Sullivan, Patrick S.a; Jones, Jeb S.a; Baral, Stefan D.b
The epidemiology of HIV is shaped by many factors, including local trends, the successes or failures of prevention and treatment efforts, and changes in the structures of populations. HIV epidemiology is further influenced by more global trends, including changes in the efficacy of available treatments, and the coverage of antiretroviral treatments (ARTs) among key populations at risk for the acquisition and transmission of HIV. Factors that drive trends in HIV epidemiology may be very different between high-incomes countries and low- and middle-income countries. These include coverage of ART, extent of HIV diagnosis among those living with HIV, access to healthcare services and social structures that facilitate accurate reporting of male–male and injection drug use sex risks. High-income countries have case-based surveillance systems , while many lower- and middle-income countries use other systems , such as demographic and household surveys and convenience surveys of key populations, to describe epidemiology. For all of these reasons, it is important to consider the epidemiology of HIV separately for high-income countries.
To describe the epidemiology of HIV in high-income countries, we used public surveillance reports and conducted a systematic literature review. We considered a layered framework: the epidemiology of HIV infections, including demographic characteristics and risk profiles, and the epidemiology of HIV prevention, including metrics of late HIV diagnoses, trends in HIV prevalence and deaths, and the complexity of denominators for key populations. Here, we summarize the major themes of our review of data and published literature.
MATERIALS AND METHODS
We utilized published literature and publicly available surveillance reports to inform our analyses.
We conducted a literature search to systematically assess the number and types of articles that have been recently published on the topic of HIV surveillance and epidemiology in high-income countries. From those countries that the World Bank identifies as high income , we restricted our search to countries in Europe, North America, Oceania and Japan because reviews of concentrated epidemics in the Caribbean  and Middle East  appear elsewhere in this issue. We conducted a systematic literature review with key terms of ‘epidemiology’, ‘HIV’ and the included countries. This search returned 329 articles; after further review, we included a total of 52 relevant manuscripts. We assessed these to determine whether they provided estimates for HIV incidence or prevalence in MSM, IDU, heterosexuals, female sex workers or transgender persons (see bibliography in supplementary digital content 1, http://links.lww.com/COH/A5).
Surveillance data and population metrics
Because many data on HIV prevalence and diagnoses are reported in surveillance reports, we also searched online for surveillance data from UNAIDS, WHO and region and country-specific sources. Population size, HIV prevalence and prevalence of female cases for 2011 were obtained from the UNAIDS AIDS Info website . HIV prevalence data for 2006 and death rates for 2001, 2006 and 2011 were obtained from the WHO Global Health Observatory Data Repository . Country populations used to calculate HIV prevalence rates for 2006 were obtained from the CIA World Factbook . Net growth in prevalence is reported as the change in cases per 100 000 population from the last two periods that prevalence data are available for all countries (2006 and 2011). Because death rates tended to be low in most high-income countries, it was not possible to analyse significant changes in the death rate. Thus, the death rates were classified categorically as generally increasing, decreasing or staying the same. Underlying data are available as a supplemental spreadsheet (supplementary digital content 2, http://links.lww.com/COH/A5).
In terms of the epidemiology of HIV infections, there were approximately 2.3 million people living with HIV (PLWH) in 2011 in the 33 high-income countries included in this review (Table 1); these PLWH represented about 6.7% of the total PLWH globally in 2011. The population prevalence among these countries was 233/100 000 population. Most cases were reported in the USA, representing 32% of cases in included high-income countries. Population prevalence was highest in Estonia (762/100 000), Portugal (449/100 000) and the United States (412/100 000). The male:female (M:F) case ratio was higher than one in all 33 countries. The median M:F case ratio was 2.5, and the highest M:F case ratio was reported from Germany (5.0). Trend in death rates was assessed from 2001 to 2011. In most (26/33; 79%) countries, there was a stable trend in death rates. However, during this time, death rates increased in Belgium, Estonia and Latvia, and decreased in France, Germany and Spain (Table 1).
There were data from 26 countries that reported probable route of HIV acquisition. Among these, the predominant acquisition mode was MSM (n = 14, 50% of countries), and for most other countries, heterosexual transmission was reported as the predominant acquisition mode (n = 10, 36%). Injecting drug use was the most common acquisition mode in only two countries: Lithuania and Estonia. The median M:F case ratio was the same in the 14 countries with reported predominant MSM transmission as in the 10 countries with reported modal heterosexual transmission (median 2.5 in each group). Examining trends in the proportions of cases by country suggested a clear pattern: in all 26 countries, the proportion of cases attributed to male–male sex increased from 2006 to 2011, and in 24 out of 26 countries, the proportion of cases attributed to male–female sex decreased from 2006 to 2011 (Table 2).
Across 33 countries with data, a median of 29% of HIV diagnoses was considered ‘late’ diagnoses, according to the relevant local definition (Fig. 1). The highest reported proportion of late diagnoses was in Portugal (50%); the lowest was in Australia (17%),
Among the 33 countries, only one country (Spain) had a decrease in HIV prevalence from 2006 to 2011; all other countries reported increased prevalence, and the median increase in prevalence was 14.8 cases/100 000 population (Table 1). The median proportional increase in prevalence was 13.6%.
Diagnosis rate per prevalent case
Among 28 countries with analysable data, the median diagnosis rate per prevalent case was 4% (data not shown in tables). The highest reported diagnosis rate per prevalence case was 13% (Japan); the lowest reported rate was 0.2% and was reported by three countries (Spain, Italy and Austria).
For European countries, substantial proportions of HIV infections diagnosed in countries may be among people who acquired their HIV infection outside of the country. For example, a large proportion of HIV cases reported in the European Union are among migrants from sub-Saharan Africa, especially among female cases [9,10]. About one-fifth of HIV cases among MSM are migrants from Western Europe and Latin America . Similarly, about only about half of newly diagnosed heterosexuals in the United Kingdom in 2011 acquired their HIV infection within the UK, up from 27% in 2002 .
Surveillance scientists and academic researchers also have increased access to molecular data about the HIV viruses involved in newly diagnosed patients, providing insights to HIV transmission and control. For example, several studies have highlighted the role of clusters of HIV transmissions among MSM [12,13]. Among college students in New England, MSM were significantly more likely to be in transmission clusters than were those with other risks , suggesting rapid HIV transmission within these networks secondary to the biology of HIV transmission during anal intercourse with HIV serodiscordant partners [15,16▪▪]. Lewis et al. suggest that about a quarter of onward transmissions in MSM occur within 6 months of infection; if confirmed, the results might support more recommendations for more frequent HIV testing among MSM.
Finally, our review illustrated the paucity of epidemiological data for some key populations. For example, only two articles reviewed provided primary data on HIV prevalence or incidence among sex workers [17,18] and one reported self-reported HIV prevalence data for transgendered women . Further, prevalence data on HIV among MSM have historically been reported using a denominator of all men, which significantly under-represented the impact of HIV in this group . MSM denominators have been recently reported for the USA [21▪], and Internet-based methods have been used to estimate MSM population size in Germany . Modelling of HIV epidemics among key population groups in France has included estimated denominators of MSM, IDU and heterosexuals using national census and nationally representative surveys .
There are several important conclusions from analysing trends in the epidemiology of HIV among the 33 high-income countries included in this analysis. HIV epidemics in high-income countries over the last decade have been characterized by growing HIV prevalence, and stable or declining death rates. These data also highlight the differential trajectories of HIV epidemics among MSM compared with other key populations. Further, gaps in data exist that hamper our ability to characterize the most highly impacted groups. Lastly, these data likely reflect the benefits of increased access to antiretroviral therapy for those living with HIV.
Globally, the M:F ratio of the burden of HIV is estimated to be 0.97 ; the median M:F ratio in high-income countries was 2.5 : 1. Previous data highlighting the disproportionate burden of HIV among men in high-income countries are primarily concentrated among MSM  and, to a lesser extent, among people who inject drugs . The relative increase in HIV cases among MSM is happening in the context of expanding HIV epidemics in these countries, supporting the absolute increase of the burden of HIV among MSM. This situation is further supported by earlier results demonstrating increases in the HIV diagnosis rate among MSM from eight of these same countries . Furthermore, increasing epidemics of HIV among MSM as the key contributor to the HIV epidemic across high-income settings is supported by the sustained or increasing published HIV incidence rates among MSM .
There are consistent data highlighting the need to optimize the continuum of HIV care for PLWH as a crucial component of comprehensive HIV prevention, treatment and care programmes . Reducing new HIV transmission by expanded, early and consistent use of ART is likely key to ending the HIV epidemic [29,30]. However, the data collected here highlight the challenges to the efficient engagement of PLWH in high-income countries. Late HIV diagnoses represent major missed opportunities for improved health outcomes  and remain common in these countries, reinforcing the need to improve HIV testing and screening programmes. In few high-income countries [32,33], routine HIV screening as part of medical care is recommended for all adults; in these countries, despite routine screening in medical care, enhanced testing services to promote frequent HIV testing among key populations are indicated. The prominence of phylogenetic clustering in epidemics among MSM suggests rapid transmission necessitating HIV testing more than annually among these men. In terms of the effectiveness of HIV prevention, the number of new HIV diagnoses per 100 persons living with HIV is a useful benchmark. Data on new diagnoses per prevalent HIV case suggest that most PLWH are not transmitting HIV to their partners: the median country-specific annual number of new diagnoses per prevalent case was 0.04.
Notable for their absence are data on transgender populations and sex workers as well as denominators for key populations in many countries. The majority of case-based HIV surveillance systems do not collect data separately on sex at birth and current sex, limiting our understanding of the risks and vulnerabilities among transgender women and men. These limited data are consistent with a recent systematic review of HIV among male to female transgender populations, which found data from only 15 countries around the world published between 2000 and 2012, including five high-income countries [34▪]. However, where data were available, transgender women appeared to bear the highest burden of HIV among any population, underpinning the need to better assess the needs of this population. Moreover, there are virtually no epidemiologic data on female to male transgender populations.
Further, there are few epidemiologic data disaggregating HIV infections among female sex workers from that of other heterosexual transmissions. The criminalization of sex work and the recently over-turned antiprostitution loyalty oath  have further complicated the implementation of sensitive case-based HIV surveillance systems for these women. Although the case-based HIV surveillance systems for MSM and people who inject drugs are likely more sensitive than those for sex workers in their ability to detect cases, the limited use of evidence-based denominators to present rates of HIV infection is an important limitation. Contextualizing the burden of HIV among these populations with denominators of population size would better demonstrate the actual magnitude of the HIV epidemic affecting these communities across high-income settings.
As previously mentioned, HIV prevalence has increased on average about 3% per year from 2006 to 2011. However, death rates were stable or declining in nearly 90% of countries (23/26). Although these data are ecological in nature, the individual-level benefit of early and sustained HIV treatment is well known . This situation suggests that in high-income settings, achieving high coverage of antiviral therapy might have causally stabilized death rates. Moreover, as people live longer with prevalent HIV infections, the cumulative prevalence increases given stable or slowly increasing HIV incidence in these settings. The lifespan of PLWH under treatment is approximating that of HIV-uninfected people [37,38], implying that the actual burden of prevalent HIV will continue to grow for the foreseeable future. To change the trajectory of this trend will require focusing on those populations with ongoing epidemic growth, and prioritize detection of HIV infection and earlier treatment, given the rapid transmission of HIV associated with acute and early HIV infection .
There are several limitations to these data and interpretations. There may be substantial misclassification of the mode of HIV acquisition, with likely differential misclassification across countries with varying case-based HIV surveillance systems. The sensitivity of HIV-surveillance systems to document mode of HIV acquisition in high-income settings is likely the lowest among the populations known to be at highest risk of HIV infection (people who inject drugs, MSM, female sex workers, and transgender populations). To the extent that these behaviours are stigmatized or criminalized, under-reporting of risks likely occurs, decreasing the sensitivity of these systems. For example, the median male:female case ratio was the same in countries reporting predominantly heterosexually epidemics and in countries reporting mostly MSM epidemics. This suggests differential misclassification of risk category in some countries. The interpretations of these data are prone to ecologic fallacy with limited ability to imply causality. However, interpreting the ecologic data within the context of complimentary data at the individual level minimizes inappropriate interpretations of data.
Each high-income country has a unique HIV epidemic given diverse socioeconomic and cultural contexts, geography and demographics. However, there also appear to be similarities in HIV epidemiologic metrics and trends over the last decade, including stable or declining HIV-related mortality rates and increasing HIV prevalence. Moreover, HIV epidemics in high-income settings continue to concentrate among MSM, even as the HIV epidemic among average-risk reproductive age adults is stable or slowing in most settings. More than 30 years into the HIV pandemic, we know relatively little about the risks and vulnerabilities among female sex workers and transgender populations. Changing the trajectory of the growing HIV epidemics in high-income countries will require addressing the biology of HIV transmission including acute HIV infection and transmission through anal intercourse with a viremic serodiscordant partner . The epidemiology of HIV in high-income countries depicts both the hope for improved clinical outcomes and the population-level impacts of treatment, and the sobering reality of the challenges with reducing new HIV infections among the mostly heavily impacted key populations.
This work was supported, in part, by the Emory Center for AIDS Research (P30 AI050409). The authors acknowledge Tom Coggia for his assistance in preparingFig. 1.
Conflicts of interest
The authors declare no conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
▪ of special interest
▪▪ of outstanding interest
1. Nakashima AK, Fleming PL. HIV/AIDS surveillance in the United States, 1981-2001. J Acquir Immune Defic Syndr 2003; 32 (Suppl 1):S68–S85.
2. Pisani E, Lazzari S, Walker N, Schwartlander B. HIV surveillance: a global perspective. J Acquir Immune Defic Syndr 2003; 32 (Suppl 1):S3–S11.
4. De Boni R, Veloso VG, Grinsztejn B. Epidemiology of HIV in Latin America and the Caribbean. Curr Opin HIV AIDS 2014; 9:192–198.
5. Mumtaz GR, Riedner G, Abu-Raddad LJ. The emerging face of the HIV epidemic in the Middle East and North Africa. Curr Opin HIV AIDS 2014; 9:183–191.
9. Del Amo J, Likatavicius G, Perez-Cachafeiro S, et al. The epidemiology of HIV and AIDS reports in migrants in the 27 European Union countries, Norway and Iceland: 1999-2006. Eur J Public Health 2011; 21:620–626.
10. McGraw-Hill Medical, Beyrer C, Baral D, Zenilman J. Holmes K. STDs, HIV/AIDS, and migrant populations. Sexually transmitted diseases 4th ed.2008; 257–268.
12. Lewis F, Hughes GJ, Rambaut A, et al. Episodic sexual transmission of HIV revealed by molecular phylodynamics. PLoS Med 2008; 5:e50.
13. Brenner BG, Roger M, Stephens D, et al. Transmission clustering drives the onward spread of the HIV epidemic among men who have sex with men in Quebec. J Infect Dis 2011; 204:1115–1119.
14. Chan PA, Kazi S, Rana A, et al. Short communication: new HIV infections at Southern New England academic institutions: implications for prevention. AIDS Res Hum Retroviruses 2013; 29:25–29.
15. Baggaley RF, White RG, Boily MC. HIV transmission risk through anal intercourse: systematic review, meta-analysis and implications for HIV prevention. Int J Epidemiol 2010; 39:1048–1063.
16▪▪. Beyrer C, Baral SD, van Griensven F, et al. Global epidemiology of HIV infection in men who have sex with men. Lancet 2012; 380:367–377.
This systematic review of the epidemiology of MSM globally includes summaries of surveillance data and studies of MSM, as well as agent-based modelling to illustrate the unique challenges of HIV prevention among MSM.
17. Parker RD, Ruutel K. A surveillance report of HIV status and high risk behaviors among rapid testing participants in Tallinn, Estonia. AIDS Behav 2011; 15:761–766.
18. Laisaar KT, Avi R, DeHovitz J, Uuskula A. Estonia at the threshold of the fourth decade of the AIDS era in Europe. AIDS Res Hum Retroviruses 2011; 27:841–851.
19. Bauer GR, Travers R, Scanlon K, Coleman TA. High heterogeneity of HIV-related sexual risk among transgender people in Ontario, Canada: a province-wide respondent-driven sampling survey. BMC Public Health 2012; 12:292.
20. Sullivan PS, Chu SY, Fleming PL, Ward JW. Changes in AIDS incidence for men who have sex with men, United States 1990–1995. AIDS 1997; 11:1641–1646.
21▪. Purcell DW, Johnson CH, Lansky A, et al. Estimating the population size of men who have sex with men in the United States to obtain HIV and syphilis rates. Open AIDS J 2012; 6:98–107.
This article is an illustration of how the use of denominators to report rates of HIV infection for key populations illustrates the true magnitude of disparities.
22. Marcus U, Schmidt A, Kollan C, Hamouda O. The denominator problem: estimating MSM-specific incidence of sexually transmitted infections and prevalence of HIV using population sizes of MSM derived from Internet surveys. BMC Public Health 2009; 9:181.
23. Le Vu S, Le Strat Y, Barin F, et al. Population-based HIV-1 incidence in France, 2003-08: a modelling analysis. Lancet Infect Dis 2010; 10:682–687.
25. Beyrer C, Sullivan P, Sanchez J, et al. The global HIV epidemics in MSM: time to act. AIDS 2013; [Epub ahead of print].
26. Strathdee SA, Stockman JK. Epidemiology of HIV among injecting and noninjecting drug users: current trends and implications for interventions. Curr HIV/AIDS Rep 2010; 7:99–106.
27. Sullivan PS, Hamouda O, Delpech V, et al. Reemergence of the HIV epidemic among men who have sex with men in North America, Western Europe, and Australia, 1996–2005. Ann Epidemiol 2009; 19:423–431.
28. Rebe K, De Swardt G, Struthers H, McIntyre J. Towards ‘Men who have sex with men-appropriate’ Health Services in South Africa. S African J HIV Med 2013; 14:52–57.
29. 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.
30. Padian NS, McCoy SI, Karim SSA, et al. HIV prevention transformed: the new prevention research agenda. Lancet 2011; 378:269–278.
31. Fisher M. Late diagnosis of HIV infection: major consequences and missed opportunities. Curr Opin Infect Dis 2008; 21:1–3.
32. Branson BM, Handsfield HH, Lampe MA, et al. CDC: Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep 2006; 55:1–17.
33. Deblonde J, Meulemans H, Callens S, et al. HIV testing in Europe: mapping policies. Health Policy (Amsterdam, The Netherlands) 2011; 103:101–110.
34▪. Baral S, Beyrer C, Muessig K, et al. Burden of HIV among female sex workers in low-income and middle-income countries: a systematic review and meta-analysis. Lancet Infect Dis 2012; 12:538–549.
This is a comprehensive review and summary of data on HIV prevalence among female sex workers, highlighting the high prevalence disparities, even in countries with medium or high background prevalence.
35. Devi S. US Supreme Court strikes down antiprostitution pledge. Lancet 2013; 381:2241.
36. Palella JFJ, Deloria-Knoll M, Chmiel JS, et al. Survival benefit of initiating antiretroviral therapy in HIV-infected persons in different CD4 + cell strata. Ann Intern Med 2003; 138:620–626.
37. Lohse N, Hansen A-BE, Pedersen G, et al. Survival of persons with and without HIV Infection in Denmark, 1995–2005. Ann Intern Med 2007; 146:87–95.
38. Harrison KM, Song R, Zhang X. Life expectancy after HIV diagnosis based on National HIV Surveillance Data from 25 States, United States. J Acquir Immune Defic Syndr 2010; 53:124–130.
39. Pilcher CD, Tien HC, Eron JJ, et al. Brief but efficient: acute HIV infection and the sexual transmission of HIV. J Infect Dis 2004; 189:1785–1792.
© 2014 Lippincott Williams & Wilkins, Inc.