Sexual transmission of HIV between men who have sex with men (MSM) is still a predominant route of HIV transmission in Western World settings. Several studies indicate that the risk of sexual transmission of the virus is almost negligible from HIV patients with low viral load (VL) in vaginal and conceivably also in anal sex.1–3 This has led to the hypothesis that highly active antiretroviral therapy (HAART) reduces not only morbidity and mortality among persons with HIV but also reduces the risk of sexual transmission substantially.4,5 A Taiwanese study from 2004 showed declining HIV transmission rates after introduction of HAART,6 and in 2 recent studies from Canada and San Francisco, declining community VL levels have been linked with reduced transmission of HIV.7,8
The number of HIV-infected MSM living in Denmark has increased substantially after the introduction of HAART, but the raising population of MSM with HIV has not transformed into an increased incidence of HIV diagnoses in this population. At the same time, reports on incidence of syphilis and sexual behavior in Danish MSM indicate that sexual risk behavior has increased among Danish MSM in the last decade.9,10 The paradox of stable incidence of HIV diagnosis despite increased HIV prevalence and unsafe sex could be explained by increased effectiveness of HAART, which would lead to a declining number of HIV-positive MSM capable of transmitting the virus sexually.
We used 3 unique Danish data sources to calculate prevalence and incidence of HIV, incidence of syphilis, and self-reported frequency of sexual risk behavior in Danish MSM. Furthermore, we calculated the number of MSM with HIV who had high VL and thereby were able to transmit HIV sexually.
Denmark has a population of 5.5 million, and the estimated prevalence of HIV infection in the total population is 0.1 %.11 It has been estimated that 500 undiagnosed HIV-positive MSM reside in Denmark, an estimate that is assumed to have been stable over the last 10 years.12
Denmark's tax-funded health care system provides antiretroviral treatment free-of-charge to all HIV-positive residents. Treatment of HIV infection is restricted to 8 specialized medical centers, where patients are seen on an outpatient basis at intended intervals of 12 weeks. During our study period, January 1995 to January 2010, national criteria for HAART initiation were any of the following: presence of an HIV-related disease, acute HIV infection, pregnancy, CD4 cell count <300 cells per microliter until May 2008 and <350 cells per microliter thereafter, and VL >100,000 HIV RNA copies per milliliter (until 2001).
The Danish HIV Cohort Study is a nationwide, prospective, population-based cohort study of all Danish HIV-infected patients treated at Danish hospitals since January 1, 1995.13 The data are updated yearly, and include demographics, route of infection, CD4 counts, VL, and antiretroviral treatment. The unique 10-digit Central Person Registration number assigned to all Danish citizens at birth or immigration is used to avoid multiple registrations and to track individuals.
The Sex Life Surveys 2000–2009
Since 2000, 5 Sex Life Surveys monitoring sexual behavior and responses to HIV issues among MSM in Denmark have been carried out. Data were collected on self-reported sexual behavior and HIV prevalence among MSM in Denmark. The surveys were carried out as internet-based questionnaires on several gay and HIV-related websites and paper questionnaires handed out during the annual Copenhagen Gay Pride event, in gay bars, clubs, and other venues and distributed as inserts together with prestamped envelopes in both gay and HIV-related journals. The questionnaires contained questions concerning sexual behavior during the last 12 months including frequency of anal sex, number of anal sex partners, frequency of unprotected anal intercourse (UAI), and number of UAI partners, stratified by partners' HIV status. Also included were questions on time and result of respondents' latest HIV test, and questions about other sexually transmitted infections. The surveys have been partly published previously on the internet.9,10
The National Syphilis Registration System was established in 1993 and registers all residents in Denmark diagnosed with syphilis. The notification to the register is mandatory and is done by the treating physician. Data include route and country of infection and result of latest HIV test.
Cohort Community Reproductive Rate
We defined Cohort Community Reproductive Rate (CCRR) as the number of newly HIV-infected MSM in a calendar year per HIV-infected MSM living in Denmark in that year. For the Danish MSM population, the number of individuals diagnosed for the first time with HIV and reporting MSM as route of HIV infection was used as a proxy for incidence of new infections, and the number of HIV-positive MSM alive, living in Denmark, and registered in the Danish HIV Cohort was used as proxy for the HIV-positive MSM population living in Denmark.
To explore the explanatory factors of CCRR, we used a Poisson regression model allowing for overdispersion. Both age and proportion of MSM in the cohort considered able to transmit HIV (VL > 400 HIV RNA copies/mL) were included as explanatory variables and included in the final model if they were statistically significant on a 5% level.
From “The Danish HIV Cohort Study,” we identified all MSM treated at the 8 HIV-treating clinics in Denmark. Patients were considered under care from date of first HIV-positive test, immigration, or January 1, 1995, which ever was last; until date of death, emigration, or January 1, 2010, which ever was first. For the period January 1, 1995, to January 1, 2010, we calculated the total number of MSM with undetectable VL (≤ 400 HIV RNA copies/mL) using the principle of last value carried forward. Patients were considered to have detectable VL until the first measurement of undetectable VL. The cut off of 400 HIV RNA copies per milliliter was used as it was the lower limit of detection in the initial period of the study.
From “The Sex Life Surveys,” we calculated the fractions of patients reporting having engaged in UAI stratified by HIV status of the respondent and of his partners, and self-reported HIV-testing behavior.9,10 Differences between groups were analyzed using χ2 tests and logistic regression analyses.
From “The National Syphilis Registration System,” we calculated the yearly number of new cases of syphilis stratified on HIV-positive and HIV-negative individuals.
The analyses were done using SPSS (SPSS Inc, Chicago, IL), and Stata version 11.2 MP (Stata Corporation, College Station, TX).
In the study period, the median number of MSM diagnosed with HIV and subsequently registered in The Danish HIV Cohort Study was 93 per year (range: 71–137) (Table 1). The yearly number of new HIV diagnoses decreased slightly during the late 1990s (Table 1), and thereafter increased vaguely and stabilized from 2005. The median CD4 count at diagnosis was stable from 1995 to 1999, but thereafter increased substantially until 2004 when it stabilized around 400 CD4 cells per milliliter (Table 1). The median age at diagnosis was 37.6 (range: 32.6–38.8) and did not change over calendar time (P = 0.15).
The number of HIV-positive MSM alive, living in Denmark, and registered in The Danish HIV Cohort Study increased from 1035 in 1995 to 1813 in 2010 (75%). Before this time, that is, in the pre-HAART era, the number of HIV-positive MSM alive was stable because the number of AIDS deaths balanced the number of new HIV cases. In the study period, the number of MSM with detectable VL (VL > 400 HIV RNA copies/mL) decreased by 75% from 1035 in 1995 to 262 in 2010 (Table 1), as HAART treatment improved the immune status of the cohort members. The median age of the cohort increased steadily with 0.53 [95% confidence interval (CI): 0.51 to 0.55] years per calendar year from 39 years in 1995 to 47 years in 2010, also an effect of HAART treatment that keeps cohort members alive.
In 1995, CCRR was 0.099 (95% CI: 0.092 to 0.108). CCRR decreased significantly through the late 1990s, and stabilized at 0.071 (95% CI: 0.065 to 0.079) from 2005 and onwards (Fig. 1).
The proportion of MSM in The Danish HIV Cohort Study with a VL >400 HIV RNA copies per milliliter followed the decreasing CCRR until CCRR leveled off in 2005 (Fig. 1). Thereafter, the proportion of MSM in The Danish HIV Cohort Study with VL >400 HIV RNA copies per milliliter continued to decrease without any further changes in CCRR. The increasing age in the cohort had no significant association with CCRR in the study period (P = 0.53).
During the study period, the Sex Life Surveys showed a trend toward increasing unsafe sex (Table 2). Respondents reported an increasing yearly number of anal sex partners (P < 0.01) and increased frequency of UAI (P < 0.01) also with partners of unknown HIV serostatus (P < 0.01). This trend was especially marked in the HIV-infected MSM population (Fig. 2) compared with the HIV-negative MSM (data not shown). In the “Sex Life Surveys,” we observed no trend in the proportion of HIV-positive respondents over calendar time (Table 2). The proportion of MSM who reported to have been ever tested for HIV did not change over calendar time, but we observed an increasing trend in the proportion not already diagnosed with HIV, who were recently tested (P < 0.01) (Table 2).
In The National Syphilis Registration System, the number of MSM diagnosed with syphilis increased substantially in the study period from 2 patients diagnosed in 1995 to 208 in 2009, and the increase was observed for HIV-positive MSM and in recent years even more for the HIV-negative MSM (Fig. 3).
In this nation-wide population-based study, we demonstrate that although unsafe sex among MSM has increased substantially and the number of HIV-positive MSM living in Denmark has enlarged, the incidence of HIV diagnoses in this population has remained almost stable for more than a decade. Our findings indicate that this paradox is due to effective antiretroviral therapy and not increased awareness of safe sex. Transmission of HIV among Danish MSM seems to be sustained by individuals not on HAART.
The major strength of the study is the nation-wide design with access to national registries of HIV and syphilis, VL tests, and estimates of changes in sexual behavior for the MSM population.
Because the Danish HIV Cohort population is closely followed and very well described, the data used in the CCRR, as a measure of transmission rate from the cohort of MSM diagnosed with HIV in Denmark, are very reliable. This concept differs from other methods using national whole population surveillance data6 and also from the “community VL” concept as that refers to the “capacity for” transmission.8
Our study has some weaknesses. Obviously, we could not include undiagnosed HIV-infected in our analyses and therefore could not calculate the true incidence of HIV in Danish MSM. The official estimate of undiagnosed MSM with HIV in Denmark is in the order of 500 individuals, and it is believed to have been fairly constant over the study period.12 This limitation is not specific to the present study and also involves studies using “community VL”.8 We estimated CCRR not from the number of newly HIV-infected MSM (true incidence) and total number of HIV-infected individuals living in Denmark (true prevalence), but from number of newly diagnosed MSM and number of MSM registered in The Danish HIV Cohort. We thereby underestimated both parameters; but as CCRR is calculated as a rate, the underestimation of both parameters will tend to neutralize this bias. Also our model does not take into consideration that some Danish MSM are infected abroad. However, unlike the heterosexual transmissions, Danish MSM diagnosed with HIV mainly report to have been infected in Denmark.14 Even though the proportion of MSM who report having been infected abroad has gone up from 12% to 20% in the last 10 years (Susan Cowan, MD, written communication, 2012), it is still not high enough to determine the incidence. Finally, we do not perform incidence testing (eg, Recent Infection Testing Algorithm) in Denmark, and we cannot exclude a recent increase in incidence of HIV among Danish MSM, but the stable median CD4 count in newly diagnosed MSM justified the assumption that the incidence of HIV in the MSM population has been stable also in the last part of the study period.
In the early HAART era (1995–2000), the yearly number of MSM diagnosed with HIV decreased slightly, which may be a proxy for a decreasing incidence of HIV in MSM in the years before introduction of HAART (Table 1). Also the median CD4 count at diagnosis was stable which indicates that time from HIV infection to diagnosis did not change. In this period, the CCRR dropped proportional to the reduction in HIV-positive MSM able to sexually transmit HIV infection, which underlines the effect of HAART (Fig. 1). No Sex Life Surveys were carried out before 2000, but the syphilis incidence was very low among MSM in this period indicating that less unsafe sex may have lead to a lower incidence of HIV.
In the period 2001–2004, the number of MSM diagnosed with HIV increased slightly, but so did the median CD4 count at diagnosis, which is compatible with increased HIV testing leading to earlier diagnosis of the disease.15 In this period, the CCRR continued to decrease, as did the proportion of MSM in The Danish HIV Cohort Study with VL >400 HIV RNA copies per milliliter (Fig. 1). The Sex Life Surveys showed increased frequency of sexual risk behavior, and the effect of HAART is the most reasonable explanation for the decreased CCRR in this period.
Finally, incidence of HIV in MSM and median CD4 count at diagnosis were stable from 2005 and onwards. The proportion of MSM in The Danish HIV Cohort Study with VL >400 HIV RNA copies per milliliter continued to decrease but was not associated with changes in CCRR (Fig. 1).
We observed that the CCRR in Danish MSM decreased parallel to decreasing proportion of MSM in The Danish HIV Cohort Study with VL >400 HIV RNA copies per milliliter up to the point where less than 400 MSM in the cohort had VL >400 HIV RNA copies per milliliter. This suggests that after this point, the ongoing transmission of HIV among MSM in Denmark is to a large extent driven by HIV-positive MSM not yet diagnosed with HIV.
Earlier models have suggested that there may be a balancing point at which increasing levels of high risk sexual behavior performed by a large number of yet undiagnosed/untreated HIV-positive MSM offsets the effect of early/regular HIV testing among sexually active MSM and subsequent treatment of those diagnosed as HIV positive.16,17 This point seems to have been reached among MSM in Denmark in the early 2000.
According to this latter hypothesis, the balancing point where the benefit of HAART on HIV transmission is counteracted by sexual risk behavior does not depend on the frequency of unsafe sex among MSM in general, but on the level of unsafe sex among MSM not recently HIV tested. A major study based on mathematical modelling from United States has suggested that HIV-positive individuals unaware of their HIV status are responsible for the majority of HIV transmissions.18
The sex Life Surveys have demonstrated that the frequency of sexual risk behavior among MSM in Denmark has increased markedly during the last 10 years and an increasing proportion, especially among HIV-positive MSM reported to have engaged in UAI with partners of unknown HIV status.9,10 Thus serosorting among HIV-positive MSM, which has been put forward as an explanation for equivalent trends seen in San Francisco,19 can be dismissed as the explanation for the stable HIV incidence despite the increase in sexual risk behavior among MSM in Denmark. The rise in syphilis incidence, which has recently been most apparent among HIV-negative MSM, suggests that the notion among HIV-negative MSM that they are not having unsafe sex is not justified.
Other studies from Western countries have demonstrated declining HIV transmission rates after introduction of HAART.6–8 In San Francisco, the HIV incidence among MSM have stabilized probably as a result of effective HAART.20 In contrast, some studies have shown an increasing number of MSM diagnosed with HIV.21–26
Treating HIV-positive MSM immediately after HIV diagnosis would further decrease the pool of HIV-positive MSM able to transmit HIV, but assuming that the undiagnosed HIV-positive MSM constitute the major proportion of possible transmitters, that will not affect transmission from this group. Increasing HIV testing among MSM would probably lead to a significant decrease in HIV transmissions. This is supported by the finding that large phylogenetic clusters of HIV positive in Denmark are mainly found among Danish MSM.27 The Danish recommendations for HIV testing have recently been updated to allow for enhanced testing,28 but it is too early to determine whether testing rates will increase among the at-risk populations and eventually will have an impact on transmission rates.
The present study suggests that successful implementation of HAART has major impact on HIV incidence among MSM. Additional measures to diminish the pool of MSM who are at risk of transmitting HIV should focus both on earlier initiation of HAART and enhanced testing, especially of MSM engaged in sexual risk behavior.
1. Quinn TC, Wawer MJ, Sewankambo N, et al.. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med. 2000;342:921–929.
2. Attia S, Egger M, Müller M, et al.. Sexual transmission of HIV
according to viral load and antiretroviral therapy: systematic review and meta-analysis. AIDS. 2009;23:1397–1404.
3. Donnell D, Baeten JM, Kiarie J, et al.. Heterosexual HIV
-1 transmission after initiation of antiretroviral therapy: a prospective cohort analysis. Lancet. 2010;375:2092–2098.
4. Granich RM, Gilks CF, Dye C, et al.. Universal voluntary HIV
testing with immediate antiretroviral therapy as a strategy for elimination of HIV
transmission: a mathematical model. Lancet. 2009;373:48–57.
5. Cohen MS, Chen YQ, McCauley M, et al.. Prevention of HIV
-1 infection with early antiretroviral therapy. N Engl J Med. 2011;365:493–505.
6. Fang C-T, Hsu H-M, Twu S-J, et al.. Decreased HIV
transmission after a policy of providing free access to highly active antiretroviral therapy in Taiwan. J Infect Dis. 2004;190:879–885.
7. Montaner JS, Lima VD, Barrios R, et al.. Association of highly active antiretroviral therapy coverage, population viral load, and yearly new HIV
diagnoses in British Columbia, Canada: a population-based study. Lancet. 2010;376:532–539.
8. Das M, Chu PL, Santos G-M, et al.. Decreases in community viral load are accompanied by reductions in new HIV
infections in San Francisco. PLoS One. 2010;5:e11068.
9. Cowan SA, Haff J. HIV
and risk behaviour among men who have sex with men in Denmark—the 2006 Sex Life Survey. Euro Surveill. 2008;13:19050. Available at: http://www.ncbi.nlm.nih.gov/pubmed/19040825
. Accessed January 10, 2012.
10. Haff J, Cowan S. The 2009 Sex Life Survey. 2009. Available at: http://www.ssi.dk/Service/Sygdomsleksikon/H/∼/media/839BC82408B44489B619AFF0382AC074.ashx
. Accessed January 10, 2012.
11. Smith E. HIV
/AIDS surveillance in Denmark: the challenges ahead. J Acquir Immune Defic Syndr. 2003;32(suppl 1):S33–S38.
12. Hamers F, Phillips A. Diagnosed and undiagnosed HIV
-infected populations in Europe. HIV
Med. 2008;9(suppl 2):6–12.
13. Lohse N, Hansen A-BE, Jensen-Fangel S, et al.. Demographics of HIV
-1 infection in Denmark: results from the Danish HIV
Cohort Study. Scand J Infect Dis. 2005;37:338–343.
14. Cowan SA, Smith E. Incidence of HIV
/AIDS in Denmark, 1990–2005 [article in Danish]. Ugeskr Laeger. 2006;168:2247–2252.
15. Dougan S, Elford J, Chadborn TR, et al.. Does the recent increase in HIV
diagnoses among men who have sex with men in the UK reflect a rise in HIV
incidence or increased uptake of HIV
testing? Sex Transm Infect. 2007;83:120–125; Discussion 125.
16. Blower SM, Gershengorn HB, Grant RM. A tale of two futures: HIV
and antiretroviral therapy in San Francisco. Science. 2000;287:650–654.
17. Law MG, Prestage G, Grulich A, et al.. Modelling the effect of combination antiretroviral treatments on HIV
incidence. AIDS. 2001;15:1287–1294.
18. Marks G, Crepaz N, Janssen RS. Estimating sexual transmission of HIV
from persons aware and unaware that they are infected with the virus in the USA. AIDS. 2006;20:1447–1450.
19. Truong HM, Truong H-HM, Kellogg T, et al.. Increases in sexually transmitted infections and sexual risk behaviour without a concurrent increase in HIV
incidence among men who have sex with men in San Francisco: a suggestion of HIV
serosorting? Sex Transm Infect. 2006;82:461–466.
20. Scheer S, Kellogg T, Klausner JD, et al.. HIV
is hyperendemic among men who have sex with men in San Francisco: 10-year trends in HIV
prevalence, sexually transmitted infections and sexual risk behaviour. Sex Transm Infect. 2008;84:493–498.
21. Heijman RLJ, Stolte IG, Thiesbrummel HFJ, et al.. Opting out increases HIV
testing in a large sexually transmitted infections outpatient clinic. Sex Transm Infect. 2009;85:249–255.
22. Le Vu S, Le Strat Y, Barin F, et al.. Population-based HIV
-1 incidence in France, 2003–2008: a modelling analysis. Lancet Infect Dis. 2010;10:682–687.
23. Semaille C, Cazein F, Lot F, et al.. Recently acquired HIV
infection in men who have sex with men (MSM
) in France, 2003–2008. Euro Surveill. 2009;14:19425. Available at: http://www.ncbi.nlm.nih.gov/pubmed/20003900
. Accessed January 10, 2012.
24. Ginige S, Chen MY, Hocking JS, et al.. Rising HIV
notifications in Australia: accounting for the increase in people living with HIV
and implications for the HIV transmission rate
. Sex Health. 2007;4:31–33.
25. Prestage G, Jin F, Zablotska IB, et al.. Trends in HIV
testing among homosexual and bisexual men in eastern Australian states. Sex Health. 2008;5:119–123.
26. Marrone J, Fairley CK, Chen M, et al.. Comparisons of trends in antiretroviral use and HIV
notification rates between three Australian states. Aust N Z J Public Health. 2007;31:131–134.
27. Audelin A, Cowan S, Obel N, et al.. Phylogenetics of the Danish HIV
epidemic: the role of the very late presenters. Rev Antiviral Ther Infect Dis. 2012;2:6. Abstract O_04. Presented at the 10th European meeting on HIV
and Hepatitis, Treatment strategies and antiviral drug resistance; March 28–30, 2012; Barcelona, Spain.