AIDS Behav DOI 10.1007/s10461-015-1020-3
ORIGINAL PAPER
Trends in the HIV Epidemic Among African American Men Who Have Sex with Men, San Francisco, 2004–2011 V. Fuqua • H. Scott • S. Scheer • J. Hecht J. M. Snowden • H. Fisher Raymond
•
Ó Springer Science+Business Media New York 2015
Abstract African American men who have sex with men have been disproportionately affected by the HIV epidemic in the United States and remain to this day one of the groups with highest HIV prevalence and incidence. Our goal was to clarify the current state of HIV risk, sexual behaviors, and structural/network–network level factors that affect black MSM’s population risk of HIV, enabling the formulation of targeted and up-to-date public health messages/campaigns directed at this vulnerable population. Our approach maximized the use of local data through a process of synthesis and triangulation of multiple independent and overlapping sources of information that are sometimes separately published and often not examined side-by-side. Among African American MSM, we observed stable HIV incidence despite increases in reported individual risk behavior and STDs. An increasing proportion of African American MSM are reporting HIV testing in the past 6 months and seroadaptive behaviors, which may play a role in this observed decline in HIV among MSM in San Francisco, California. Our analysis suggests
V. Fuqua H. Scott S. Scheer H. F. Raymond (&) San Francisco Department of Public Health, 25 Van Ness, Suite 500, San Francisco, CA 94102, USA e-mail: [email protected] J. Hecht San Francisco AIDS Foundation, San Francisco, CA, USA J. M. Snowden Department of Obstetrics & Gynecology, Oregon Health & Science University, Portland, OR, USA H. F. Raymond Epidemiology and Biostatistics, University of California, San Francisco, USA
that currently the HIV epidemic is stable among African American MSM in San Francisco. However, we suggest that the observed stability is due to factors prohibiting expansion of new infections rather than decreasing risks for HIV infection among African American MSM. Keywords HIV prevalence HIV incidence Population size Men who have sex with men, population size estimation African American
Introduction African American men who have sex with men have been disproportionately affected by the HIV epidemic in the United States and remain to this day one of the groups with highest HIV prevalence and incidence. These disparities exist despite similar or lower reported individual HIV risk behaviors compared to White MSM, indicating that structural (i.e., stigma, homophobia, racism) and social (i.e., sexual networks) factors are driving these disparities [1]. This is true also for African American MSM in San Francisco where African American MSM comprise 6 % of the MSM population but account for 8 % of the MSM HIV cases in the city [2]. However, some recent findings from San Francisco have suggested that HIV prevalence among African American MSM has declined to a level consistent with that among White and Latino MSM [3]. While seemingly a positive development, it is unclear what may be driving this decline in HIV prevalence. Possible explanations include increases in mortality, emigration, changes in risk behaviors and/or changes in other structural and social factors related to HIV acquisition [4, 5]. To help begin to understand the epidemiology of HIV among African American MSM we synthesized the available data
123
AIDS Behav
on this population, assessing a number of HIV indicators, programmatic responses and community factors in order to better understand the current direction of the epidemic among this highly affected population. Our goal was to clarify the current state of HIV risk, sexual behaviors, and structural/network–network level factors that affect black MSM’s population risk of HIV, enabling the formulation of targeted and up-to-date public health messages/campaigns directed at this vulnerable population.
Methods Tracking the HIV epidemic and response among specific population requires data on HIV prevalence, HIV incidence, risk behaviors, programmatic data (i.e. data collected during the routine delivery of HIV prevention services) and the size of the population of interest. Unfortunately, data for these indicators are rarely if ever found in one dataset or originating from one study and are often not representative individually, making it difficult to understand the direction of the epidemic, or guide the programmatic decisions to respond appropriately to the epidemic. One solution for this problem is to maximize the use of local data through a process of synthesis and triangulation of multiple independent and overlapping sources of information that are sometimes separately published and often not examined side-by-side [6]. This approach, described in World Health Organization guidelines [7], has been used by San Francisco Department of Public Health for arriving at estimates of HIV prevalence and incidence used to guide the allocation of resources, set program targets, and help evaluate reach and impact [8].
short behavioral risk assessment (BRA) [11] and from the Black Brothers Esteem (BBE) program which is for African American gay, bisexual, and same gender loving men in San Francisco, California. Conducted by trained outreach volunteers the BRA is administered to approximately 1500–2000 MSM each year at venues, primarily in the Castro and South of Market areas of San Francisco where MSM congregate. BBE is a group-level HIV prevention intervention for African American men to gather and to gain support from each other, with the goal of providing education, skills and enhanced social connections to reduce sexual risk-taking. In addition to measures of HIV prevalence (via laboratory based HIV testing in community based samples) and HIV incidence, we also focused our analysis on structural factors (measures of service delivery), individual factors such as ‘‘drivers of HIV infection’’, sexual risk taking, in particular trends in serosorting/seroadaptive behaviors [12] and substance use. Drivers of HIV infection are behaviors or conditions that have an association with HIV infection and are present in at least 10 % of a population. The main drivers among MSM in San Francisco are multiple partners, methamphetamine use, STDs and binge alcohol use. These drivers are used in San Francisco HIV prevention planning (HIV Plan) and as such we limited our analyses to these indicators [14]. Lastly, we examined sexual mixing among MSM in San Francisco [3, 13] and HIV viral load suppression among MSM in San Francisco as indicators of social factors and health care access that may lead to more or fewer new HIV infections. Many of the specific calculations for the indicators are shown in the tables to ease understanding of how results were derived.
Data Sources
Results
Data sources used in the analysis include: National HIV Behavioral Surveillance (NHBS), a routine bio-behavioral survey conducted by sampling from all MSM in San Francisco every 3 years [9], Black Men Testing (BMT), an HIV testing pilot program among African American MSM [10], HIV/AIDS case reporting data, programmatic data from community based organizations and municipally funded HIV testing sites. NHBS is the CDC’s de facto gold standard to sample MSM in 20 cities in the United States. NHBS samples from any type of venue where MSM can be found including bars, clubs, cafes, gyms, religious organizations, parks, street locations and social groups. A number of these indicators have been published separately while others were collected or calculated specifically for this synthesis. San Francisco AIDS Foundation programs contributed programmatic data from STOP AIDS Project which conducts street based outreach among MSM and administers a
Population Size
123
Overall, the population size of MSM in San Francisco appears to be increasing from 2004 to 2011 by approximately 3000 men (Table 1). These estimates are from published data from San Francisco. In brief, population size estimates are derived using multiple methods including service multipliers and a component method. Details can be found elsewhere. [8, 19]. Using the proportion of MSM who reported being African American in NHBS we estimate that there were 3989, 4820 and 4055 African American MSM in 2004, 2008 and 2011, respectively. Using HIV prevalence from NHBS to determine how many HIV infected African American MSM live in San Francisco we estimated 1316, 1769 and 811 HIV infected African American MSM in 2004, 2008 and 2011. Finally, subtracting HIV infected from the total resulted in 2673,
AIDS Behav Table 1 Population size(s), African American MSM, San Francisco, 2004–2011 Indicator
2004
2008
2011
Calculation
Source
Overall MSM Population Size
63,577
66,032
66,487
na
NHBSa
% MSM that are African American
6.0
7.3
6.1
na
NHBSb
African American MSM Pop
3989
4820
4055
All MSM 9 % African American MSM in NHBS
NHBS
# HIV positive
1316
1769
811
African American MSM Population size 9 % African American MSM HIV positive
NHBSc
# HIV negative
2673
3051
3244
African American MSM pop size-# HIV positive
NHBS
a
Refs. [8, 19] 2008 mean of 2004 and 2011
b
Ref. [8]
c
Ref. [3]
Table 2 HIV Indicators among African American MSM, San Francisco, 2004–2011 Indicator
2004
% HIV infected
33
2005
2006
2007
2008
2009
2010
36.7
% HIV infected
2011
Calculation
Source
20
# HIV infections/ sample size
NHBS
na
BMT Case reporting/ NHBS NHBS
24
% HIV infected
27.8
24.1
30.0
# Cases/pop size
% Unrecognized infection
57.1
30.8
20
# Unrecognized/# HIV positive
% Unrecognized infection
23
# New HIV cases reported
35
HIV incidence %
1.3
22
24
37
39
31
Positivity rate (% positive)
5.2
3.9
4.2
3.6
2.4
2
# Deaths
28
38
28
30
25
# Living cases
1108
1118
1138
1143
1161
% HIV infected
27.8
# Unrecognized/# HIV positive 36
25
na
Case Reporting
1.0
(# New cases/HIV negative population size)
NHBS/Pop size/ Case Reporting
2
4.2
#HIV positive test/# total tests
CHEP supported HIV testing
27
19
18
na
Case Reporting
1186
1198
1217
na
Case Reporting
30.0
(# Reported cases/ population size)
NHBS/Case reporting
1.3
24.1
BMT
NHBS National HIV Behavioral Surveillance, Case Reporting mandated disease reporting, BMT Black Men Testing, Pop Size population size estimate, CHEP Community Health Equity and Promotion
3051 and 3244 HIV uninfected African American MSM in 2004, 2008 and 2011, respectively. HIV Prevalence, Unrecognized HIV Infection and HIV Incidence HIV prevalence appears to have declined from 2004 to 2011 from data from three samples of MSM (33 % in 2004 to 20 % in 2011) and one sample of African American MSM only (24 % in 2009). Using reported HIV cases and population size we estimated an HIV prevalence that appears stable around 30 % across the study period (HIV cases/population size = HIV prevalence) (Table 2). In terms of HIV incidence, we accessed or calculated HIV incidence estimates using three sources of information: population size of HIV negative African American
MSM derived from NHBS HIV-negative proportion, total African American MSM population size, and the number of new cases reported per year in mandated disease reporting. The number of newly reported cases remained mostly stable (mean 31 cases per year SD 6.6), the positivity rate among HIV testers at city funded clinics remained mostly stable (mean 3.4 % SD 1.2) and our calculated HIV incidence has remained stable (mean 1.2 % SD 0.6). Seroadaptive Behavior Among HIV uninfected African American MSM, the proportion of men reporting having sex but no anal sex in the past 6 months has declined from 4.3 % to 0 % from 2004 to 2011 and the proportion reporting using condoms 100 %
123
AIDS Behav Table 3 Seroadaptive strategies among HIV negative African American MSM, San Francisco, 2004–2011 Indicator
2004
2008
2011
% No Sex
22.6
22.6
19.2
% Sex but no anal sex
4.3
3.2
0
% Anal sex with condom only
47.3
38.7
26.9
% Any serosorting
22.5
35.5
46.2
% No seroadaptive strategy
3.2
0
7.7
Source: NHBS
of the time during anal sex has declined from 47.3 % in 2004 to 26.9 % in 2011. Meanwhile, the proportion of HIV uninfected African American MSM who report engaging in any seroadaptive behavior has increased from 22.5 to 46.2 % in the same period while the proportion who report no seroadaptive strategy has increased from 3.2 to 4.7 % (Table 3).
have not tested over the 2004–2011 period. The number of HIV tests in community settings have remained stable and the positivity rate among those tested has also remained stable at about 4–5 % from 2004 to 2011 (Table 5). Sexual Mixing Newman’s coefficient, which in this analysis was used to measure sexual mixing in and across racial groups, suggests that both MSM overall and African American MSM are mixing more readily across racial groups in 2011 compared to 2004 [3, 15]. In 2004, the coefficient was 0.4 for African American MSM while in 2011 the coefficient was 0.1, suggesting increased dissortative sexual mixing in this group. While HIV prevalence has remained stable among all MSM (24 % in 2004 to 23 % in 2011), the proportion of HIV infected MSM who are virally suppressed has increased steadily from 32 % in 2004 to 59 % in 2011. Viral suppression data are from the mandatory HIV case reporting system (Table 6).
HIV Risk Behaviors Discussion Individual Risk Factors (‘‘Drivers’’) for HIV Infection We accessed four data sources that had information on select ‘‘drivers’’ of HIV infection. Overall, among African American MSM, methamphetamine use has declined from 13 % reporting any use in 2004 to 9.7 % in both 2010 and 2011 while cocaine use and crack use appear to be stable in the study period (Table 4). HIV Testing A majority of HIV uninfected African American MSM have tested in the past 6 months however almost 40 % Table 4 HIV risk indicators, African American MSM, San Francisco, 2004–2011
Among African American MSM, we observed stable HIV incidence despite increases in reported individual risk behavior and STDs. An increasing proportion of African American MSM are reporting HIV testing in the past 6 months and seroadaptive behaviors, which may play a role in this observed decline in HIV among MSM in San Francisco, California. Seroadaptation has sometimes been characterized as fraught with potential risks of HIV transmission especially for African-American men compared to MSM of other races [16]. One such potential risk is wrongly classifying a sexual partner as the same serostatus, especially when high
Indicator
2004
% any MA use past 12 months
13
2005
2006
2007
2008
2009
18 9.7
21
27 11.3
11.3
5.4
NHBS National HIV Behavioral Surveillance, MA Methamphetamine, SAP Stop AIDS Project, SFAF San Francisco AIDS Foundation, BMT Black Men Testing, STD San Francisco Municipal STD Control, UAI unprotected anal intercourse, URAI unprotected receptive anal intercourse
123
% URAI past 6 months % BBE UAI % 2 ? partners past 6 months
4.8 10.9
14.3 61
73.3
48 62.2
% 2 ? partners past 6 months
55
51
NHBS NHBS SAP
20.8
NHBS
40 79.3
SFAF NHBS
66.0
% 2 ? partners past 6 months
NHBS
SAP 9.7
% crack past 6 month
Source
SAP 25
% cocaine use past 6 months % any crack use past 12 months
2011 9.7
% MA use past 6 months % any cocaine use past 12 months
2010
BMT 61.1
SAP
# chlamydia cases
73
78
83
88
92
129
137
128
STD
# gonorrhea cases
92
120
141
132
105
113
126
126
STD
# syphilis cases
42
24
40
28
44
38
51
54
STD
AIDS Behav Table 5 HIV testing among African American MSM, San Francisco, 2004–2011 Indicator
2004
% Testinga
62.3
2005
2006
2007
2008
2009
2010
2011
65.6
Source
72.7
% Testinga
NHBS
58
% Testinga
BMT 62.5
SAP
# African American MSM Tests
649
736
638
563
710
544
757
527
CHEP (new tests)
% HIV?b
5.2
3.9
4.2
3.6
2.4
2
2
4.2
CHEP (new tests)
a
HIV testing among those not already HIV positive, past 6 months
b
% Testing positive among those tested. HIV positivity rate (not population HIV prevalence) NHBS National HIV Behavioral Surveillance, BMT Black Men Testing, SAP Stop AIDS Project, CHEP Community Health Equity and Promotion
Table 6 Social/environment indicators, MSM, San Francisco, 2004–2011 Indicator
2004
Mixing (Newman’s Coefficient), African American MSM
0.4
Mixing (Newman’s Coefficient), all MSM
0.08
% Virally suppressed, all MSM
32
% HIV infected, all MSM
24
a
Ref. [3]
b
Ref. [8]
2005
33
2006
41
2007
48
2008
2011
Source
0.12
0.1
NHBSa
0.02
0
NHBSa
52 23
2009
54
2010
57
59
Case reporting
23
NHBSb
NHBS National HIV Behavioral Surveillance, Case Reporting mandated disease reporting Newman’s coefficient [15]
proportions of men within a sexual network are unaware of their HIV infection. We found that over the course of our study period that the proportion of African American MSM in NHBS who reported an HIV test in the past 6 months increased by approximately 10 %. While seroadaptive strategies are not currently recommended as an HIV prevention intervention for MSM, the increase in these reported behaviors with increased HIV testing may partly explain the reduction in the disparities among African American MSM in San Francisco, California. African American MSM are having sex with a wider range of other MSM in 2011 compared to 2004 while overall, viral suppression has increased among all MSM. This may suggest that African American MSM are less likely to have sex in situations that may lead to HIV acquisition compared to earlier data that suggested that because of high assortative sexual mixing African American MSM may have more likely been infected regardless of risk behavior profiles [17]. Of course, there are limitations to our synthesis of the available data. First, each indicator is only as strong as the original data from which it was drawn. Our use of multiple sources of data for individual indicators should ameliorate this limitation, as individual sources of data are unlikely to be biased in similar directions. Secondly, there may be data
sources that we were not able to access during our analysis. Future efforts to monitor the epidemic in this population can make every effort to discover and access more data sources. Thirdly, any estimates based on case reports may be low estimates due to the lag between actual infection and cases being reported. Lastly, synthesis of existing data such as this does not rely on statistical measures of uncertainty (e.g. 95 % Confidence Intervals). Findings and interpretation of the existing data should be taken cautiously. Our analysis suggests that currently the HIV epidemic is stable among African American MSM in San Francisco. Although not directly from our data, it is well worth considering that the observed stability is possibly due to factors prohibiting expansion of new infections rather than decreasing risks for HIV infection among African American MSM. Specifically, individual risk behaviors appear to currently being offset by increased dissortative mixing across races, viral suppression among all MSM, and increased serosorting in the setting where there are fewer HIV-positive men who are unaware of their infection [12, 18]. Should continuing decreases in condom use, increases in the proportion of men who do not engage in a strategy to reduce their risk of infection and increases in imperfect serosorting behaviors continue, these behaviors may
123
AIDS Behav
overwhelm the protection of mixing and viral suppression leading to an increase in new HIV infections among African American MSM. Acknowledgments HD079658-01.
JMS was supported by NIH Grant # K99
References 1. Millett GA, Peterson JL, Wolitski RJ, Stall R. Greater risk for HIV infection of black men who have sex with men: a critical literature review. Am J Public Health. 2006;96(6):1007–19. 2. San Francisco Department of Public Health (2013) 2012 HIV/ AIDS Epidemiology Annual Report. Available at http://www. sfdph.org/dph/files/reports/default.asp. Accessed 25 July 2014. 3. Sudhinaraset M, Raymond HF, McFarland W. The convergence of HIV prevalence and inter-racial sexual mixing among men who have sex with men, San Francisco, 2004-2011. AIDS Behav. 2012;17(4):1550–6. doi:10.1007/s10461-012-0370-3. 4. Magnus M, Kuo I, Phillips G, Shelley K, Rawls A, Montanez L, et al. Elevated HIV prevalence despite lower rates of sexual risk behaviors among African American men in the District of Columbia who have sex with men. AIDS Patient Care STDS. 2010;24(10):615–22. 5. Poundstone KE, Strathdee SA, Celentano DD. The social epidemiology of human immunodeficiency virus/acquired immunodeficiency syndrome. Epidemiol Rev. 2004;26:22–35. 6. Presanis AM, De Angelis D, Goubar A, et al. Bayesian evidence synthesis for a transmission dynamic model for HIV among men who have sex with men. Biostatistics. 2011;12(4):666–81. doi:10. 1093/biostatistics/kxr006. 7. WHO. Guidelines for second generation HIV surveillance. 2005. Available at http://whqlibdoc.who.int/hq/2000/WHO_CDS_CSR_ EDC_2000.5.pdf Accessed 26 Aug 2014. 8. Raymond HF, Bereknyei S, Berglas N, Hunter J, Ojeda N, McFarland W. Estimating population size, HIV prevalence and HIV incidence among men who have sex with men in San Francisco: a case example of triangulation methods. STI. 2013;89(5):383–7.
123
9. Gallagher KM, Sullivan PS, Lansky A, Onorato IM. Behavioral surveillance among people at risk for HIV infection in the U.S.: the National Behavioral Surveillance System. Public Health Rep. 2007;122(Sup. 1):32–8. 10. Fuqua V, Chen YH, Packer T, Dowling T, Ick TO, Nguyen BH, Colfax GN, Raymond HF. Using social networks to reach Black MSM for HIV testing and linkage to care. AIDS and Behav. 2012;16(2):255–65. 11. Kellogg TA, Bernstein K, Hecht J, Connors A, Perlhof L, Ick TO, Le B, Mehrtens J, Rive J Jr, McFarland W, Raymond HF. Comparison of HIV behavioral indicators among men who have sex with men across two survey methodologies, San Francisco, 2004 and 2008. STD. 2013;40(9):689–94. doi:10.1097/01.olq. 0000431354.96087.50. 12. Snowden JM, Wei CY, McFarland W, Raymond HF. Prevalence, correlates and trends in seroadaptive behaviors among men who have sex with men, San Francisco, 2004-2011. STI. 2014. doi:10. 1136/sextrans-2013-051368. 13. Bohl D, McFarland W, Raymond HF. Improved measures of racial mixing among men who have sex with men using Newman’s assortivity coefficient. STI. 2010;87(7):616–20. 14. San Francisco HIV Prevention Planning Council. 2004 San Francisco HIV Prevention Plan. 2004. Available at http://www. sfhiv.org/wp-content/uploads/HIV-Prev-Plan2010_final.pdf. Accessed 26 Aug 2014. 15. Newman ME. Mixing patterns in networks. Phys Rev E. 2003;67(2 Pt 2):026126. 16. Golden MR, Dombrowski JC, Kerani JD, Stekler JP. Failure of serosorting to protect African American men who have sex with men from HIV infection. Sex Transm Dis. 2012;39(9):659–64. doi:10.1097/OLQ.0b013e31825727cb. 17. Raymond HF, McFarland W. Racial mixing among men who have sex with men. AIDS Behav. 2009;13(4):671–6. 18. Raymond HF, Chen Y-H, Ick TO, Scheer S, Bernstein K, Liska S, Louie B, Pandori M, McFarland W. A new trend in the HIV epidemic among men who have sex with men, San Francisco, 2004-2011. JAIDS. 2013;62(5):584–9. doi:10.1097/QAI.0b013e318285febf. 19. San Francisco Department of Public Health (2006). HIV/AIDS Epidemiology Annual Report. Available at http://www.sfdph.org/ dph/files/ContractRequest4/RFP022011/RptsHIVAIDS/HIVAI DAnnlRpt2005.pdf. Accessed 15 Aug 2014.
ORIGINAL PAPER
Trends in the HIV Epidemic Among African American Men Who Have Sex with Men, San Francisco, 2004–2011 V. Fuqua • H. Scott • S. Scheer • J. Hecht J. M. Snowden • H. Fisher Raymond
•
Ó Springer Science+Business Media New York 2015
Abstract African American men who have sex with men have been disproportionately affected by the HIV epidemic in the United States and remain to this day one of the groups with highest HIV prevalence and incidence. Our goal was to clarify the current state of HIV risk, sexual behaviors, and structural/network–network level factors that affect black MSM’s population risk of HIV, enabling the formulation of targeted and up-to-date public health messages/campaigns directed at this vulnerable population. Our approach maximized the use of local data through a process of synthesis and triangulation of multiple independent and overlapping sources of information that are sometimes separately published and often not examined side-by-side. Among African American MSM, we observed stable HIV incidence despite increases in reported individual risk behavior and STDs. An increasing proportion of African American MSM are reporting HIV testing in the past 6 months and seroadaptive behaviors, which may play a role in this observed decline in HIV among MSM in San Francisco, California. Our analysis suggests
V. Fuqua H. Scott S. Scheer H. F. Raymond (&) San Francisco Department of Public Health, 25 Van Ness, Suite 500, San Francisco, CA 94102, USA e-mail: [email protected] J. Hecht San Francisco AIDS Foundation, San Francisco, CA, USA J. M. Snowden Department of Obstetrics & Gynecology, Oregon Health & Science University, Portland, OR, USA H. F. Raymond Epidemiology and Biostatistics, University of California, San Francisco, USA
that currently the HIV epidemic is stable among African American MSM in San Francisco. However, we suggest that the observed stability is due to factors prohibiting expansion of new infections rather than decreasing risks for HIV infection among African American MSM. Keywords HIV prevalence HIV incidence Population size Men who have sex with men, population size estimation African American
Introduction African American men who have sex with men have been disproportionately affected by the HIV epidemic in the United States and remain to this day one of the groups with highest HIV prevalence and incidence. These disparities exist despite similar or lower reported individual HIV risk behaviors compared to White MSM, indicating that structural (i.e., stigma, homophobia, racism) and social (i.e., sexual networks) factors are driving these disparities [1]. This is true also for African American MSM in San Francisco where African American MSM comprise 6 % of the MSM population but account for 8 % of the MSM HIV cases in the city [2]. However, some recent findings from San Francisco have suggested that HIV prevalence among African American MSM has declined to a level consistent with that among White and Latino MSM [3]. While seemingly a positive development, it is unclear what may be driving this decline in HIV prevalence. Possible explanations include increases in mortality, emigration, changes in risk behaviors and/or changes in other structural and social factors related to HIV acquisition [4, 5]. To help begin to understand the epidemiology of HIV among African American MSM we synthesized the available data
123
AIDS Behav
on this population, assessing a number of HIV indicators, programmatic responses and community factors in order to better understand the current direction of the epidemic among this highly affected population. Our goal was to clarify the current state of HIV risk, sexual behaviors, and structural/network–network level factors that affect black MSM’s population risk of HIV, enabling the formulation of targeted and up-to-date public health messages/campaigns directed at this vulnerable population.
Methods Tracking the HIV epidemic and response among specific population requires data on HIV prevalence, HIV incidence, risk behaviors, programmatic data (i.e. data collected during the routine delivery of HIV prevention services) and the size of the population of interest. Unfortunately, data for these indicators are rarely if ever found in one dataset or originating from one study and are often not representative individually, making it difficult to understand the direction of the epidemic, or guide the programmatic decisions to respond appropriately to the epidemic. One solution for this problem is to maximize the use of local data through a process of synthesis and triangulation of multiple independent and overlapping sources of information that are sometimes separately published and often not examined side-by-side [6]. This approach, described in World Health Organization guidelines [7], has been used by San Francisco Department of Public Health for arriving at estimates of HIV prevalence and incidence used to guide the allocation of resources, set program targets, and help evaluate reach and impact [8].
short behavioral risk assessment (BRA) [11] and from the Black Brothers Esteem (BBE) program which is for African American gay, bisexual, and same gender loving men in San Francisco, California. Conducted by trained outreach volunteers the BRA is administered to approximately 1500–2000 MSM each year at venues, primarily in the Castro and South of Market areas of San Francisco where MSM congregate. BBE is a group-level HIV prevention intervention for African American men to gather and to gain support from each other, with the goal of providing education, skills and enhanced social connections to reduce sexual risk-taking. In addition to measures of HIV prevalence (via laboratory based HIV testing in community based samples) and HIV incidence, we also focused our analysis on structural factors (measures of service delivery), individual factors such as ‘‘drivers of HIV infection’’, sexual risk taking, in particular trends in serosorting/seroadaptive behaviors [12] and substance use. Drivers of HIV infection are behaviors or conditions that have an association with HIV infection and are present in at least 10 % of a population. The main drivers among MSM in San Francisco are multiple partners, methamphetamine use, STDs and binge alcohol use. These drivers are used in San Francisco HIV prevention planning (HIV Plan) and as such we limited our analyses to these indicators [14]. Lastly, we examined sexual mixing among MSM in San Francisco [3, 13] and HIV viral load suppression among MSM in San Francisco as indicators of social factors and health care access that may lead to more or fewer new HIV infections. Many of the specific calculations for the indicators are shown in the tables to ease understanding of how results were derived.
Data Sources
Results
Data sources used in the analysis include: National HIV Behavioral Surveillance (NHBS), a routine bio-behavioral survey conducted by sampling from all MSM in San Francisco every 3 years [9], Black Men Testing (BMT), an HIV testing pilot program among African American MSM [10], HIV/AIDS case reporting data, programmatic data from community based organizations and municipally funded HIV testing sites. NHBS is the CDC’s de facto gold standard to sample MSM in 20 cities in the United States. NHBS samples from any type of venue where MSM can be found including bars, clubs, cafes, gyms, religious organizations, parks, street locations and social groups. A number of these indicators have been published separately while others were collected or calculated specifically for this synthesis. San Francisco AIDS Foundation programs contributed programmatic data from STOP AIDS Project which conducts street based outreach among MSM and administers a
Population Size
123
Overall, the population size of MSM in San Francisco appears to be increasing from 2004 to 2011 by approximately 3000 men (Table 1). These estimates are from published data from San Francisco. In brief, population size estimates are derived using multiple methods including service multipliers and a component method. Details can be found elsewhere. [8, 19]. Using the proportion of MSM who reported being African American in NHBS we estimate that there were 3989, 4820 and 4055 African American MSM in 2004, 2008 and 2011, respectively. Using HIV prevalence from NHBS to determine how many HIV infected African American MSM live in San Francisco we estimated 1316, 1769 and 811 HIV infected African American MSM in 2004, 2008 and 2011. Finally, subtracting HIV infected from the total resulted in 2673,
AIDS Behav Table 1 Population size(s), African American MSM, San Francisco, 2004–2011 Indicator
2004
2008
2011
Calculation
Source
Overall MSM Population Size
63,577
66,032
66,487
na
NHBSa
% MSM that are African American
6.0
7.3
6.1
na
NHBSb
African American MSM Pop
3989
4820
4055
All MSM 9 % African American MSM in NHBS
NHBS
# HIV positive
1316
1769
811
African American MSM Population size 9 % African American MSM HIV positive
NHBSc
# HIV negative
2673
3051
3244
African American MSM pop size-# HIV positive
NHBS
a
Refs. [8, 19] 2008 mean of 2004 and 2011
b
Ref. [8]
c
Ref. [3]
Table 2 HIV Indicators among African American MSM, San Francisco, 2004–2011 Indicator
2004
% HIV infected
33
2005
2006
2007
2008
2009
2010
36.7
% HIV infected
2011
Calculation
Source
20
# HIV infections/ sample size
NHBS
na
BMT Case reporting/ NHBS NHBS
24
% HIV infected
27.8
24.1
30.0
# Cases/pop size
% Unrecognized infection
57.1
30.8
20
# Unrecognized/# HIV positive
% Unrecognized infection
23
# New HIV cases reported
35
HIV incidence %
1.3
22
24
37
39
31
Positivity rate (% positive)
5.2
3.9
4.2
3.6
2.4
2
# Deaths
28
38
28
30
25
# Living cases
1108
1118
1138
1143
1161
% HIV infected
27.8
# Unrecognized/# HIV positive 36
25
na
Case Reporting
1.0
(# New cases/HIV negative population size)
NHBS/Pop size/ Case Reporting
2
4.2
#HIV positive test/# total tests
CHEP supported HIV testing
27
19
18
na
Case Reporting
1186
1198
1217
na
Case Reporting
30.0
(# Reported cases/ population size)
NHBS/Case reporting
1.3
24.1
BMT
NHBS National HIV Behavioral Surveillance, Case Reporting mandated disease reporting, BMT Black Men Testing, Pop Size population size estimate, CHEP Community Health Equity and Promotion
3051 and 3244 HIV uninfected African American MSM in 2004, 2008 and 2011, respectively. HIV Prevalence, Unrecognized HIV Infection and HIV Incidence HIV prevalence appears to have declined from 2004 to 2011 from data from three samples of MSM (33 % in 2004 to 20 % in 2011) and one sample of African American MSM only (24 % in 2009). Using reported HIV cases and population size we estimated an HIV prevalence that appears stable around 30 % across the study period (HIV cases/population size = HIV prevalence) (Table 2). In terms of HIV incidence, we accessed or calculated HIV incidence estimates using three sources of information: population size of HIV negative African American
MSM derived from NHBS HIV-negative proportion, total African American MSM population size, and the number of new cases reported per year in mandated disease reporting. The number of newly reported cases remained mostly stable (mean 31 cases per year SD 6.6), the positivity rate among HIV testers at city funded clinics remained mostly stable (mean 3.4 % SD 1.2) and our calculated HIV incidence has remained stable (mean 1.2 % SD 0.6). Seroadaptive Behavior Among HIV uninfected African American MSM, the proportion of men reporting having sex but no anal sex in the past 6 months has declined from 4.3 % to 0 % from 2004 to 2011 and the proportion reporting using condoms 100 %
123
AIDS Behav Table 3 Seroadaptive strategies among HIV negative African American MSM, San Francisco, 2004–2011 Indicator
2004
2008
2011
% No Sex
22.6
22.6
19.2
% Sex but no anal sex
4.3
3.2
0
% Anal sex with condom only
47.3
38.7
26.9
% Any serosorting
22.5
35.5
46.2
% No seroadaptive strategy
3.2
0
7.7
Source: NHBS
of the time during anal sex has declined from 47.3 % in 2004 to 26.9 % in 2011. Meanwhile, the proportion of HIV uninfected African American MSM who report engaging in any seroadaptive behavior has increased from 22.5 to 46.2 % in the same period while the proportion who report no seroadaptive strategy has increased from 3.2 to 4.7 % (Table 3).
have not tested over the 2004–2011 period. The number of HIV tests in community settings have remained stable and the positivity rate among those tested has also remained stable at about 4–5 % from 2004 to 2011 (Table 5). Sexual Mixing Newman’s coefficient, which in this analysis was used to measure sexual mixing in and across racial groups, suggests that both MSM overall and African American MSM are mixing more readily across racial groups in 2011 compared to 2004 [3, 15]. In 2004, the coefficient was 0.4 for African American MSM while in 2011 the coefficient was 0.1, suggesting increased dissortative sexual mixing in this group. While HIV prevalence has remained stable among all MSM (24 % in 2004 to 23 % in 2011), the proportion of HIV infected MSM who are virally suppressed has increased steadily from 32 % in 2004 to 59 % in 2011. Viral suppression data are from the mandatory HIV case reporting system (Table 6).
HIV Risk Behaviors Discussion Individual Risk Factors (‘‘Drivers’’) for HIV Infection We accessed four data sources that had information on select ‘‘drivers’’ of HIV infection. Overall, among African American MSM, methamphetamine use has declined from 13 % reporting any use in 2004 to 9.7 % in both 2010 and 2011 while cocaine use and crack use appear to be stable in the study period (Table 4). HIV Testing A majority of HIV uninfected African American MSM have tested in the past 6 months however almost 40 % Table 4 HIV risk indicators, African American MSM, San Francisco, 2004–2011
Among African American MSM, we observed stable HIV incidence despite increases in reported individual risk behavior and STDs. An increasing proportion of African American MSM are reporting HIV testing in the past 6 months and seroadaptive behaviors, which may play a role in this observed decline in HIV among MSM in San Francisco, California. Seroadaptation has sometimes been characterized as fraught with potential risks of HIV transmission especially for African-American men compared to MSM of other races [16]. One such potential risk is wrongly classifying a sexual partner as the same serostatus, especially when high
Indicator
2004
% any MA use past 12 months
13
2005
2006
2007
2008
2009
18 9.7
21
27 11.3
11.3
5.4
NHBS National HIV Behavioral Surveillance, MA Methamphetamine, SAP Stop AIDS Project, SFAF San Francisco AIDS Foundation, BMT Black Men Testing, STD San Francisco Municipal STD Control, UAI unprotected anal intercourse, URAI unprotected receptive anal intercourse
123
% URAI past 6 months % BBE UAI % 2 ? partners past 6 months
4.8 10.9
14.3 61
73.3
48 62.2
% 2 ? partners past 6 months
55
51
NHBS NHBS SAP
20.8
NHBS
40 79.3
SFAF NHBS
66.0
% 2 ? partners past 6 months
NHBS
SAP 9.7
% crack past 6 month
Source
SAP 25
% cocaine use past 6 months % any crack use past 12 months
2011 9.7
% MA use past 6 months % any cocaine use past 12 months
2010
BMT 61.1
SAP
# chlamydia cases
73
78
83
88
92
129
137
128
STD
# gonorrhea cases
92
120
141
132
105
113
126
126
STD
# syphilis cases
42
24
40
28
44
38
51
54
STD
AIDS Behav Table 5 HIV testing among African American MSM, San Francisco, 2004–2011 Indicator
2004
% Testinga
62.3
2005
2006
2007
2008
2009
2010
2011
65.6
Source
72.7
% Testinga
NHBS
58
% Testinga
BMT 62.5
SAP
# African American MSM Tests
649
736
638
563
710
544
757
527
CHEP (new tests)
% HIV?b
5.2
3.9
4.2
3.6
2.4
2
2
4.2
CHEP (new tests)
a
HIV testing among those not already HIV positive, past 6 months
b
% Testing positive among those tested. HIV positivity rate (not population HIV prevalence) NHBS National HIV Behavioral Surveillance, BMT Black Men Testing, SAP Stop AIDS Project, CHEP Community Health Equity and Promotion
Table 6 Social/environment indicators, MSM, San Francisco, 2004–2011 Indicator
2004
Mixing (Newman’s Coefficient), African American MSM
0.4
Mixing (Newman’s Coefficient), all MSM
0.08
% Virally suppressed, all MSM
32
% HIV infected, all MSM
24
a
Ref. [3]
b
Ref. [8]
2005
33
2006
41
2007
48
2008
2011
Source
0.12
0.1
NHBSa
0.02
0
NHBSa
52 23
2009
54
2010
57
59
Case reporting
23
NHBSb
NHBS National HIV Behavioral Surveillance, Case Reporting mandated disease reporting Newman’s coefficient [15]
proportions of men within a sexual network are unaware of their HIV infection. We found that over the course of our study period that the proportion of African American MSM in NHBS who reported an HIV test in the past 6 months increased by approximately 10 %. While seroadaptive strategies are not currently recommended as an HIV prevention intervention for MSM, the increase in these reported behaviors with increased HIV testing may partly explain the reduction in the disparities among African American MSM in San Francisco, California. African American MSM are having sex with a wider range of other MSM in 2011 compared to 2004 while overall, viral suppression has increased among all MSM. This may suggest that African American MSM are less likely to have sex in situations that may lead to HIV acquisition compared to earlier data that suggested that because of high assortative sexual mixing African American MSM may have more likely been infected regardless of risk behavior profiles [17]. Of course, there are limitations to our synthesis of the available data. First, each indicator is only as strong as the original data from which it was drawn. Our use of multiple sources of data for individual indicators should ameliorate this limitation, as individual sources of data are unlikely to be biased in similar directions. Secondly, there may be data
sources that we were not able to access during our analysis. Future efforts to monitor the epidemic in this population can make every effort to discover and access more data sources. Thirdly, any estimates based on case reports may be low estimates due to the lag between actual infection and cases being reported. Lastly, synthesis of existing data such as this does not rely on statistical measures of uncertainty (e.g. 95 % Confidence Intervals). Findings and interpretation of the existing data should be taken cautiously. Our analysis suggests that currently the HIV epidemic is stable among African American MSM in San Francisco. Although not directly from our data, it is well worth considering that the observed stability is possibly due to factors prohibiting expansion of new infections rather than decreasing risks for HIV infection among African American MSM. Specifically, individual risk behaviors appear to currently being offset by increased dissortative mixing across races, viral suppression among all MSM, and increased serosorting in the setting where there are fewer HIV-positive men who are unaware of their infection [12, 18]. Should continuing decreases in condom use, increases in the proportion of men who do not engage in a strategy to reduce their risk of infection and increases in imperfect serosorting behaviors continue, these behaviors may
123
AIDS Behav
overwhelm the protection of mixing and viral suppression leading to an increase in new HIV infections among African American MSM. Acknowledgments HD079658-01.
JMS was supported by NIH Grant # K99
References 1. Millett GA, Peterson JL, Wolitski RJ, Stall R. Greater risk for HIV infection of black men who have sex with men: a critical literature review. Am J Public Health. 2006;96(6):1007–19. 2. San Francisco Department of Public Health (2013) 2012 HIV/ AIDS Epidemiology Annual Report. Available at http://www. sfdph.org/dph/files/reports/default.asp. Accessed 25 July 2014. 3. Sudhinaraset M, Raymond HF, McFarland W. The convergence of HIV prevalence and inter-racial sexual mixing among men who have sex with men, San Francisco, 2004-2011. AIDS Behav. 2012;17(4):1550–6. doi:10.1007/s10461-012-0370-3. 4. Magnus M, Kuo I, Phillips G, Shelley K, Rawls A, Montanez L, et al. Elevated HIV prevalence despite lower rates of sexual risk behaviors among African American men in the District of Columbia who have sex with men. AIDS Patient Care STDS. 2010;24(10):615–22. 5. Poundstone KE, Strathdee SA, Celentano DD. The social epidemiology of human immunodeficiency virus/acquired immunodeficiency syndrome. Epidemiol Rev. 2004;26:22–35. 6. Presanis AM, De Angelis D, Goubar A, et al. Bayesian evidence synthesis for a transmission dynamic model for HIV among men who have sex with men. Biostatistics. 2011;12(4):666–81. doi:10. 1093/biostatistics/kxr006. 7. WHO. Guidelines for second generation HIV surveillance. 2005. Available at http://whqlibdoc.who.int/hq/2000/WHO_CDS_CSR_ EDC_2000.5.pdf Accessed 26 Aug 2014. 8. Raymond HF, Bereknyei S, Berglas N, Hunter J, Ojeda N, McFarland W. Estimating population size, HIV prevalence and HIV incidence among men who have sex with men in San Francisco: a case example of triangulation methods. STI. 2013;89(5):383–7.
123
9. Gallagher KM, Sullivan PS, Lansky A, Onorato IM. Behavioral surveillance among people at risk for HIV infection in the U.S.: the National Behavioral Surveillance System. Public Health Rep. 2007;122(Sup. 1):32–8. 10. Fuqua V, Chen YH, Packer T, Dowling T, Ick TO, Nguyen BH, Colfax GN, Raymond HF. Using social networks to reach Black MSM for HIV testing and linkage to care. AIDS and Behav. 2012;16(2):255–65. 11. Kellogg TA, Bernstein K, Hecht J, Connors A, Perlhof L, Ick TO, Le B, Mehrtens J, Rive J Jr, McFarland W, Raymond HF. Comparison of HIV behavioral indicators among men who have sex with men across two survey methodologies, San Francisco, 2004 and 2008. STD. 2013;40(9):689–94. doi:10.1097/01.olq. 0000431354.96087.50. 12. Snowden JM, Wei CY, McFarland W, Raymond HF. Prevalence, correlates and trends in seroadaptive behaviors among men who have sex with men, San Francisco, 2004-2011. STI. 2014. doi:10. 1136/sextrans-2013-051368. 13. Bohl D, McFarland W, Raymond HF. Improved measures of racial mixing among men who have sex with men using Newman’s assortivity coefficient. STI. 2010;87(7):616–20. 14. San Francisco HIV Prevention Planning Council. 2004 San Francisco HIV Prevention Plan. 2004. Available at http://www. sfhiv.org/wp-content/uploads/HIV-Prev-Plan2010_final.pdf. Accessed 26 Aug 2014. 15. Newman ME. Mixing patterns in networks. Phys Rev E. 2003;67(2 Pt 2):026126. 16. Golden MR, Dombrowski JC, Kerani JD, Stekler JP. Failure of serosorting to protect African American men who have sex with men from HIV infection. Sex Transm Dis. 2012;39(9):659–64. doi:10.1097/OLQ.0b013e31825727cb. 17. Raymond HF, McFarland W. Racial mixing among men who have sex with men. AIDS Behav. 2009;13(4):671–6. 18. Raymond HF, Chen Y-H, Ick TO, Scheer S, Bernstein K, Liska S, Louie B, Pandori M, McFarland W. A new trend in the HIV epidemic among men who have sex with men, San Francisco, 2004-2011. JAIDS. 2013;62(5):584–9. doi:10.1097/QAI.0b013e318285febf. 19. San Francisco Department of Public Health (2006). HIV/AIDS Epidemiology Annual Report. Available at http://www.sfdph.org/ dph/files/ContractRequest4/RFP022011/RptsHIVAIDS/HIVAI DAnnlRpt2005.pdf. Accessed 15 Aug 2014.
