Secondary Logo

Journal Logo

Prevention Research

Changes in Racial and Ethnic Disparities in Estimated Diagnosis Rates of Heterosexually Acquired HIV Infection Among Heterosexual Males in the United States, 2014–2018

McCree, Donna Hubbard PhD, MPH, RPha; Chesson, Harrell W. PhDb; Eppink, Samuel T. PhDb; Beer, Linda PhDa; Henny, Kirk D. PhD, MAa

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: December 15, 2020 - Volume 85 - Issue 5 - p 588-592
doi: 10.1097/QAI.0000000000002495
  • Free

Abstract

INTRODUCTION

Diagnoses of HIV infection attributed to heterosexual contact among male adults and adolescents aged ≥13 years decreased 13% from 2014 to 2018.1 Despite this progress, however, racial/ethnic disparities exist. In 2018, approximately 9% of diagnosed infections among men were attributed to heterosexual contact; by race/ethnicity, Black/African American men accounted for 61% of these infections.1 Monitoring and addressing HIV-related disparities are national HIV prevention goals.2 There are no data that measure changes in disparities in heterosexually acquired HIV infection diagnosis among heterosexual males in the United States. We used 12 absolute and relative measures of disparity (Table 1) to calculate racial/ethnic disparity changes in estimated HIV diagnosis rates among men with infection attributed to heterosexual contact [number of newly diagnosed HIV infections attributed to heterosexual contact per 100,000 non‐men who have sex with men (MSM) population] from 2014 to 2018.

TABLE 1. - List of Disparity Measures and Descriptions of Calculations
Disparity Measure Description of Calculation (See Footnotes for Definitions of Terms) Description of Measure
Relative disparity measures
 Black-to-White rate ratio The Black-to-White rate ratio is the HIV diagnosis rate among Blacks divided by the rate among Whites. A value of 1 indicates no disparity. Values above 1 indicate that rates are higher among Blacks than Whites, and the magnitude of the rate ratio reflects the relative degree of this disparity. The Hispanic-to-White ratio can be defined and interpreted in an analogous manner.
 Hispanic-to-White rate ratio
 Index of disparity (ID) The ID summarizes the average gap between the HIV diagnosis rate for each racial/ethnic group and the overall rate, expressed as a percentage of the overall rate. A value of 0 indicates no disparity, and higher values indicate greater disparity. The population-weighted ID is similar but summarizes the population-weighted average gap between the HIV diagnosis rate in each racial/ethnic and the overall rate.
 ID, weighted by population
 Mean log deviation (MLD) For each racial/ethnic group, the difference between the group's rate of HIV diagnosis and the population's rate of HIV diagnosis is calculated using natural logarithms. The MLD and Thiel index are population-weighted summary measures based on the sum of these calculations across all groups. A value of 0 indicates no disparity; higher values indicate greater disparity.
 Theil index
 Population attributable proportion (PAP) The PAP ranges from 0 (no disparity) to 1 (maximum disparity) and shows the proportion of HIV diagnoses attributable to disparity (ie, the proportion of HIV diagnoses that would be averted if all racial/ethnic groups had the same HIV diagnosis rate as the group with the lowest rate).
 Gini coefficient* The Gini coefficient ranges from 0 (no disparity) to 1 (maximum disparity). It reflects the degree to which the Lorenz curve (which plots the cumulative proportion of HIV diagnoses accounted for by the cumulative proportion of the population) differs from a diagonal line of equality in which HIV diagnosis rates are equal across all racial/ethnic groups.
Absolute disparity measures
 Black-to-White rate difference The Black-to-White rate difference is the HIV diagnosis rate among Blacks minus the rate among Whites. A value of 0 indicates no disparity. Values above 0 indicate that rates are higher among Blacks than Whites, and the magnitude of the rate difference reflects the absolute degree of this disparity. The Hispanic-to-White rate difference can be defined and interpreted in an analogous manner.
 Hispanic-to-White rate difference
 Absolute ID The only difference between the absolute ID and the traditional ID described above is that the absolute ID is not divided by the overall rate. The absolute ID summarizes the average gap between the HIV diagnosis rate for each racial/ethnic group and the overall rate. A value of 0 indicates no disparity, and higher values indicate greater disparity. The population-weighted absolute ID is similar but summarizes the population-weighted average gap between the HIV diagnosis rate for each racial/ethnic group differs from the overall rate.
 Absolute ID, weighted by population
Rate is the HIV diagnosis rate; the subscripts Black, White, and Hispanic indicate the racial/ethnic groups Black/African American, White, and Hispanic/Latino, respectively; Population indicates the population size; the subscript i indicates racial/ethnic group (American Indian/Alaska Native, Asian, Black/African American, Hispanic/Latino, Native Hawaiian/Other Pacific Islander, White, and Multiple Races); the subscript overall refers to all 7 racial/ethnic groups combined; actual cases is the number of HIV diagnoses reported for the given racial/ethnic group in a given year; hypothetical cases is the number of HIV diagnoses that would have occurred in a given racial/ethnic group in a given year if that group's HIV diagnosis rate was that of the group with the lowest HIV diagnosis rate in the given year.
Disparity measures were calculated for each year from 2014 through 2018. All calculations in this study focused on disparities in heterosexually acquired HIV infections in heterosexual males.
*For this calculation, the racial/ethnic groups were ranked according to their HIV diagnosis rates from lowest (i = 1) to highest (i =7). Y1 reflects the percentage of HIV diagnoses accounted for by group 1, Y2 reflects the percentage of HIV diagnoses accounted for by group 1 plus group 2, Y3 reflects the percentage of HIV diagnoses accounted for by group 1 plus group 2 plus group 3, and so on. Xi is calculated in an analogous manner and reflects the cumulative percentage of the population accounted for by group 1 through group I (X0 = 0, Y0 = 0, X7 = 1, and Y7 = 1).

METHODS

We used Atlas Plus (https://www.cdc.gov/nchhstp/atlas/index.htm) to obtain (1) annual case counts of diagnoses of HIV infection attributed to heterosexual contact among men and (2) population size for years 2014 through 2018 for male adults and adolescents aged ≥13 years (United States excluding territories) by the racial/ethnic group. We used an adjusted population denominator to calculate rates of diagnosed HIV infections acquired through heterosexual contact per 100,000 men. For 2014, the adjusted population denominator for each racial/ethnic group was calculated as the total male population aged ≥13 years for the given group in 2014 minus the estimated number of MSM for the given group in 2014. The estimated number of MSM for the given group was obtained from an updated application3 of a previously published method for estimating the MSM population size.4 For years 2015 through 2018, for which the MSM population size estimates were not available by race/ethnicity, we calculated the adjusted population denominators for a given group by assuming that MSM accounted for the same percentage of the population as in 2014.

Given the uncertainty in the estimates of the number of MSM by race/ethnicity, the method used to account for MSM was varied in 4 additional analyses: first, by applying lower bound estimates of the number of MSM; second, by applying upper bound estimates of the number of MSM; third, by assuming MSM accounted for 3.9% of the male population for all racial/ethnic groups, as opposed to the base case using group-specific national estimates of the percentage of men reporting same-sex contact in the past 5 years5; and fourth, by making no adjustment for MSM (such that the denominator included all men aged ≥13 years). The lower and upper bound estimates of the number of MSM by race/ethnicity were obtained from the same study from which the base case estimates of the number of MSM were obtained; the number of MSM in the base case corresponded to 3.6% of the overall male population with a range of 2.8%–4.6% across the racial/ethnic groups.3

Table 1 provides a summary of the 8 measures of relative and 4 measures of absolute disparity used to calculate changes in disparities from 2014 to 2018 among the following 7 racial/ethnic groups: American Indian/Alaska Native, Asian, Black/African American, Hispanic/Latino, Native Hawaiian/Other Pacific Islander, White, and Multiple Races. There is no broad consensus in the field on the best method for measuring and monitoring progress toward eliminating HIV-related health disparities. Until a broad consensus emerges on the most appropriate disparity measures to use, we think it is useful for studies of trends in disparities to include a wide range of measures from the published literature. Including 12 different measures not only allows for a more comprehensive assessment of trends in racial/ethnic disparities in heterosexually acquired HIV in men, but also helps to inform future deliberations regarding the most appropriate methods to use for assessing trends in health disparities.

The descriptions of the disparity measures were adapted from previous studies of health disparities.6–9 For all the disparity measures, a score of 0 corresponds to no disparity, and a higher score indicates greater disparity. Scores for the Gini coefficient and the Population Attributable Proportion (see Table 1) range from 0 (no disparity) to 1 (maximum disparity); for the remaining measures of disparity, there is no established upper bound.

RESULTS

Table 2 presents the number of HIV diagnoses and estimated HIV diagnosis rates per 100,000 population among male adults and adolescents ≥13 years with infection attributed to heterosexual contact by race/ethnicity for years 2014 through 2018. Black men had the highest estimated diagnosis rate in each of the years; the lowest rates were among Native Hawaiian/Other Pacific Islanders in 2014 and White men in 2015–2018. Estimated diagnosis rates decreased from 2014 to 2018 among Black men (13.90–11.39) and men of multiple races (5.19–2.23). Rates also decreased among Hispanics/Latinos (3.18–2.50) and American Indians/Alaskan Natives (1.11–0.85). Estimated diagnosis rates increased (0.75–0.82) among Asians and varied among Native Hawaiians/Other Pacific Islanders, with an increase from 0.47 in 2014 to 1.73 in 2018. The overall trends for these 2 groups are less clear because of the small number of HIV diagnoses each year for each group. A large relative increase in the HIV diagnosis rate among Native Hawaiians/Other Pacific Islanders was observed because of an increase from 1 HIV diagnosis in 2014 to 4 diagnoses in 2018. Rates among White men remained stable over the time period, 0.53 in 2018 compared with 0.53 in 2014.

TABLE 2. - HIV Diagnoses (n) and Estimated Rates* (per 100,000) Among Male Adults and Adolescents aged ≥13 Years With Infection Attributed to Heterosexual Contact, 2014 Through 2018, United States, by Race/Ethnicity
2014 2015 2016 2017 2018 Percent Change in Rates, 2014–2018
n Rate n Rate n Rate n Rate n Rate
American Indian/Alaska Native 10 1.11 10 1.10 9 0.98 8 0.86 8 0.85 −23.1
Asian 48 0.75 59 0.90 59 0.87 38 0.54 59 0.82 8.3
Black/African American 2027 13.90 2094 14.19 2037 13.65 1792 11.88 1736 11.39 −18.0
Hispanic/Latino 644 3.18 599 2.89 605 2.84 529 2.43 558 2.50 −21.3
Multiple races 99 5.19 81 4.09 83 4.05 64 3.01 49 2.23 −57.1
Native Hawaiian/Other Pacific Islander 1 0.47 6 2.77 2 0.90 4 1.77 4 1.73 267.0
White 425 0.53 410 0.51 472 0.58 437 0.54 431 0.53 0.9
Total 3254 2.60 3259 2.58 3267 2.56 2872 2.24 2845 2.20 −15.5
*Rates were calculated as the number of diagnoses of HIV infection in men attributed to heterosexual contact divided by the estimated number of males aged ≥13 years who are non-MSM, multiplied by 100,000.
Race/ethnicity groups are mutually exclusive; Hispanics/Latinos can be of any race.

Outcomes from the disparity measures are provided in Table 3. The main finding of the analysis is that all 12 disparity measures were lower in 2018 than in 2014, suggesting racial/ethnic disparities in estimated HIV diagnosis rates among males with infection attributed to heterosexual contact decreased over this time frame. All of the absolute measures of disparity decreased by at least 18.8%. Among the relative measures of disparity, the largest decreases were observed in the Hispanic/Latino-to-White rate ratio, the Black-to-White rate ratio, and the Index of Disparity.

TABLE 3. - Relative and Absolute Changes* in Racial/Ethnic Disparities of Estimated HIV Diagnosis Rates Among Male Adults and Adolescents aged ≥13 Years With Infection Attributed to Heterosexual Contact, 2014 Through 2018, United States
2014 2015 2016 2017 2018 Percent Change, 2014–2018
Relative disparity measures
 Black-to-White rate ratio 26.5 28.1 23.5 22.1 21.5 −18.8
 Hispanic-to-White rate ratio 6.1 5.7 4.9 4.5 4.7 −22.0
 Index of disparity 120.9 104.4 110.2 101.3 93.5 −22.7
 Index of disparity, weighted by population 111.5 110.9 106.8 105.4 103.3 −7.4
 Mean log deviation 0.868 0.883 0.792 0.772 0.740 −14.8
 Theil index 0.875 0.900 0.836 0.825 0.791 −9.6
 Population attributable proportion 0.819 0.804 0.773 0.760 0.759 −7.3
 Gini coefficient 0.647 0.654 0.629 0.621 0.613 −5.3
Absolute disparity measures
 Black-to-White rate difference 13.4 13.7 13.1 11.3 10.9 −18.8
 Hispanic-to-White rate difference 2.7 2.4 2.3 1.9 2.0 −25.7
 Absolute index of disparity 314.4 269.3 282.5 226.5 205.5 −34.6
 Absolute index of disparity, weighted by population 289.9 286.1 273.8 235.7 227.1 −21.7
*For all measures, a higher score indicates greater disparity.

The results did not change in any meaningful way when the method of accounting for MSM in the population denominators was varied. The percentage change in each disparity measure from 2014 to 2018 varied by less than one-tenth of a percentage point for all measures, except the population attributable proportion, which varied by four-tenths of a percentage point.

DISCUSSION

Results from multiple disparity measures provide evidence that racial/ethnic disparities in estimated HIV diagnosis rates among males with infection attributed to heterosexual contact decreased in 2018 compared with 2014. The decreases ranged from 18.8% to 34.6% among the 4 absolute disparity measures and from 5.3% to 22.7% among the 8 relative disparity measures. However, despite the decrease, disparities still exist. These results support a need for tailored strategies and interventions to address the disproportionate rates of HIV infection among heterosexual Black males to continue progress toward reducing the disparity. Available data suggest that some heterosexual Black males are at higher risk for HIV infection compared with heterosexual males of other races/ethnicities because of social and structural factors such as lower socioeconomic status, higher incarceration rates, medical mistrust, greater degree of HIV stigma, and lack of available and lower use of HIV prevention, screening, and treatment services.10,11 There is also a paucity of behavioral interventions specifically designed for and that address the sociocultural factors associated with HIV risk among heterosexual males and specifically, Black males.12 Strategies to address the disparity should ensure that all heterosexual men, including Black men, who are at risk for HIV infection, have knowledge of and access to effective prevention methods, such as condoms; behavioral interventions to reduce risk behaviors, diagnosis, and treatment of other sexually transmitted infections; and pre-exposure prophylaxis, if appropriate. Strategies could also include linking at risk men to employment services, vocational rehabilitation, and job centers; support for policies that reduce incarceration rates; and development of campaigns that decrease HIV stigma.13

These data are subject to limitations. First, the annual case counts of new HIV diagnoses from national surveillance data are likely affected by some under-reporting and reporting delays. Second, we used population estimates for MSM to determine the heterosexual population estimates for each racial/ethnic group and assumed the MSM population estimates were uniform for the years 2014 through 2018. Our results, however, are robust to a range of alternative methods for accounting of the MSM population. Third, because diagnoses of HIV among heterosexual men who inject drugs are attributed to injection drug use and not to heterosexual contact,1 the comparison of rates of heterosexually acquired HIV by race/ethnicity could be biased by differences in injection drug use and in reporting of injection drug use by race/ethnicity. However, such differences should not alter the general patterns reported in the trends in racial/ethnic disparities in heterosexually acquired HIV rates, particularly if these differences are fairly stable over time. Missing risk factors are imputed and add. to uncertainty.

Notwithstanding the limitations, results indicate progress toward reducing racial/ethnic disparities in new HIV diagnoses among males with infection attributed to heterosexual contact. The results also provide support for tailored HIV prevention efforts to further decrease disparities, particularly between heterosexual Black men at risk for HIV infection and heterosexual men of other races/ethnicities.

REFERENCES

1. Centers for Disease Control and Prevention. HIV Surveillance Report, 2018. Vol. 31. Atlanta, GA: CDC; 2020. Available at: http://www.cdc.gov/hiv/library/reports/hiv-surveillance.html. Accessed June 1, 2020.
2. National HIV/AIDS Strategy for the United States 2020. Available at: https://www.hiv.gov/federal-response/national-hiv-aids-strategy/nhas-update. Accessed June 1, 2020.
3. Grey JA, Bernstein KT, Sullivan PS, et al. Rates of primary and secondary syphilis among white and Black non-Hispanic men who have sex with men, United States, 2014. J Acquir Immune Defic Syndr. 2017;76:e65–e73.
4. Grey JA, Bernstein KT, Sullivan PS, et al. Estimating the population sizes of men who have sex with men in US states and counties using data from the American Community Survey. JMIR Public Health Surveill. 2016;2:e14.
5. Purcell DW, Johnson CH, Lansky A, et al. Estimating the population size of men who have sex with men in the United States to obtain HIV and syphilis rates. Open AIDS J. 2012;6:98–107.
6. Harper S, Lynch J, Meersman SC, et al. An overview of methods for monitoring social disparities in cancer with an example using trends in lung cancer incidence by area-socioeconomic position and race-ethnicity, 1992–2004. Am J Epidemiol. 2008;167:889–899.
7. An Q, Prejean J, Hall HI. Racial disparity in U.S. diagnosis of acquired immune deficiency syndrome, 2000–2009. Am J Prev Med. 2012;43:461–466.
8. Rossen LM, Schoendorf KC. Measuring health disparities: trends in racial-ethnic and socioeconomic disparities in obesity among 2- to 18-year old youth in the United States, 2001–2010. Ann Epidemiol. 2012;22:698–704.
9. Harper S, Lynch J. Selected Comparisons of Measures of Health Disparities: A Review Using Databases Relevant to Healthy People 2010 Cancer-Related Objectives. NCI Cancer Surveillance Monograph Series, Number 7. Bethesda, MD: National Cancer Institute; 2007. NIH Pub No. 07-6281.
10. Bout ML, Docena JP, Ratemo BK, et al. Beyond race and place: distal sociological determinants of HIV disparities. PLoS One. 2014;9:e91711.
11. Bogart LM, Derose KP, Kanouse DE, et al. Correlated of HIV Testing among African American and Latino church congregants; the role of HIV stigmatizing attitudes and discussions about HIV. J Urban Health. 2015;92:93–107.
12. Henny KD, Crepaz N, Lyles CM, et al. Efficacy of HIV/STI behavioral interventions for heterosexual African American men in the United States: a meta-analysis. AIDS Behav. 2012;16:1092–1114.
13. Beer L, McCree DH, Jeffries WL, et al. A review of recent Centers for Disease Control and Prevention activities to reduce HIV stigma in the United States, J Int Assoc Provid AIDS Care. 2019;18:1–5.
Keywords:

heterosexual males; HIV diagnosis; disparities; social determinants of health

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.