Elsevier

The Lancet HIV

Volume 4, Issue 7, July 2017, Pages e311-e320
The Lancet HIV

Articles
Sources of racial disparities in HIV prevalence in men who have sex with men in Atlanta, GA, USA: a modelling study

https://doi.org/10.1016/S2352-3018(17)30067-XGet rights and content

Summary

Background

In the USA, men who have sex men (MSM) are at high risk for HIV, and black MSM have a substantially higher prevalence of infection than white MSM. We created a simulation model to assess the strength of existing hypotheses and data that account for these disparities.

Methods

We built a dynamic, stochastic, agent-based network model of black and white MSM aged 18–39 years in Atlanta, GA, USA, that incorporated race-specific individual and dyadic-level prevention and risk behaviours, network attributes, and care patterns. We estimated parameters from two Atlanta-based studies in this population (n=1117), supplemented by other published work. We modelled the ability for racial assortativity to generate or sustain disparities in the prevalence of HIV infection, alone or in conjunction with scenarios of observed racial patterns in behavioural, care, and susceptibility parameters.

Findings

Race-assortative mixing alone could not sustain a pre-existing disparity in prevalence of HIV between black and white MSM. Differences in care cascade, stigma-related behaviours, and CCR5 genotype each contributed substantially to the disparity (explaining 10·0%, 12·7%, and 19·1% of the disparity, respectively), but nearly half (44·5%) could not be explained by the factors investigated. A scenario assessing race-specific reporting differences in risk behaviour was the only one to yield a prevalence in black MSM (44·1%) similar to that observed (43·4%).

Interpretation

Racial assortativity is an inadequate explanation for observed disparities. Work to close the gap in the care cascade by race is imperative, as are efforts to increase serodiscussion and strengthen relationships among black MSM particularly. Further work is urgently needed to identify other sources of, and pathways for, this disparity, to integrate concomitant epidemics into models, and to understand reasons for racial differences in behavioural reporting.

Funding

The Eunice Kennedy Shriver National Institute of Child Health and Development, the National Institute of Allergy and Infectious Diseases, the National Institute of Minority Health and Health Disparities, and the National Institute of Mental Health.

Introduction

Men who have sex with men (MSM) account for most new HIV diagnoses in the USA.1 Concentrations of HIV infections among MSM are highest in southern USA, and Georgia is the only state that ranks in the top five for both percentage of MSM and absolute number of MSM living with a diagnosis of HIV.2 The HIV epidemic in MSM is characterised by marked, long-standing racial disparities between black and white populations: in a study in Atlanta, GA, the estimated prevalence was 43% in black MSM and 13% in white MSM, a 3·3-fold disparity.3, 4

Several explanations for the disproportionate effect of HIV among black MSM have been offered and thoroughly reviewed,5, 6 including distal structural factors such as poverty, stigma, and institutionalised racism. However, the proximal causal pathways through which these factors enact disparate epidemics have proven challenging to elucidate. For structural factors to cause differentials in incidence of HIV infection, they must mediate one or both of two more proximal causes: the frequency with which HIV-negative individuals are potentially exposed to HIV, and the probability of transmission if exposed, which comprises factors associated with either the HIV-negative partner (eg, circumcision status) or the HIV-positive partner (eg, viral suppression).

Much work about racial disparities in HIV infection among MSM focuses on self-reported individual risk behaviours (eg, number of sex partners, substance use), and thus has limited explanatory power, because most behaviours are not more common in black MSM than white MSM.5, 6 Inadequate HIV suppression among HIV-positive partners of HIV-negative MSM places them at increased risk of acquisition, with racial disparities in the care continuum probably contributing to racial disparaties in incidence.7 Few studies of susceptibility differences have been done, although the CCR5Δ32 mutation, which is more prevalent in populations of European ancestry than of African ancestry, is protective against infection.8 Differences in sexual-network properties are another potential set of explanations, although the primary evidence for meaningful differences by race is mixed.4, 6, 9, 10

Stigma related to sexuality and HIV can also affect the health of black MSM and influence their apparent and real HIV risks by shortening partnership durations and suppressing discussion of HIV status before sex (ie, serodiscussion). Stigma and mistrust of research might cause under-reporting of risk-enhancing behaviours by black MSM: evidence suggests that correction for under-reporting leads to equivalent, rather than lower, risks for black MSM compared with white MSM.11

Research in Context

Evidence before this study

Disparities in HIV infection between black and white populations have been a hallmark of the US HIV epidemic for decades, but to account for them fully has proved challenging. We searched PubMed with combinations of the keywords “HIV”, “AIDS”, “MSM”, “men who have sex with men”, “disparity”, “disparities”, “race”, “racial”, “ethnic”, “ethnicity”, “black”, “African American”, “white”, and “Caucasian” for articles published in English on or before June 5, 2016 (the date of our final search). We focused on studies that discussed either the empirical evidence for, or sources and causal mechanisms of, disparities in incidence and prevalence of HIV infection between black and white men who have sex with men (MSM) in the USA. We found many empirical studies over decades that show the existence of these disparities, and several critical literature reviews and meta-analyses that show the existence and magnitude of many potential sources, including racial assortativity, proximal differences in the care cascade, sexual networking, and biological cofactors, and distal factors such as stigma and poverty. However, quantification of the magnitude of disparity that the many proximal sources could generate or sustain over time necessitates dynamic modelling, which the authors of many of the papers acknowledge and call for. One group of investigators developed a data-driven network model to assess disparities in very young (age 16·0–21·8 years) MSM during a 15 year timeframe, but they did not attempt to partition the detected disparities attributed to each of the proposed sources, or to estimate the proportion unexplained. Additionally, the authors of a series of theoretical modelling papers laid out the expected relations between the generation and maintenance of disparities, although these relations have not been verified in the specific context of racial disparities in HIV infection in US MSM.

Added value of this study

Our study is the first to show in a dynamic model that a combination of many proposed proximal sources of racial disparities in HIV among US MSM generate a reverse disparity. We quantified the amount that each component, in combination with race assortativity, contributes to observed or reverse disparities, with the care cascade, biological cofactors, HIV serostatus disclosure, and some aspects of partnership dynamics each contributing a substantial amount to the observed disparity. We showed that misclassification within other behavioural components would be sufficient to generate the observed prevalence of HIV among black MSM. We examined the power of these causes to generate and sustain disparities, and showed that the ability to sustain pre-existing disparities on their own is short-lived.

Implications of all the available evidence

We provide the most thorough assessment so far of the ability for proposed sources to either generate or sustain observed racial disparities in HIV among US MSM in the long term. High incidence of HIV infection among black MSM is one of the most pressing public health concerns, and reduction of this burden is a priority in the US National HIV/AIDS Strategy. We provide novel evidence for the relative importance of the proximal sources to this high burden, a necessary first step in determining the effectiveness of efforts to reduce that burden. We also show how much these sources cannot explain, highlighting the areas where more evidence is crucially needed.

Race-assortative mixing (the tendency to select same-race partners) can enable disparities arising from other factors to remain concentrated within groups. Partner race is a strong explanatory factor in incidence studies,12, 13 suggesting a role in facilitating persistence for existing disparities.6, 7, 10 However, race-assortative mixing alone probably cannot sustain disparities arising from other sources. Modelling theory, including work on HIV and other sexually transmitted infections (STIs), predicts that a given epidemic typically heads towards a fixed equilibrium, irrespective of its current state.14 For each subpopulation, a specific prevalence represents the point at which incident infections are balanced by deaths among people with HIV. We have built a web tool that shows this concept to build further intuition in the context of a simpler model (appendix). The theory suggests that, if race-assortative mixing alone cannot generate a disparity then it also cannot sustain a disparity indefinitely or cause increasing disparities over time. However, exceptions to this theory exist, and epidemic dynamics can be slow for lifelong infections like HIV.15 The ability for reported levels of race-assortative mixing, in combination with other reported proximal factors, to generate racial disparities in HIV in MSM or maintain disparities over the long timeframes that they have been reported, has not been assessed. A study in which data-driven models were used to examine this question for young MSM over 15 years showed that racial differences in incidence of HIV infection narrowed over the course of the simulated epidemic.16

Dynamic network models are well positioned to assess multiple proposed mechanisms for the ability to generate or sustain disparities, and have been called for.7 We implemented a model parameterised by data from young MSM in Atlanta to answer two questions. First, assuming that disparities emerged from some unmeasured differences early in the epidemic, how long could they be sustained under reported race-assortative mixing, with or without other recorded racial differences? Second, how much of the 3·3-fold racial disparity in prevalence of HIV infection could be generated by specific measured proximal factors, alone or in combination? We also further explore our findings for the second question to consider the potential effect of behaviour misclassification caused by societal factors.

Section snippets

Model design

We used dynamic, stochastic network models that extend previous work.17 We began with 10 000 MSM, each of whom possessed fixed (eg, race, circumcision status, CCR5Δ32 status) and dynamic (eg, age, infection status) attributes. Men with HIV had additional dynamic attributes (eg, diagnosis status, treatment status, stage, viral load). We simulated multiple transitions for each man by week, concurrently with relational dynamics (panel). Models simulated three contact networks for anal intercourse:

Results

For our null model beginning with observed race-specific prevalence, incidence was initially higher in black MSM because of the difference in prevalence in partners. Subsequently, incidence and prevalence converged, with prevalence disparity falling to half the initial disparity in 6·7 years, and by over 90% in 22 years (figure 1A, 1B). When all parameters were set to observed race-specific values (ie, as observed), disparities disappeared even more quickly (the disparity in prevalence fell by

Discussion

We used data from Atlanta-based studies focused on racial disparities in HIV prevalence and a previously described model structure17 to assess whether the hypotheses that have been put forward in scientific literature (average differences in networks of sexual relationships, sexual behaviours within relationships, the care cascade, disclosure, and CCR5 genotype frequencies) are sufficient to explain the observed disparities. We found that these hypothesised explanatory factors accounted for, at

References (30)

  • M Marmor et al.

    Homozygous and heterozygous CCR5-Delta32 genotypes are associated with resistance to HIV infection

    J Acquir Immune Defic Syndr

    (2001)
  • HV Tieu et al.

    Do sexual networks of men who have sex with men in New York City differ by race/ethnicity?

    AIDS Patient Care STDs

    (2016)
  • B Mustanski et al.

    The role of geographic and network factors in racial disparities in HIV among young men who have sex with men: an egocentric network study

    AIDS Behav

    (2015)
  • R Garofalo et al.

    Incidence of HIV infection and sexually transmitted infections and related risk factors among very young men who have sex with men

    J Acquir Immune Defic Syndr

    (2016)
  • CP Simon et al.

    Reproduction numbers and the stability of equilibria of si models for heterogeneous populations

    Siam J Appl Math

    (1992)
  • Cited by (0)

    View full text