was significantly higher in people below 45 years of age than people older than 45 years, the IRR being 0.29.
Six hundred and twelve (268 men, 344 women) of the participants who were HIV-2-seronegative in 1996 survey were reexamined in 2006. The follow-up time was 5770 person-years. Fourteen had become HIV-2-seropositive (3 men and 11 women), giving an overall incidence of HIV-2 infection of 0.24 per 100 person-years, being lower in men than in women (Table 4). The female-male IRR was 2.87. The IRR between the age groups 45+ and 15–44 was 1.16.
HIV 1+2 incidence
Six hundred and fifty four (279 men and 375 women) of the participants, who were negative for dual infection in the 1996 survey, were reexamined in 2006, contributing a total of 6222 person-years. Four female participants had become HIV 1+2-seropositive, giving an incidence of HIV 1+2 infections of 0.06 per 100 person-years. One participant was HIV-2-positive in the last survey and became HIV dual infected, and the other three participants were HIV-negative in the 1996 survey and seroconverted to both HIV-1 and HIV-2 infection in the 10-year period.
Changes in HIV incidence between two periods: 1987–1996 and 1996–2006
There was a nonsignificant increase in the incidence of HIV-1 between the first period (1987–1996) and the second period (1996–2006). The incidence was 0.34 per 100 PYO in the first period  and 0.50 per 100 PYO in the second period, the IRR for the two periods being 1.46 (95% CI, 0.78–2.76). For people less than 44 years old, the incidence of HIV-1 increased between the two periods being 0.37 per 100 PYO and 0.61 per 100 PYO respectively, the IRR being 1.65 (95% CI, 0.85–3.20). For people older than 45 the incidence rates were 0.22 and 0.18 respectively. For men the HIV-1 incidence rate decreased from 0.45 per 100 PYO in the first period to 0.31 per 100 PYO in the second period, with an IRR of 0.68 (95% CI, 0.26–1.82). For women the HIV-1 incidence rate increased from 0.25 per 100 PYO to 0.64 per 100 PYO, the IRR being 2.51 (95% CI, 1.02–6.19). The difference between men and women was borderline significant, P = 0.05. For women younger than 45 the incidence rate increased from 0.30 per 100 PYO to 0.76 per 100 PYO, the IRR being 2.58 (95% CI, 1.03–6.47).
There was a significant decline in the HIV-2 overall incidence from 0.53 per 100 PYO in the first period to 0.24 per 100 PYO in the second period, the IRR being 0.45 (95% CI, 0.23–0.90). For men, the HIV-2 incidence declined from 0.31 per 100 PYO to 0.12 per 100 PYO, the IRR being 0.39 (95% CI, 0.09–1.61). For the women, the HIV-2 incidence had the same pattern, decreasing from 0.71 per 100 PYO to 0.34 per 100 PYO, the IRR being 0.48 (95% CI, 0.22–1.04).
Studies of pregnant women form the predominant source of data on the HIV-1 prevalence in different African countries. However, they are less reliable for generating population-level estimates [19,20] due to the sex bias and the bias associated with studying only a fertile population . Therefore, community studies with extensive follow-up can offer a more accurate overview of the dynamics of the spread of the HIV infection in different strata and age groups in a population.
The Bandim Health Project carried out the first community study of HIV-2 covering the whole population. In the initial survey in 1987  children were included but the prevalence was so low that it was subsequently decided to restrict the study to adults. Over the years, we have tried to maintain the same methodology representatively covering around 10% of the houses in the study area. The number of houses included in the study has therefore been gradually enlarged. Although the participation rate in 2006 was slightly lower (78.6%) compared with the 1987 (86.0%)  and 1996 (85.0%)  serosurveys carried out in the Bandim study area, the patterns of participation were the same. As in previous studies, men were more likely to be absent at the time of the interview at home due to school or work attendance in all age groups. With improvement in transport traveling has become more common, and absence due to traveling was an important reason for nonparticipation. We tested whether the individuals who had been most difficult to reach had a higher prevalence of HIV. If so we might have underestimated the prevalence due to the lower participation rate. However, there was no indication of this being the case (data available on request). Hence, it seems unlikely that the lower participation rate has seriously affected our observations.
Our study  has documented the continuation of the trends we first observed in 1996, namely an increase in the HIV-1 prevalence and a constant decline in the prevalence of HIV-2 in Guinea-Bissau. The prevalence of HIV-1 has increased two-fold globally in the past 10 years and this augmentation occurred in all age groups and in both sexes. The 25–34-year age group was the most affected in women whereas in men the highest prevalence was observed in the age group 10 years older (35–44).
A similar pattern of increasing prevalence rates of HIV-1 and declining prevalence rates of HIV-2 were also reported in a cohort of police officers from Guinea-Bissau in 1997  and among pregnant women attending the National Hospital in Bissau between 1987 and 2004 . This trend has also been observed in Gambia  and other West African countries .
Currently in West Africa only Côte d'Ivoire has a national adult HIV prevalence that surpasses 4% (no distinction was made between HIV-1 and HIV-2) . In Nigeria about 4.4% of the women attending antenatal clinics nationally were found to be infected with HIV . Thus, Côte d'Ivoire, Nigeria and Mali are the countries with the most serious epidemic in West Africa as far as documented in studies. In contrast to neighboring countries (Senegal and Guinea-Conakry) with a low HIV-1 prevalence (<1 and 1.5% respectively) , Guinea-Bissau has a prevalence that may reach the levels of the other most affected countries in the sub-region. We have only data from the capital, but the HIV-1 prevalence was as high as 4.8% in women giving birth at the National Hospital  and 4.6% for both sexes in our urban cohort.
The present study also documented that the overall HIV-1 incidence increased from the 1987–1996 period to the 1996–2006 period but not significantly. The HIV-1 incidence rate as expected was higher in people younger than 35 years of age.
Generally there was more HIV infection among women. For both HIV-1 and HIV-2 the prevalence and incidence were higher in women, and HIV-1 prevalence increased more in women. Interestingly the HIV-1 incidence increased for women but declined for men. Between the two periods the HIV-1 incidence more than doubled for women. Higher HIV-1 seroprevalence and incidence in women than in men have previously been reported from others parts of Africa .
While HIV-1 infection has increased rapidly over the last decade, the HIV-2 infection has an opposite trend. The decline in the prevalence of HIV-2 that we noted in 1996  has continued. The overall HIV-2 prevalence among adults had declined from 8.9% in 1987  to 7.4% in 1996  and 4.4% in 2006. Initially this decline was only for men, now the prevalence is declining in both sexes. Regarding incidence, we observed that the HIV-2 incidence rate between 1996 and 2006 was half the incidence in the preceding 10-year period. Hence, the decrease in the HIV-2 risk was genuine and not only due to mortality or migration from a previously established cohort.
Nevertheless, it should be noted that the prevalence of HIV-2 is not declining among older people. This may indicate a cohort effect, previously infected young adults coming of age due to the longer survival of HIV-2 infected individuals. Furthermore, older women continue to become infected (Table 4). Hence, HIV-2 may continue as an independent infection among the oldest for some time, especially in women. An increased susceptibility to HIV and human T cell lymphotropic virus (HTLV) infections in older women, possibly due to changes in vaginal mucosal immunity, has previously been suggested to be the cause of this pattern [15,26,27].
The reasons why the HIV-2 is disappearing are probably complex and could be related to different factors. First of all, screening of blood donors for HIV-1 and HIV-2 was introduced in Bissau in March 1987, so the spread of HIV through blood transfusion was mainly prevented from that date. In the 1996 survey, transfusions were no longer a general risk factor for HIV-2 transmission . This change may have been very important for transmission of HIV-2 as around 20% of blood donors were infected when screening was initiated in 1987. Second, HIV-2 is probably not very effective as a sexually transmitted disease [28,29]. We have previously found that women married to an HIV-2-infected man only became infected after 45 years of age .
In conclusion, the HIV epidemic in Guinea-Bissau is still evolving and further studies are needed in order to monitor the trends. Understanding the behavioral patterns of older people and providing practical prevention guidelines based on national data will help the National HIV/AIDS Control Program to control the situation. Information, counseling and voluntary HIV testing programmes should be reinforced in all age groups in order to stop the further spread of HIV-1 epidemic in the country.
We are thankful to the laboratory technicians at the National Public Health Laboratory, field assistants and office staff at the Bandim DSS site for making this study possible. The study was supported by grants from the Danish International Development Agency (DANIDA)/ENRECA and Swedish Agency for Research Cooperation with Developing Countries (SAREC).
Contributions: The study was planned by Z. J. da Silva and P. Aaby and executed by Z. J. da Silva. A. Rodrigues and I. Oliveira were involved with supervision and interpretation of data. A. Andersen was responsible for the statistical analyses. B. Holmgren had carried out the previous cohort studies. S. Andersson and F. Dias have been responsible for the laboratory testing strategies. Z. J. da Silva wrote the first draft and all authors contributed to the final version of the paper.
There are no conflicts of interest.
1. UNAIDS/WHO. AIDS epidemic update: special report on HIV/AIDS. Geneva, Switzerland. December 2006. Available at http://www.unaids.org
2. Bouckenooghe A, Shandera W. HIV trends in African blood donors. J Infect 1999; 39:122–128.
3. Meda N, Ndoye I, M'Boup S, Wade A, Ndiaye S, Niang C, et al
. Low and stable HIV infection rates in Senegal: natural course of the epidemic or evidence for success of prevention? AIDS 1999; 13:1397–1405.
4. Schim van der Loeff MF, Sarge-Njie R, Ceesay S, Awasana AA, Jaye P, Sam O, et al
. Regional differences in HIV trends in The Gambia: results from sentinel surveillance among pregnant women. AIDS 2003; 17:1841–1846.
5. Poulsen AG, Aaby P, Gottschau A, Kvinesdal BB, Dias F, Molbak K, Lauritzen E. HIV-2 infection in Bissau, West Africa, 1987–1989: incidence, prevalences, and routes of transmission. J Acquir Immune Defic Syndr 1993; 6:941–948.
6. Poulsen AG, Kvinesdal B, Aaby P, Molbak K, Frederiksen K, Dias F, Lauritzen E. Prevalence of and mortality from human immunodeficiency virus type 2 in Bissau, West Africa. Lancet 1989; 1:827–831.
7. Wilkins A, Ricard D, Todd J, Whittle H, Dias F, Paulo Da Silva A. The epidemiology of HIV infection in a rural area of Guinea-Bissau. AIDS 1993; 7:1119–1122.
8. Poulsen AG, Kvinesdal BB, Aaby P, Lisse IM, Gottschau A, Molbak K, et al
. Lack of evidence of vertical transmission of human immunodeficiency virus type 2 in a sample of the general population in Bissau. J Acquir Immune Defic Syndr 1992; 5:25–30.
9. Larsen O, da Silva Z, Sandstrom A, Andersen PK, Andersson S, Poulsen AG, et al
. Declining HIV-2 prevalence and incidence among men in a community study from Guinea-Bissau. AIDS 1998; 12:1707–1714.
10. Larsen O. Epidemiology of Retroviral Infection in Guinea-Bissau, cross-sectional and longitudinal community studies [PhD Thesis]. University of Copenhagen; 1998.
11. da Silva ZJ. Trends in HIV infection in Guinea-Bissau. Sentinel Surveillance of pregnant women [Master Thesis]. Umea University; 2002.
12. Poulsen AG, Aaby P, Larsen O, Jensen H, Naucler A, Lisse IM, et al
. 9-year HIV-2-associated mortality in an urban community in Bissau, West Africa. Lancet 1997; 349:911–914.
13. Larsen O, Andersson S, da Silva Z, Hedegaard K, Sandstrom A, Naucler A, et al
. Prevalences of HTLV-1 infection and associated risk determinants in an urban population in Guinea-Bissau, West Africa. J Acquir Immune Defic Syndr 2000; 25:157–163.
14. Holmgren B, da Silva Z, Larsen O, Vastrup P, Andersson S, Aaby P. Dual infections with HIV-1, HIV-2 and HTLV-I are more common in older women than in men in Guinea-Bissau. AIDS 2003; 17:241–253.
15. Holmgren B, Aaby P, Jensen H, Larsen O, da Silva Z, Lisse IM. Increased prevalence of retrovirus infections among older women in Africa. Scand J Infect Dis 1999; 31:459–466.
16. Interim proposal for a WHO Staging System for HIV infection and Disease.Wkly Epidemiol Rec
17. Andersson S, da Silva Z, Norrgren H, Dias F, Biberfeld G. Field evaluation of alternative testing strategies for diagnosis and differentiation of HIV-1 and HIV-2 infections in an HIV-1 and HIV-2-prevalent area. AIDS 1997; 11:1815–1822.
18. Aaby P, Ariyoshi K, Buckner M, Jensen H, Berry N, Wilkins A, et al
. Age of wife as a major determinant of male-to-female transmission of HIV-2 infection: a community study from rural West Africa. AIDS 1996; 10:1585–1590.
19. Fylkesnes K, Ndhlovu Z, Kasumba K, Mubanga Musonda R, Sichone M. Studying dynamics of the HIV epidemic: population-based data compared with sentinel surveillance in Zambia. AIDS 1998; 12:1227–1234.
20. Kigadye RM, Klokke A, Nicoll A, Nyamuryekung'e KM, Borgdorff M, Barongo L, et al
. Sentinel surveillance for HIV-1 among pregnant women in a developing country: 3 years' experience and comparison with a population serosurvey. AIDS 1993; 7:849–855.
21. Norrgren H, Andersson S, Biague AJ, da Silva ZJ, Dias F, Naucler A, Biberfeld G. Trends and interaction of HIV-1 and HIV-2 in Guinea-Bissau, West Africa: no protection of HIV-2 against HIV-1 infection. AIDS 1999; 13:701–707.
22. Mansson F, Alves A, da Silva ZJ, Dias F, Andersson S, Biberfeld G, et al
. Trends of HIV-1 and HIV-2 prevalence among pregnant women in Guinea-Bissau, West Africa: possible effect of the civil war 1998–1999. Sex Transm Infect 2007; 83:463–467.
23. van der Loeff MF, Awasana AA, Sarge-Njie R, van der Sande M, Jaye A, Sabally S, et al
. Sixteen years of HIV surveillance in a West African research clinic reveals divergent epidemic trends of HIV-1 and HIV-2. Int J Epidemiol 2006; 35:1322–1328.
24. 2005 National HIV/syphilis seroprevalence sentinel survey among pregnant women attending antenatal clinics:technical report. In: Abuja, Nigeria: Federal Ministry of Health Nigeria; 2006.
25. Boerma JT, Ghys PD, Walker N. Estimates of HIV-1 prevalence from national population-based surveys as a new gold standard. Lancet 2003; 362:1929–1931.
26. White HD, Yeaman GR, Givan AL, Wira CR. Mucosal immunity in the human female reproductive tract: cytotoxic T lymphocyte function in the cervix and vagina of premenopausal and postmenopausal women. Am J Reprod Immunol 1997; 37:30–38.
27. Wira CR, Rossoll RM. Antigen-presenting cells in the female reproductive tract: influence of sex hormones on antigen presentation in the vagina. Immunology 1995; 84:505–508.
28. De Cock KM, Adjorlolo G, Ekpini E, Sibailly T, Kouadio J, Maran M, et al
. Epidemiology and transmission of HIV-2. Why there is no HIV-2 pandemic? JAMA 1993; 270:2083–2086.
29. Kanki PJ, Travers KU, MBoup S, Hsieh CC, Marlink RG, Gueye NA, et al
. Slower heterosexual spread of HIV-2 than HIV-1. Lancet 1994; 343:943–946.
Keywords:© 2008 Lippincott Williams & Wilkins, Inc.
community study; Guinea-Bissau; HIV-1; HIV-2; incidence; prevalence