HIV disproportionately affects black men in the United States: most diagnoses are for black gay, bisexual, and other men who have sex with men (collectively referred to as MSM). A better understanding of the social conditions in which black men live and work may better explain why HIV incidence and diagnosis rates are higher than expected in this population.
Using data from the National HIV Surveillance System and the US Census Bureau's American Community Survey, we examined the relationships of HIV diagnosis rates and 5 census tract–level social determinants of health variables for 21,948 black MSM and non-MSM aged ≥15 years residing in 17 areas in the United States. We examined federal poverty status, marital status, education level, employment status, and vacancy status and computed rate ratios (RRs) and prevalence odds ratios (PORs), using logistic regression with zero-inflated negative binomial modeling.
Among black MSM, HIV diagnosis rates decreased as poverty increased (RR: 0.54). At the time of HIV diagnosis, black MSM were less likely than black non-MSM to live in census tracts with a higher proportion below the poverty level (POR: 0.81) and with a higher proportion of vacant houses (POR: 0.86). In comparison, housing vacancy was positively associated with HIV diagnosis rates among black non-MSM (RR: 1.65). HIV diagnosis rates were higher for black MSM (RR: 2.75) and non-MSM (RR: 4.90) whose educational level was low. Rates were significantly lower for black MSM (RR: 0.06) and non-MSM (RR: 0.26) as the proportion unemployed and the proportion married increased.
This exploratory study found differences in the patterns of HIV diagnosis rates for black MSM and non-MSM and provides insight into the transmission of HIV infection in areas that reflect substantial disadvantage in education, housing, employment, and income.
Citation: Gant Z, Gant L, Song R, Willis L, Johnson AS (2014) A Census Tract–Level Examination of Social Determinants of Health among Black/African American Men with Diagnosed HIV Infection, 2005–2009—17 US Areas. PLoS ONE 9(9): e107701. doi:10.1371/journal.pone.0107701
Editor: Chongyi Wei, UCSF, United States of America
Received: June 9, 2014; Accepted: August 20, 2014; Published: September 30, 2014
This is an open-access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Data Availability: The authors confirm that, for approved reasons, some access restrictions apply to the data underlying the findings. CDC ensures that these data are in compliance with the Data Security and Confidentiality Guidelines for HIV, Viral Hepatitis, Sexually Transmitted Disease, and Tuberculosis Programs: Standards to Facilitate Sharing and Use of Surveillance Data for Public Health Action.The data shared with CDC from the state and local jurisdictions are only be used for statistical summary purposes. Additional protections are provided by exemptions to the Freedom of Information Act of 1966—specifically, 5 USC §552(b)(6)—and by the Privacy Act of 1974. Most importantly, the Assurance of Confidentiality authorized by 308(d) of the Public Health Service Act guarantees that any information collected as part of HIV surveillance that would permit direct or indirect identification of an individual or institution will be held in confidence and will not otherwise be disclosed or released without the consent of the individual or institution. Therefore, we are not permitted to share these data. Additionally, the general data sources for each dataset are located here: * Centers for Disease Control and Prevention's (CDC's) National HIV Surveillance System (NHSS) http://www.cdc.gov/hiv/statistics/ and http://gis.cdc.gov/GRASP/NCHHSTPAtlas/main.html * CDC's HIV Geographic Information System (GIS) Supplemental Surveillance Project (referred to in http://www.cdc.gov/hiv/pdf/statistics_2005_2009_HIV_Surveillance_Report_vol_18_n4.pdf * US Census Bureau's American Community Survey (ACS) http://www.census.gov/acs/www/data_documentation/data_main/ Since these data in this report are at the census-tract level, we do not have permission to release data at this unit of analysis. However, the NCHHSTP Atlas (http://gis.cdc.gov/GRASP/NCHHSTPAtlas/main.html) provides county-level data, in which this project can be replicated at that unit of analysis.
Funding: The Centers for Disease Control and Prevention (CDC) provides funding to state and local health departments to conduct surveillance of HIV disease in accordance with their own disease reporting regulations. CDC provides technical guidance for the collection of data by the state and local health departments and data are sent to CDC for national-level analyses. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
HIV disproportionately affects black/African American (black) men who accounted for 12% of the male population and 42% of HIV infections diagnosed among men in the United States during 2011 , . In 2011, estimated rates of diagnoses of HIV infection among black men were nearly 8 times the rate for whites and more than twice the rate for Hispanics/Latinos. For black men, the lifetime risk of HIV is estimated to be 1 in 16 . Gay, bisexual, and other men who have sex with men (collectively referred to as MSM) account for the highest number of cases among black men , . Although the rates of HIV infection are higher for black MSM than for MSM of any other race/ethnicity, research has shown no significant differences in the sexual risk behaviors of black and white MSM –. This observation suggests that factors other than individual risk behavior are contributing to the rate disparity. These disproportionate rates of HIV may be partly explained by a better understanding of the social determinants that affect black men's lives and work –. Social determinants of health (SDH) serve as accessible proxies for social factors (e.g., poverty, education, marital status, and employment) and physical environmental factors (e.g., property values, housing vacancies), and they allow us to contextualize HIV risk among at-risk populations –.
Research addressing disproportionate rates of HIV infection among black men has focused mainly on the sexual behaviors of black MSM , , , not on the environment in which these behaviors occur. This research tends to assess the risk of HIV infection solely by examining sexual behavior such as unprotected anal intercourse. Whereas, most studies examining the risk of HIV infection among heterosexual black men have not assessed risk solely with a specific sexual behavior , , , ; instead they compound behavioral risk factors with other factors such as drug use, crime, and excess mortality , , , . This assessment may not truly encompass the behavioral and environmental paradigms that exist for heterosexual black men. Additionally, most previous research that has examined HIV risk among the black male population has addressed MSM only or included MSM and non-MSM as a combined population. Given that black MSM accounted for approximately 70% of newly diagnosed HIV infections among black males annually , , examination of the combined populations could skew the conclusions toward those applicable to the MSM population.
We had 2 objectives: (1) to assess selected SDH that may provide insight into the disparate rates of HIV transmission among black MSM and non-MSM subpopulations separately; and (2) to examine the differences in proportion of HIV diagnoses between black MSM and non-MSM and the SDH that may be associated with these differences.
Data were obtained from 3 sources: the Centers for Disease Control and Prevention's (CDC's) National HIV Surveillance System (NHSS) (http://www.cdc.gov/nchhstp/Atlas/Index.htm), CDC's HIV Geographic Information System (GIS) Supplemental Surveillance Project (http://www.cdc.gov/hiv/pdf/statistics_2005_2009_HIV_Surveillance_Report_vol_18_n4.pdf), and the US Census Bureau's American Community Survey (ACS) (http://www.census.gov/acs/www/data_documentation/data_main/). The numbers, percentages, and rates of diagnoses of HIV infection are based on cases reported to NHSS through December 2010 and included cases among black males aged 15 years and older whose HIV infection was diagnosed during 2005–2009, regardless of the stage of disease at diagnosis. The population of black MSM included men who had ever had sexual contact with other men, men who had ever had sexual contact with both men and women, and men who had sexual contact with other men and injected drugs. The population of black non-MSM included men with infection attributable to heterosexual contact (i.e., men who had ever had heterosexual contact with a person known to have, or to be at high risk for, HIV infection), injection drug use (IDU), or other modes of transmission (i.e., hemophilia, blood transfusion, and perinatal exposure).
The analysis was based on data from the 17 (of 29) areas that were funded for the 2010 HIV GIS Supplemental Surveillance Project and that provided 5 years (2005–2009) of geocoded data for residence at diagnosis: Colorado, District of Columbia, Illinois (excluding Chicago), Iowa, Los Angeles County, Louisiana, Michigan, Minnesota, Mississippi, New York City, New York State, North Carolina, San Francisco, South Carolina, Virginia, Washington, and Wisconsin. We linked census tract SDH indicator data to individual cases on the basis of residence at the time of diagnosis. HIV diagnosis data were statistically adjusted for missing transmission category , but not for delayed or incomplete reporting.
Census tract–level SDH data were obtained from the ACS 2005–2009 5-year estimates . We examined 5 SDH variables: federal poverty status, marital status, education level, employment status, and housing vacancy status. We measured
- federal poverty status by the “proportion of residents in the census tract who were living below the US poverty level (i.e., below a specified threshold) within the last 12 months of the survey response”
- marital status by the “proportion of people currently married among individuals 15 years and older”
- education level by the “proportion with less than high school education for individuals 18 years and older”
- employment status by the “proportion in the workforce without a job for individuals 16 years and older”
- housing vacancy status by the “proportion of vacant houses within a census tract”
We included these SDH variables because they are generally recognized in the scientific literature as population determinants of health. Although other social determinants may affect health, research on social determinants among black males has emphasized the need to incorporate these specific structural and societal factors into analyses of public health data , , . Additionally, authors of studies evaluating SDH have noted the need to include analysis of the geographic environment to help explain how these factors moderate individual-level behaviors that influence health outcomes , , .
Rates per 100,000 population were calculated for the numbers of diagnoses of HIV infection in each census tract. The population denominators used to compute these rates for the 17 areas were based on the US Census ACS 5-year estimates of the black male populations . The 5-year average annual HIV diagnosis rate was calculated by dividing the total number of diagnoses among black males aged 15 years and older during 2005–2009 by the total black male population aged 15 years and older for the same 5 years. The result was then multiplied by 100,000 and divided by 5 to obtain the annualized 5-year rate. Because of the lack of US Census data stratified by transmission category, HIV diagnosis rates among black MSM and non-MSM were calculated by using the general black male population as the denominator.
The association between HIV diagnosis rates among black MSM and non-MSM and SDH variables was determined by using stratified logistic regression with zero-inflated negative binomial modeling . Zero-inflated modeling was used to account for the excess of zero counts (i.e., census tracts with no black males with diagnosed HIV infection) to reduce bias due to the skewing of data toward zero counts. Rate ratios (RRs) were computed to determine the HIV diagnosis rates for black MSM and non-MSM when the proportion change for the SDH variable of interest (e.g., proportion of people living below poverty level) increased from zero (i.e., 0%, or no one living below poverty level) to one (i.e., 100%, or everyone living below poverty level). Consequently, we constructed 2 models and calculated average annual HIV diagnosis RRs and 95% confidence intervals (CIs) stratified for black MSM and non-MSM at the census tract–level. We report the effect of each SDH variable on the HIV diagnosis rate when the SDH variable was increased by 1 percentage point (i.e., 0.01).
Because we were unable to determine the total populations of black MSM and non-MSM in the 17 areas to compute transmission category–specific rates, we computed prevalence odds ratios (PORs). The POR was defined as the prevalence odds of HIV diagnosis for black MSM vs. non-MSM based on the census tract at the time of diagnosis, when the proportion change for the specified SDH variable of interest increased from 0 (or 0%) to 1 (or 100%). We calculated PORs and 95% CIs for HIV diagnosis among black non-MSM vs MSM, and we report the effect of each SDH variable on the prevalence odds of HIV diagnosis for black MSM vs non-MSM when the SDH variable increased by 1 percentage point (i.e., 0.01). The prevalence odds of HIV diagnosis for black MSM vs. non-MSM was defined as (No. MSM + 1)/(No. non-MSM + 1). The numeral 1 was added to the numerator and the denominator to avoid an undefined POR when the number of non-MSM was zero. A POR of>1 indicated that the proportion of black MSM compared with black non-MSM with diagnosed HIV increased as the proportion of the SDH variable of interest increased. However, a POR of <1 indicated that the proportion of black MSM compared with black non-MSM with diagnosed HIV decreased as the proportion of the SDH variable of interest increased.
Recognizing the complex relationships between these SDH variables, we retained all SDH variables in our models (i.e., we analyzed each individual social determinant while controlling for all other SDH) to determine the overall impact of contributing factors. We also controlled for 3 indicators with possible confounding effects in our model: proportion of males in the general population in a census tract, proportion of blacks in the general population in a census tract, and proportion of people aged 15–49 years in a census tract. All results for each outcome of interest in the logistic regression models are based on controlling for all other variables. Statistical analyses were based on regression with a significance level of 0.05 and were performed by using SAS version 9.2 (SAS Institute, Inc., Cary, NC). All reported outcomes were statistically significant unless otherwise noted.
During 2005–2009, in the 17 HIV Supplemental Surveillance Project areas, HIV infection was diagnosed for 22,396 black males aged 15 years and older, and residential addresses had been geocoded to the census tract level in 18,330 census tracts for 98% (21,948) (Table 1). Of the 15,276 (69.6%) classified as MSM, 29.4% were aged 18–24 years at diagnosis, and 28.3% were aged 25–34 years. Of the 6,672 black males classified as non-MSM, 31.6% were aged 45–54 years, and 29.0% were aged 35–44 years. The average annual HIV diagnosis rate for black males was 416.6 per 100,000 population.
In the stratified logistic regression models, some of the patterns for the 2 populations of black males (Table 2) differed. For every percentage point increase in the proportion below the poverty level, the likelihood of HIV diagnosis among black MSM decreased by 0.46% [(RRMSM = 0.54)–1]. The reverse pattern, although not statistically significant, was observed among non-MSM. For every percentage point increase in the proportion of vacant houses, the likelihood of HIV diagnosis among black non-MSM increased by 0.65% [(RRnon-MSM = 1·65)–1]. The reverse pattern, although not statistically significant, was observed among black MSM.
In contrast, the stratified models revealed some similarities among black MSM and non-MSM (Table 2). For every percentage point increase in the proportion married and the proportion unemployed, the likelihood of HIV diagnosis among black MSM decreased by 0.94% [(RRMSM = 0.06)–1] and 0.74% [(RRMSM = 0.26)–1], respectively, and among black non-MSM, decreased by 0.90% [(RR non-MSM = 0·10)–1] and 0.94% [RRnon-MSM = 0·06)–1], respectively. For every percentage point increase in the proportion with less than a high school education, the likelihood of HIV diagnosis increased 1.75% [(RRMSM = 2.75)–1] among black MSM and 3.90% [(RRnon-MSM = 4.90)–1] among black non-MSM.
From our POR calculations (Table 3), for every percentage point increase in the proportion below the poverty level, the proportion married, and the proportion of vacant houses, the change in odds of HIV diagnosis for black MSM versus non-MSM were lower by 0.19% (POR: 0.81), 0.16% (POR: 0.84), and 0.14% (POR: 0.86), respectively. However, for every percentage point increase in the proportion with less than a high school education, the odds of HIV diagnosis for black MSM versus black non-MSM were higher by 0·14% (POR: 1·14).
Our analysis is one of few large-scale studies examining SDH among black males on the basis of HIV transmission category. Overall, we found that HIV diagnosis rates followed differing patterns for black MSM and non-MSM when we examined poverty and housing vacancies. Similar patterns emerged when we examined education level, unemployment status, and marital status.
According to an earlier study of more than 9,000 people in 23 US cities, heterosexuals living below the poverty level were 5 times as likely as the nation's general population to be HIV-positive, regardless of race or ethnicity . We found a similar, but statistically nonsignificant, pattern among black non-MSM in our analysis. Also, our results showed that as poverty increased, HIV diagnosis rates decreased among black MSM. Although previous research suggests that community involvement in HIV-related organizations and groups (e.g., social services, support groups) may protect against the association of poverty and risky sexual behavior , a more plausible explanation could be that poverty is masking HIV infection. Additionally, we found that at the time of HIV diagnosis, black MSM were less likely than black non-MSM to live in census tracts in which a higher proportion of persons were living below the poverty level. On the basis of geographic location, the structural barriers of poverty appear to have differing effects on HIV diagnosis rates among black MSM compared with non-MSM.
Housing vacancy, one of many measures of neighborhood distress, can serve as a proxy measurement of material deprivation, lack of resources, households in poverty, and socioeconomic status –. A geospatial methods study found that housing vacancy rates were positively associated with HIV diagnosis rates and high-risk heterosexual transmission rates . Our study supports these findings for black non-MSM, but not for black MSM. Additionally, we found that at the time of diagnosis, black MSM were less likely than black non-MSM to live in census tracts with a higher proportion of vacant houses. Further assessment is needed to better understand the possible relationship between housing vacancies and HIV diagnosis rates among black MSM.
A low level of education (<high school education) followed the expected associations for black MSM and non-MSM: HIV diagnoses rates for black men were higher when educational attainment was low –. Despite the similar pattern for both populations and contrary to our expectations, our results show that at the time of HIV diagnosis, MSM were more likely than non-MSM to live in census tracts in which residents had a lower level of education. This finding is consistent with the findings of other studies in which patterns of lower educational access, performance, and outcomes persisted for black MSM compared with black non-MSM , . Persistent fears and documented reports of harassment and physical violence toward gay and bisexual youth in educational settings leads to increased missed days of school and dropouts at all educational levels (i.e., presecondary, secondary, and postsecondary) , . Additionally, a relationship may exist between low educational attainment and the proportion of persons who live below the poverty level. Further research might examine the extent to which the interaction of these structural factors affects rates of HIV infection among black MSM and non-MSM.
Unemployment rates and HIV infection varied in ways that we did not expect. We expected unemployment to follow the same pattern as education for black MSM and non-MSM, but HIV diagnosis rates were significantly lower as unemployment rates increased. We suspect that unemployed persons may not be able to access HIV testing and treatment and therefore do not get tested. For example, Mayer et al. found that unemployed black MSM were more likely to have undiagnosed HIV infection . If rates of risky behaviors are similar among employed and unemployed MSM, then lower rates of testing among unemployed MSM could explain lower rates of HIV diagnosis in that group. Employed persons have a greater chance of receiving testing through their jobs and are more likely to be engaged in the health care system through the provision of employer-sponsored health insurance.
Furthermore, a relationship may exist between the proportion unemployed and the proportion living below the poverty level. This relationship could suggest that impoverished men have lower social mobility and increased chances for unemployment, thus fewer opportunities to meet prospective partners. Regardless of the outcome of partner engagement (e.g. short-term or long-term relationships), possession of, or access to, material resources and economic opportunities may be a critical component in partner seeking and partner engagement. As previously discussed, poverty may be masking HIV infections among black MSM. Additional research is needed to further assess the relationship between unemployment and HIV diagnosis rates.
For black MSM and non-MSM, HIV diagnosis rates decreased as the proportion of married persons in the census tract increased. At the time of diagnosis, black MSM were less likely than non-MSM to live in a census tract in which more persons were married. The protective effects of marriage (e.g., reduction in the number of partners, sex with partner at low risk) for heterosexual partners continues to be debated ,  but appears to be supported by our findings. It is unclear how marriage would protect black MSM, given the absence of legalized marriage among same-sex partners in some of the participating states during the time frame of data collection. It is possible that the rates of HIV diagnosis among MSM were low because the populations of MSM who lived in the census tracts in these 17 areas were small; however, this question could not be answered because we were unable to compute transmission category–specific HIV rates by census tract. Little explanation in current relevant literature could be found regarding the latter finding, and should be a focus of further study and possibly an in-depth examination of some areas to assess the relationship.
Our analysis had several limitations. First, diagnoses of HIV infection do not represent incidence, or new infections. The time from infection to diagnosis varies by individual, and residence at HIV diagnosis may not be the residence at the time HIV infection was acquired. Second, data for this analysis were not adjusted for reporting delays. This may have resulted in an underestimate of the number of cases diagnosed during 2005–2009. Third, data were limited to persons in 17 areas whose residential addresses were complete and thus could be geocoded; therefore, results may not reflect the population of black males with diagnosed HIV in those areas. However, according to the surveillance databases in these 17 areas, 98% of persons with diagnosed HIV had residential addresses that were geocoded. Fourth, the US Census Bureau does not collect data by transmission category (e.g., injection drug use) or sexual orientation (e.g., MSM). Our estimated HIV rates for black MSM and non-MSM are based on a denominator population of all black males aged 15 years and older. Although a previous study estimated that approximately 2.0% of the US population are MSM , we could not extrapolate the percentage of MSM in the 17 areas or in the US black male population. Finally, given that SDH information is not available at the level of the individual, we used census tract data as a surrogate for the environment in which persons with diagnosed HIV infection live.
This exploratory study shows the importance of understanding how SDH may frame HIV diagnosis rates among black men. These determinants may provide greater knowledge about where, when, and how to tailor and deploy prevention and care resources to populations at risk, particularly black males. Although, some of the patterns of HIV rates and SDH among black MSM and non-MSM in this study could not be easily explained, they highlight the current gaps in SDH research for this population, thereby identifying the need for additional research to better understand how SDH affects HIV risk.
Overall, we found differences in the patterns of HIV diagnosis rates for black MSM and non-MSM. These differences provide insight into the transmission of HIV infection in areas that reflect substantial disadvantage in education, housing, employment, and income. Differing patterns among these SDH also show the importance of examining several determinants together to better understand their associations with risk behavior in the population of interest. Although causality could not be determined in this study, this study opens up opportunity for further research to address structural differences that may affect HIV infection among black MSM and non-MSM.
The authors thank H. Irene Hall, PhD, John Brooks, MD, and Marie Morgan for their valuable comments and editorial suggestions on this article. Also, this manuscript would not have been possible without the contributions of the state and local health departments and the HIV surveillance programs that provided surveillance data to CDC.
The findings and conclusions in this paper are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.
Conceived and designed the experiments: ZG RS. Analyzed the data: ZG RS. Contributed reagents/materials/analysis tools: ZG RS. Wrote the paper: ZG LG RS LW ASJ. Intepretation of data: ZG LG RS LW ASJ.
- 1. Centers for Disease Control and Prevention (2013) HIV Surveillance Report, 2011. Available: http://www.cdc.gov/hiv/topics/surveillance/resources/reports/. Accessed 2013 Aug 1.
- 2. Centers for Disease Control and Prevention (2013) HIV Surveillance by Race/Ethnicity (slide set), 2011. Available: http://www.cdc.gov/hiv/library/slideSets/index.html. Accessed 2013 Sep 2.
- 3. Hall HI, An Q, Hutchinson AB, Sansom S (2008) Estimating the lifetime risk of a diagnosis of the HIV infection in 33 states, 2004–2005. J Acquir Immune Defic Syndr 49: 294–297. doi: 10.1097/qai.0b013e3181893f17
- 4. Millett GA, Flores SA, Peterson JL, Bakeman R (2007) Explaining disparities in HIV infection among black and white men who have sex with men: a meta-analysis of HIV risk behaviors. Aids 21: 2083–2091. doi: 10.1097/qad.0b013e3282e9a64b
- 5. Millett GA, Peterson JL, Wolitski RJ, Stall R (2006) Greater risk for HIV infection of black men who have sex with men: A critical literature review. American Journal of Public Health 96: 1007–1019. doi: 10.2105/ajph.2005.066720
- 6. Centers for Disease Control and Prevention (2013) HIV Testing and Risk Behaviors Among Gay, Bisexual, and Other Men Who Have Sex with Men-United States. MMWR Morbidity and mortality weekly report 62: 958.
- 7. Adimora AA, Schoenbach VJ (2002) Contextual factors and the black-white disparity in heterosexual HIV transmission. Epidemiology 13: 707–712. doi: 10.1097/00001648-200211000-00016
- 8. Adimora AA, Schoenbach VJ, Floris-Moore MA (2009) Ending the epidemic of heterosexual HIV transmission among African Americans. Am J Prev Med 37: 468–471. doi: 10.1016/j.amepre.2009.06.020
- 9. Diez-Roux AV (1998) Bringing context back into epidemiology: Variables and fallacies in multilevel analysis. American Journal of Public Health 88: 216–222. doi: 10.2105/ajph.88.2.216
- 10. Oparanozie A, Sales JM, DiClemente RJ, Braxton ND (2012) Racial Identity and Risky Sexual Behaviors Among Black Heterosexual Men. Journal of Black Psychology 38: 32–51. doi: 10.1177/0095798410397542
- 11. Raj A, Reed E, Santana MC, Welles SL, Horsburgh CR, et al. (2008) History of incarceration and gang involvement are associated with recent sexually transmitted disease/HIV diagnosis in African American men. J Acquir Immune Defic Syndr 47: 131–134. doi: 10.1097/qai.0b013e31815a5731
- 12. Gant Z, Lomotey M, Hall H, Hu X, Guo X, et al. (2012) A County-Level Examination of the Relationship Between HIV and Social Determinants of Health: 40 States, 2006-2008. The Open AIDS Journal 6: 1–7. doi: 10.2174/1874613601206010001
- 13. National Center for HIV/AIDS Viral Hepatitis STD and TB Prevention (U.S.) (2010) Establishing a holistic framework to reduce inequities in HIV, viral hepatitis, STDs, and tuberculosis in the United States an NCHHSTP white paper on social determinants of health, 2010. [Atlanta, GA]: National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention (CDC). Available: http://www.cdc.gov/socialdeterminants/docs/SDH-White-Paper-2010.pdf. Accessed 2013 Mar 12.
- 14. WHO Commission on Social Determinants of Health., World Health Organization. (2008) Closing the gap in a generation: health equity through action on the social determinants of health: Commission on Social Determinants of Health final report. Geneva, Switzerland: World Health Organization, Commission on Social Determinants of Health. 246 p. p. Available: http://whqlibdoc.who.int/publications/2008/9789241563703%5Feng.pdf. Accessed 2013 Mar 25.
- 15. Centers for Disease Control and Prevention (2013) Social Determinants of Health among Adults with Diagnosed HIV Infection in 18 Areas, 2005–2009. HIV Surveillance Supplemental Report. Available: http://www.cdc.gov/hiv/pdf/statistics_2005_2009_HIV_Surveillance_Report_vol_18_n4.pdf. Accessed 2013 Mar 25.
- 16. Raj A, Bowleg L (2012) Heterosexual risk for HIV among black men in the United States: a call to action against a neglected crisis in black communities. Am J Mens Health 6: 178–181. doi: 10.1177/1557988311416496
- 17. Bowleg L (2011) The “Forgotten”. Director 202: 6176.
- 18. Bowleg L, Teti M, Massie JS, Patel A, Malebranche DJ, et al. (2011) 'What does it take to be a man? What is a real man?': ideologies of masculinity and HIV sexual risk among Black heterosexual men. Culture Health & Sexuality 13: 545–559. doi: 10.1080/13691058.2011.556201
- 19. Raj A, Reed E, Welles SL, Santana MC, Silverman JG (2008) Intimate Partner Violence Perpetration, Risky Sexual Behavior, and STI/HIV Diagnosis Among Heterosexual African American Men. American Journal of Mens Health 2: 291–295. doi: 10.1177/1557988308320269
- 20. O'Leary A, Fisher HH, Purcell DW, Spikes PS, Gomez CA (2007) Correlates of risk patterns and Race/Ethnicity among HIV-Positive men who have sex with men. Aids and Behavior 11: 706–715. doi: 10.1007/s10461-006-9205-4
- 21. Warner L, Ghanem KG, Newman DR, Macaluso M, Sullivan PS, et al. (2009) Male Circumcision and Risk of HIV Infection among Heterosexual African American Men Attending Baltimore Sexually Transmitted Disease Clinics. Journal of Infectious Diseases 199: 59–65. doi: 10.1086/595569
- 22. Prejean J, Song RG, Hernandez A, Ziebell R, Green T, et al. (2011) Estimated HIV Incidence in the United States, 2006–2009. Plos One 6. doi: 10.1371/journal.pone.0017502
- 23. McDavid Harrison K, Kajese T, Hall HI, Song R (2008) Risk factor redistribution of the national HIV/AIDS surveillance data: an alternative approach. Public Health Rep 123: 618–627.
- 24. U.S. Census Bureau American Community Survey: 2005–2009 5-Year estimates (data files). Available: http://www.census.gov/acs/www/data_documentation/2009_release/. Published 2010 Dec 10. Accessed 2013 Mar 8.
- 25. Mayer KH, Wang L, Koblin B, Mannheimer S, Magnus M, et al. (2014) Concomitant Socioeconomic, Behavioral, and Biological Factors Associated with the Disproportionate HIV Infection Burden among Black Men Who Have Sex with Men in 6 US Cities. PLOS ONE 9: e87298. doi: 10.1371/journal.pone.0087298
- 26. Auerbach JD, Parkhurst JO, Caceres CF (2011) Addressing social drivers of HIV/AIDS for the long-term response: conceptual and methodological considerations. Glob Public Health 6 Suppl 3S293–309. doi: 10.1080/17441692.2011.594451
- 27. Poundstone KE, Strathdee SA, Celentano DD (2004) The social epidemiology of human immunodeficiency virus/acquired immunodeficiency syndrome. Epidemiologic Reviews 26: 22–35. doi: 10.1093/epirev/mxh005
- 28. Yau KKW, Wang K, Lee AH (2003) Zero-Inflated Negative Binomial Mixed Regression Modeling of Over-Dispersed Count Data with Extra Zeros. Biometrical Journal 45: 437–452. doi: 10.1002/bimj.200390024
- 29. Denning P, DiNenno E (2010) Communities in crisis: is there a generalized HIV epidemic in impoverished urban areas of the United States. International AIDS conference. Available: http://www.cdc.gov/hiv/risk/other/poverty.html. Accessed 2013 Nov 4.
- 30. Ramirez-Valles J (2002) The protective effects of community involvement for HIV risk behavior: a conceptual framework. Health Educ Res 17: 389–403. doi: 10.1093/her/17.4.389
- 31. Cohen D, Spear S, Scribner R, Kissinger P, Mason K, et al. (2000) "Broken windows" and the risk of gonorrhea. Am J Public Health 90: 230–236. doi: 10.2105/ajph.90.2.230
- 32. Fitzpatrick K, Gory ML (2010) Unhealthy cities: poverty, race, and place in America: Routledge.
- 33. Scribner RA, Johnson SA, Cohen DA, Robinson W, Farley TA, et al. (2008) Geospatial methods for identification of core groups for HIV/AIDS. Subst Use Misuse 43: 203–221. doi: 10.1080/10826080701690607
- 34. Millett GA, Jeffries WL, Peterson JL, Malebranche DJ, Lane T, et al. (2012) Common roots: a contextual review of HIV epidemics in black men who have sex with men across the African diaspora. Lancet 380: 411–423. doi: 10.1016/s0140-6736(12)60722-3
- 35. Millett GA, Peterson JL, Flores SA, Hart TA, Jeffries WLt, et al. (2012) Comparisons of disparities and risks of HIV infection in black and other men who have sex with men in Canada, UK, and USA: a meta-analysis. Lancet 380: 341–348. doi: 10.1016/s0140-6736(12)60899-x
- 36. Peterman TA, Lindsey CA, Selik RM (2005) This place is killing me: A comparison of counties where the incidence rates of AIDS increased the most and the least. Journal of Infectious Diseases 191: S123–S126. doi: 10.1086/425284
- 37. Siddiqi A, Nguyen QC (2010) A cross-national comparative perspective on racial inequities in health: the USA versus Canada. Journal of Epidemiology and Community Health 64: 29–35. doi: 10.1136/jech.2008.085068
- 38. Wilton L (2009) Men Who Have Sex With Men of Color in the Age of AIDS:The Sociocultural Contexts of Stigma, Marginalization, and Structural Inequalities. In: Smith KY, Rawlings MK, Ojikutu B, Stone V, editors. HIV/AIDS in US Communities of Color: Springer New York. pp. 179–211.
- 39. Savin-Williams RC (1994) Verbal and physical abuse as stressors in the lives of lesbian, gay male, and bisexual youths: associations with school problems, running away, substance abuse, prostitution, and suicide. Journal of consulting and clinical psychology 62: 261. doi: 10.1037//0022-006x.62.2.261
- 40. Varjas K, Dew B, Marshall M, Graybill E, Singh A, et al. (2008) Bullying in schools towards sexual minority youth. Journal of school violence 7: 59–86. doi: 10.1300/j202v07n02_05
- 41. Stein J, Nyamathi A, Ullman J, Bentler P (2007) Impact of Marriage on HIV/AIDS Risk Behaviors Among Impoverished, At-Risk Couples: A Multilevel Latent Variable Approach. AIDS and Behavior 11: 87–98. doi: 10.1007/s10461-005-9058-2
- 42. Waldron I, Hughes ME, Brooks TL (1996) Marriage protection and marriage selection—Prospective evidence for reciprocal effects of marital status and health. Social Science & Medicine 43: 113–123. doi: 10.1016/0277-9536(95)00347-9
- 43. Purcell DW, Johnson CH, Lansky A, Prejean J, Stein R, et al. (2012) Estimating the population size of men who have sex with men in the United States to obtain HIV and syphilis rates. Open AIDS J 6: 98–107. doi: 10.2174/1874613601206010098