New Faces of HIV Infection: Age,Race,and Timing of Entry into HIV Care in the Southeastern United States

Abstract

Among younger men who have sex with men (MSM), the incidence of HIV is rising nationally. Of the 281 persons who entered into care at a large HIV clinic in the southeastern United States in 2010 to 2012, 78 (27.8%) were <25 years old at the time of diagnosis. Those in the younger group were more likely than those aged ≥25 to be black (59.0% versus 37.4%), MSM (78.2% versus 55.2%), and to have a longer median time from diagnosis to entry into care (71 versus 53 days; P < .05 each). In adjusted survival analysis, persons of black race were less likely to enter care after diagnosis than those of nonblack race (hazard ratio = 0.75, P = .02). Young MSM represent an important target population for prevention and HIV testing interventions, and there is a need to shorten the time from diagnosis to linkage to care, particularly in persons aged <25 and of black race.

Keywords

HIV young adults race entry into care care continuum

Introduction

Since its peak in 1992, the incidence of HIV infection in the United States has been steadily declining due to advances in prevention and treatment. Recent data from the National HIV Surveillance System of the Centers for Disease Control and Prevention (CDC) revealed an overall decrease by 33.2% in the annual diagnosis rate, from 24 new diagnoses per 100 000 population in 2002 to 16.1 per 100 000 in 2011.¹ Significant progress has been made, as reflected in declining incidence rates over the 10-year period, in women and across multiple racial and ethnic groups, although African Americans continue to have disproportionately higher rates of HIV. Compared to 2002, there were 35% fewer diagnoses among injection drug users in 2011 and 70% fewer persons who acquired HIV through heterosexual contact.

In contrast, however, there was a rise of 38% in the incidence of HIV diagnosed in young persons, aged 13 to 24, compared with those aged 25 and over. The trend was even more pronounced in persons aged 13 to 24 whose transmission route was male-to-male sexual contact, with a 133% increase in the number of diagnoses in this population in 2011 compared to 2002. Among men who have sex with men (MSM), those less than 25 years of age comprised 27% of new diagnoses in 2011.¹ The percentage of young persons affected may be even higher than these statistics suggest, as it is estimated that only 49% of persons <25 living with HIV in the United States have been diagnosed, compared with 88% of persons aged 25 and older.² These data suggest an urgent need to target screening and prevention efforts toward this younger population in the United States.

Understanding the characteristics of the young HIV-infected population in this country may help elucidate some of the reasons for this increased incidence and identify potential areas for intervention, particularly related to linkage to care.^3
–5 Although some of the characteristics of this younger cohort have been reported, we sought to further evaluate the lag between diagnosis and entry to care in this group after diagnosis with HIV. Here we present a retrospective study of patients presenting for initial HIV care at a large southeastern HIV/AIDS clinic. We identified new patients (never before enrolled in care) attending at least 1 provider visit for HIV care at the Vanderbilt Comprehensive Care Clinic (VCCC) from October 2010 to June 2012. Those <25 years old at the time of HIV diagnosis were compared with those ≥25 years old.

Methods

The VCCC in Nashville is the largest HIV/AIDS clinic in Tennessee, with approximately 2900 patients in active care, and its patient population has been described previously.⁶ This study evaluated all 281 persons who presented for at least 1 provider visit from October 2010 to June 2012 for initial care after HIV diagnosis. Demographic, clinical, medication information, health insurance, HIV testing history, education level, substance use, and laboratory data are routinely collected from the electronic medical record and validated through standardized chart abstraction procedures.

Age was calculated at HIV diagnosis date and was dichotomized at the 25-year threshold to describe the differences between age-groups compatible with CDC descriptions of vulnerable young populations.^4,7 Follow-up time began at diagnosis and concluded at the first provider visit at the VCCC. Race/ethnicity (black versus nonblack), sex, HIV transmission risk factor (MSM versus non-MSM), prior negative HIV test, CD4 count at first provider visit, and active substance abuse as determined by chart review of provider notes and laboratory screening were included as potential confounders of the relationship between age at diagnosis and entry into care. Secondary analyses examining disparities in delayed linkage (CD4 count <200 cells/mm³ at enrollment), antiretroviral therapy (ART) receipt (regimen containing ≥3 antiretroviral agents ≤1 year after enrollment), and viral suppression (HIV-1 RNA <200 copies/mL ≤1 year after enrollment) were also conducted.⁸ Race/ethnicity and HIV transmission risk factor were dichotomized according to highly vulnerable populations with greater burdens of disease, both in the southeastern United States and nationally.^9
–11

Differences across age categories were detected by χ² test of differences in proportion for categorical variables and by the Wilcoxon rank sum test of differences in medians for continuous variables. Adjusted hazard ratios for entering care and median times from diagnosis to first provider visit by age category were assessed using Cox proportional hazards models, accounting for race, sex, HIV risk factor, prior HIV-negative test, CD4 count at first provider visit, and active substance abuse. Scaled Schoenfeld residuals were used to assess the proportionality assumption by age category in the fully adjusted model. Regression models accommodating interactions between age category and race or with HIV risk factor were also used to assess potential differences in entry to care across multiple strata. Secondary analyses were conducted using modified Poisson regression for binary outcomes.¹² Additional analyses extrapolating adjusted model predictions to older age-groups, where our sample was sparse, modeled age using restricted cubic splines with 3 knots.¹³ All analyses were conducted in Stata 12 (StataCorp, College Station, Texas).

Results

Of the 281 persons included, 27.8% were <25 years old at HIV diagnosis (Table 1). Those in the younger group had a similar percentage of males (85.9%) as in the older group (80.3%; P = .28), though they had a higher percentage of black patients (59.0% versus 37.4%, respectively; P < .01). Those <25 were more likely than those ≥25 years to have MSM as HIV risk factor (78.2% versus 55.2%, respectively; P < .01) and to have higher median CD4 counts at entry to care (425 cells/mm³ versus 359.5, respectively; P < .01), though they were not significantly more likely to report a prior negative HIV test (59.7% versus 45.5%, respectively; P = .11). Rates of active substance abuse were similar in the 2 age groups (21.8% in the younger group versus 22.2%, P = .95; Table 1). The median number of days between positive HIV diagnosis and entry to the clinic was significantly lower among older versus younger (53 versus 71 days; P = .02) and nonblack versus black (51 versus 75 days; P < .01) patients.

Table 1.

Characteristics of Entrants to Care at Vanderbilt Comprehensive Care Clinic from October 2010 to June 2012, by Age Category.

	Age ≥ 25 Years		Age < 25 Years
	N = 203	Col. %	N = 78	Col. %	P Value^a
Race					<.01
Nonblack	127	62.6%	32	41.0%
Black	76	37.4%	46	59.0%
Sex					.28
Male	163	80.3%	67	85.9%
Female	40	19.7%	11	14.1%
Risk factor					<.01
Non-MSM	91	44.8%	17	21.8%
MSM	112	55.2%	61	78.2%
Prior negative HIV test					.11
Yes	92	45.5%	46	59.7%
No	58	28.7%	16	20.8%
Unknown	53	25.7%	16	19.5%
Active substance abuse					.95
No	158	77.8%	61	78.2%
Yes	45	22.2%	17	21.8%
CD4 count^b	359.5	(111-530)	425	(342-595)	<.01
Days from diagnosis to clinic entry^c	53	(34-113)	71	(43-223)	.02

Abbreviation: MSM, men who have sex with men.

^aχ² test for difference in proportions used for categorical variables and Wilcoxon rank sum test for difference in medians used for continuous variables.

^bBaseline CD4 count presented as median (interquartile range) in units of cells/mm³

^cDays from HIV diagnosis to clinic entry presented as median (interquartile range).

In adjusted regression analyses including age, black race, MSM status, and active substance use, there was a trend toward longer time between diagnosis and first provider visit for those <25 years (median = 69 days) compared to those ≥25 (median = 58 days; Figure 1), though this did not meet statistical significance (hazard ratio [HR] = 0.79; 95% confidence interval [CI]: 0.61-1.04; P = .09; Table 2). In the adjusted regression analysis, persons of black race were less likely to enter into care at any given time point after diagnosis than those of nonblack race (HR = 0.75; 95% CI: 0.59-0.95; P = .02), which remained statistically significant after accounting for active substance abuse and HIV transmission risk factor (Table 2). There was no evidence for violation of the proportionality assumption in the final regression model (global P = .89; Figure 2). Extrapolating model predictions to older age-groups, among whom our data were sparse, there was a dose–response increase in the adjusted probability of entering care with older age above 50, with 50-year-olds (HR = 1.40; 95% CI: 1.01-1.94), 60-year-olds (HR = 1.96; 95% CI: 1.20-3.21), and 70-year-olds (HR = 2.76; 95% CI: 1.22-6.23), all significantly more likely to enter care at a given time point compared to 25-year-olds. Despite this finding, there remained no significant difference between 25-year-olds and those between ages 25 and 50 or below age 25.

Figure 1.

Adjusted survival curves for time from HIV diagnosis (Dx) to first clinic visit, by age category, accounting for race, sex, HIV risk factor, and active substance abuse.

Figure 2.

Diagnosis of the proportionality assumption in a Cox regression model, using the scaled Schoenfeld residuals. The horizontal line at 0 on the Y-axis indicates that hazards are proportional by exposure of interest (age) across the study period.

Table 2.

Hazard Ratios for Entry to Care After HIV Diagnosis, by Demographic and Risk Characteristics.^a

	HR	95% CI	P Value	aHR^b	95% CI	P Value
Age
≥25 (years)	Ref			Ref
<25 (years)	0.80	(0.61-1.04)	.10	0.79	(0.60-1.04)	.09
Race
Nonblack	Ref			Ref
Black	0.74	(0.58-0.94)	.01	0.75	(0.59-0.95)	.02
Sex
Male	Ref			Ref
Female	1.03	(0.75-1.40)	.87	0.96	(0.70-1.31)	.78
Risk factor
Non-MSM	Ref			Ref
MSM	1.11	(0.87-1.41)	.42	1.16	(0.90-1.49)	.24
Active substance abuse
No	Ref			Ref
Yes	1.01	(0.76-1.34)	.97	1.00	(0.75-1.33)	.99

Abbreviations: aHR, adjusted hazard ratio from multivariable Cox regression model including all covariates; CI, confidence interval; HR, crude hazard ratio from bivariate Cox regression model; MSM, men who have sex with men; Ref, reference category.

^aBoldface estimates are statistically significant (P < .05).

^bEstimates for all factors except for sex are from adjusted model including age, race, risk, and active substance abuse; the adjusted estimate for sex is derived from a model including all other factors but excluding MSM HIV risk factor, since male sex and MSM risk factor are collinear.

In effect of modification models that were fit using interactions of age by race and of age by HIV risk factor, differences by black race persisted, with those who were ≥25 years old and black having similar delays in entering care to all patients who were <25 (compared to those who were ≥25 and nonblack; Figure 3). There were no differences detected in age-by-MSM interaction models (Figure 4).

Figure 3.

Adjusted survival curves for time from HIV diagnosis to first clinic visit, with age by race interactions, accounting for sex, HIV risk factor, and active substance abuse.

Figure 4.

Adjusted survival curves for time from HIV diagnosis to first clinic visit, with age by risk interactions, accounting for sex, race, and active substance abuse.

In secondary analyses, adjusting for all factors, there was an age disparity in delayed entry, with patients <25 at HIV diagnosis significantly less likely to have a CD4 count <200 cells/mm³ at enrollment (risk ratio [RR] = 0.39; 95% CI: 0.20-0.76; P = .01) compared to those ≥25 years old. There were no disparities in delayed entry by any other factor, nor were there significant associations of any kind between other factors and outcomes of delayed ART receipt or viral suppression.

Discussion

In this study from a large single center in Nashville, persons who were <25 years of age and presenting for initial HIV care were more likely to be MSM and to have higher median CD4 counts at entry to care versus those who were ≥25 years of age; this pattern of younger patients presenting with higher CD4 counts than older patients remained in adjusted multivariable regression. The younger age-group had a longer time to entry into care compared to the older group, although there was no statistical significance after adjusting for race, risk factor, and substance use. Other studies have shown that younger persons with HIV infection are less likely to be diagnosed, retained in care, and virally suppressed, nationally, though we did not observe significant age patterns related to ART receipt and viral suppression outcomes in this study population.^14
–16 In fact, our findings related to linkage to care among those >50 years at the time of HIV diagnosis match the findings at other stages of the HIV care continuum in other cohorts—namely, that older patients were more likely to be successfully engaged in care than their younger counterparts. A recent US surveillance study of HIV diagnosis showed that only 73% of newly diagnosed persons aged 13 to 24 were linked to care within 3 months of diagnosis, versus 84% of those aged ≥35. Of all persons with HIV in the United States in 2011, only an estimated 22% of those aged 18 to 24 were engaged in care versus 40% across all age-groups. Most notably, only an estimated 13% of those aged 18 to 24 with HIV were virally suppressed, versus 30% overall.² Our study captured the characteristics of patients who are successfully linked to care, regardless of delay; future studies would ideally look at a broader group, including those who are diagnosed but not linked to care.

In our study, persons in the younger age-group were more likely to be of black race than the older age-group. In the United States, HIV incidence has been increasing in younger black MSM. A meta-analysis of studies in the United States, United Kingdom, and Canada found that young (<30), black MSM were more likely to have had younger age at debut of sexual activity, to be sexually active with older partners, and to have a history of a sexually transmitted infection (STI) than their white counterparts, in addition to being more likely to be HIV seropositive.¹⁷ In a survey of 200 predominantly black MSM aged 16 to 24, the belief that an undetectable viral load reduced infectiousness was associated with unprotected receptive anal intercourse with HIV-positive partners.¹⁸ In one study of black MSM prospectively followed in 6 US cities, HIV incidence was nearly twice as high in those ≤30 years of age (5.9% versus 3.0% overall for black MSM). Those ≤30 were more likely to report unprotected receptive anal intercourse with an HIV-positive or unknown status partner, more likely to have an STI, and less likely to have regular health care or access to health care.¹⁹ Black MSM and young MSM (age 20-29) in the United States were also found to be disproportionately affected by syphilis, whose incidence has been increasing in recent decades.²⁰

Notably, in our study, persons of black race had a longer time to entry into care after HIV diagnosis compared to those of nonblack race, which remained statistically significant in the adjusted regression model. This may reflect disparate access to health care, stigma, or other factors.²¹ It is known that black Americans suffer health disparities and inequalities disproportionately when compared with other racial and ethnic groups, and that there is a legacy of poorer HIV outcomes in black patients nationally and among those in the southeastern United States, where a larger proportion of the population is black than elsewhere.^11,22 Black persons are 50% more likely to die of heart disease or stroke and have double the infant mortality when compared to non-Hispanic whites and yet are significantly more likely to be uninsured.²³ Stigma is also a barrier to health care, with more than 60% of black Americans endorsing discrimination against people living with HIV/AIDS, in the United States.²¹ In 1 study, nearly half of black MSM reported mistrust of the medical establishment. Experiencing stigma was associated with longer gaps in care since their last HIV appointment.²⁴ In black men, stigma has been associated with lower rates of HIV testing, which may translate to poorer outcomes at each step along with the care continuum for black HIV-infected individuals.^17,25 Data from the Medical Monitoring Project showed that rates of sustained viral suppression for patients in clinical care were lower for blacks (25% lower rate than whites) and for younger persons (46% lower rate for those aged 18-29 compared to those ≥50 years of age), although rates of ART prescription and viral suppression are improving.²⁶

Limitations of the current study include a type of selection bias—those who were diagnosed but never presented to care were not included here. This limitation is inherent in most observational settings. Furthermore, though we regret that adequately detailed data on patient diagnosis, treatment, and linkage history were not available in more recent years for the current analysis, we believe that it may yet serve as a baseline assessment, informing future analyses to more accurately describe progress toward US National HIV/AIDS Strategy goals first articulated, along with the dawn of universal treatment guidelines, during the study period. Future directions for study could include surveillance data to detect linkages to care or deaths for those diagnosed but never linked into care at our site. Nationally, the <25-year-old group represents a crucial demographic target toward which improved HIV prevention, testing, and linkage efforts should be tailored. Our study also reinforced that barriers to care for black individuals with HIV remain despite advances in treatment. Further study is needed to better elucidate the factors causing disparities in these vulnerable populations so that targeted interventions may be developed accordingly.

Footnotes

Authors’ Note

Peter F. Rebeiro and Kelsey S. Ivey contributed equally to the conduct of this research.

Acknowledgments

The authors gratefully acknowledge all patients, caregivers, and data managers involved in the Vanderbilt Comprehensive Care Clinic cohort, without whom this work would not have been possible.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Institute of Allergy and Infectious Diseases (NIAID) through the Tennessee Center for AIDS Research (P30 AI110527); by the NIAID, the National Cancer Institute (NCI) and the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) through the Caribbean, Central, and South American network for HIV epidemiology (CCASAnet; U01 AI069923) and the North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD; U01 AI069918); and by the Health Resources and Services Administration (HRSA) through the Southeast AIDS Education and Training Center (U01 OHA28686).

References

Johnson

Hall

Lansky

Holtgrave

Mermin

. Trends in diagnoses of HIV infection in the United States, 2002-2011. JAMA. 2014;312(4):432–434.

Bradley

Hall

Wolitski

. Vital Signs: HIV diagnosis, care, and treatment among persons living with HIV--United States, 2011. MMWR Morb Mortal Wkly Rep. 2014;63(47):1113–1117.

Gardner

McLees

Steiner

Del Rio

Burman

. The spectrum of engagement in HIV care and its relevance to test-and-treat strategies for prevention of HIV infection. Clin Infect Dis. 2011;52(6):793–800.

Centers for Disease Control and Prevention. Monitoring selected national HIV prevention and care objectives by using HIV surveillance data—United States and 6 dependent areas—2011. HIV Surveillance Supplemental Report. 2013;18(5):1–47.

Millett

Crowley

Koh

. A way forward: the National HIV/AIDS strategy and reducing HIV incidence in the United States. J Acquir Immune Defic Syndr. 2010;55(Suppl 2):S144–S147.

McGowan

Weinstein

Samenow

. Drug use and receipt of highly active antiretroviral therapy among HIV-infected persons in two U.S. clinic cohorts. PLoS One. 2011;6(4):e18462.

Singh

Bradley

. Men living with diagnosed HIV who have sex with men: progress along the continuum of HIV care--United States, 2010. MMWR Morb Mortal Wkly Rep. 2014;63(38):829–833.

Valdiserri

Forsyth

Yakovchenko

Koh

. Measuring what matters: development of standard HIV core indicators across the U.S. Department of Health and Human Services. Public Health Rep. 2013;128(5):354–359.

Hanna

Selik

Tang

Gange

. Disparities among US states in HIV-related mortality in persons with HIV infection, 2001-2007. AIDS. 2012;26(1):95–103.

10.

Reif

Whetten

Wilson

Gong

. HIV/AIDS Epidemic in the South Reaches Crisis Proportions in Last Decade. Duke Center for Health Policy and Inequalities Research. 2011. Retrieved from https://www.hivlawandpolicy.org/sites/www.hivlawandpolicy.org/files/Epidemic%20in%20the%20South%20Reaches%20Crisis%20Proportions%20in%20Last%20Decade.pdf.

11.

Reif

Whetten

Wilson

. HIV/AIDS in the Southern USA: a disproportionate epidemic. AIDS Care. 2014;26(3):351–359.

12.

Zou

. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol. 2004;159(7):702–706.

13.

Harrell

. Regression Modeling Strategies: With Applications to Linear Models, LOGISTIC Regression, and Survival Analysis. New York, NY: Springer; 2001.

14.

Rebeiro

Althoff

Buchacz

. Retention among North American HIV-infected persons in clinical care, 2000-2008. J Acquir Immune Defic Syndr. 2013;62(3):356–362.

15.

Yehia

Rebeiro

Althoff

. Impact of age on retention in care and viral suppression. J Acquir Immune Defic Syndr. 2015;68(4):413–419.

16.

Althoff

Rebeiro

Brooks

. Disparities in the quality of HIV care when using US Department of Health and Human Services indicators. Clin Infect Dis. 2014;58(8):1185–1189.

17.

Millett

Peterson

Flores

. 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. 2012;380(9839):341–348.

18.

Bruce

Harper

Suleta

, The Adolescent Medicine Trials Network for HIV/AIDS Interventions. Sexual risk behavior and risk reduction beliefs among HIV-positive young men who have sex with men. AIDS Behav. 2013;17(4):1515–1523.

19.

Koblin

Mayer

Eshleman

. Correlates of HIV acquisition in a cohort of Black men who have sex with men in the United States: HIV prevention trials network (HPTN) 061. PLoS One. 2013;8(7):e70413.

20.

Abara

Hess

Neblett Fanfair

Bernstein

Paz-Bailey

. Syphilis trends among men who have sex with men in the United States and Western Europe: a systematic review of trend studies published between 2004 and 2015. PLoS One. 2016;11(7):e0159309.

21.

Kaiser Family Foundation. Survey of Americans on HIV/AIDS: Part Three - Experiences and Opinions by Race/Ethnicity and Age. The Henry J. Kaiser Family Foundation, 2004.

22.

Stone

. HIV/AIDS in Women and racial/ethnic minorities in the U.S. Curr Infect Dis Rep. 2012;14(1):53–60.

23.

Moonesinghe

Chang

Truman

, Centers for Disease Control and Prevention. Health insurance coverage - United States, 2008 and 2010. MMWR Suppl. 2013;62(3):61–64.

24.

Eaton

Driffin

Kegler

. The role of stigma and medical mistrust in the routine health care engagement of black men who have sex with men. Am J Public Health. 2015;105(2):e75–e82.

25.

Bond

Frye

Taylor

. Knowing is not enough: a qualitative report on HIV testing among heterosexual African-American men. AIDS Care. 2015;27(2):182–188.

26.

Bradley

Mattson

Beer

Huang

Shouse

, Medical Monitoring Project. Increased antiretroviral therapy prescription and HIV viral suppression among persons receiving clinical care for HIV infection. AIDS. 2016;30(13):2117–2124.