Abstract
Homeless youth experience increased risk of contracting HIV, making HIV testing imperative in this population. We analyzed factors associated with HIV testing among homeless youth in Atlanta, Georgia using data from the 2015 Atlanta Youth Count and Needs Assessment. The analysis included 693 homeless youth aged 14–25 years, of whom 88.4% reported ever being tested for HIV, and 74.6% reported being tested within the previous year. Prevalence of ever testing for HIV was significantly higher among youth who reported risk factors for HIV (sexually active, transactional sex, or ever having an STI). Higher prevalence of testing within the last year was significantly associated with experiencing physical abuse or transactional sex. However, reporting ≥4 sexual partners or not using condoms were not associated with higher testing. Although testing prevalence among homeless youth was high, homeless youth engaging in certain high risk behaviors could benefit from further promotion of HIV testing.
Keywords: homeless youth, Atlanta, Georgia, human immunodeficiency virus, adolescent, risk factors
In metropolitan Atlanta, Georgia, an estimated 3,374 homeless and runaway youth (aged < 25 years) are living on the streets, in shelters, or in other unstable housing situations (Wright et al., 2016). Unstable housing and poverty can negatively affect the health of youth (Farrow, Deisher, Brown, Kulig, & Kipke, 1992), particularly homeless youth (Beharry, 2012; Edidin, Ganim, Hunter, & Karnik, 2012). In particular, homeless youth are at an increased risk for sexually transmitted infections (STIs) and HIV specifically (Caccamo, Kachur, & Williams, 2017; Carmona, Slesnick, Guo, & Letcher, 2014; Ferguson, Bender, Thompson, Xie, & Pollio, 2011; Marshall, Kerr, Shoveller, Montaner, & Wood, 2009; Martino et al., 2011; Mastro, Cunningham, Medrano, & van Dam, 2012; Tucker, Hu, et al., 2012). For homeless youth, increased exposure to drugs, early sexual debut, being a sexual minority (lesbian, gay, bisexual, transgender, or other), having multiple sex partners, engaging in high-risk survival behaviors (e.g., exchanging sex for food, drugs, shelter, or money) to secure resources, and being at an increased risk for sexual exploitation, can all contribute to an increased risk for HIV (Cochran et al., 2002; Ferguson et al., 2011; Martino et al,. 2011; Mastro et al., 2012; Tucker, Hu, et al., 2012; Tyler, 2013).
During 2016, youth aged 13–24 years accounted for 21% of all new HIV diagnoses (Centers for Disease Control and Prevention [CDC], 2018a), but true prevalence of HIV among homeless youth is difficult to determine because it is difficult to conduct health research in this population. Some studies have estimated that HIV prevalence among homeless youth is 2–12 times higher than their stably-housed peers (Hsu et al., 2018; Pfeifer & Oliver, 1997; Roteram-Borus et al., 2003; Santa Maria et al., 2018; Stricof, Kennedy, Nattell, Weisfuse, & Novick, 1991; Young & Rice, 2011). Atlanta’s homeless youth, in particular, are at an increased risk for infection, given the concentrated rates of HIV in the area. Among metropolitan statistical areas in the United States and Puerto Rico during 2016, Atlanta ranked fourth in rates of new diagnoses of HIV infection (CDC, 2017). Therefore, HIV testing is important for this population to ensure that they are aware of their status and linked to care and treatment (Lally et al., 2018). Additionally, the CDC recommends that everyone between the ages of 13 and 64 should get tested for HIV at least once, but people at higher risk (men who have sex with men, people who exchange sex for drugs or money, people who inject drugs and share needles and works) should get tested more often (CDC, 2019); some homeless youth should be tested annually based on their level of risk.
According to the CDC’s Youth Risk Behavior Survey (YRBS), prevalence of ever having been tested for HIV was 9.3% during 2017 among a nationally representative sample of students enrolled in high school (Kann et al., 2018). The prevalence of HIV testing among youth who are unstably housed can vary by location, but often is higher than the HIV testing prevalence among their high school peers. In three Midwestern cities in the United States, 67% of homeless youth reported testing for HIV (Tyler & Melander, 2010), and 85% of sexually active homeless youth in a Los Angeles study had ever been tested for HIV (Ober, Martino, Ewing, & Tucker, 2012). Research demonstrates that if offered an HIV test, homeless youth are likely to take the test (Ober et al., 2012). Homeless youth with a high likelihood of HIV testing include females, older youth, Black or mixed race youth, gay or bisexual males, and youth who have a history of pregnancy or of fathering a child (Solorio, Milburn, Weiss, & Batterham, 2006). Nevertheless, certain homeless youth have reported confidentiality concerns and a lack of trust of adults as barriers to HIV testing (Tyler, Akinyemi, & Kort-Butler, 2012).
Given the high rates of HIV diagnosis in the South, overall, and among youth aged 13–24 (CDC, 2016, 2018b), it is important to focus efforts among youth in the South, Atlanta in particular (Camacho-Gonzalez et al., 2016). Therefore, understanding factors associated with HIV testing among homeless youth in Atlanta is critical to devise strategies to increase testing and prevent new infections in this vulnerable population. We conducted a study to examine demographic and other relevant factors (sexual- and violence-related) associated with HIV testing among homeless youth in metropolitan Atlanta.
METHODS
POPULATION SAMPLE AND SETTING
We used data from the Atlanta Youth Count and Needs Assessment (AYCNA) conducted in 2015. The 2015 AYCNA was designed to quantify the number, and assess the state, of homeless, precariously housed, and runaway youth in Atlanta, Georgia. AYCNA surveyed homeless and runaway youth aged 14–25 years, who reported not having a permanent residence of their own, and who were living independently without consistent parental and family support (Wright et al., 2016). The study design used capture–recapture field-based sampling methods, and participants were recruited by outreach and service workers who conducted sweeps of shelters and community locations where homeless youth spend time and reside (Wright et al., 2016). Participants were invited to complete a 15-minute, face-to-face interview and to answer questions about their personal and social background, health status, and contact with health and social service systems. In addition, youth were asked about current and past experiences with homelessness and factors that led to their being homeless. All data were collected anonymously, and no information was collected that could be used to identify or trace participants. Verbal consent was obtained from participants by trained researchers. The study was approved by the Institutional Review Board at Georgia State University (Wright et al., 2016).
MEASURES
Demographic Variables.
In the 2015 AYCNA, age was measured through a free-response question with responses ranging from 15 to 25 years. Age was coded into two groups, 15–19 and 20–25 years. Sex was measured by the question, “What sex were you assigned at birth?” Response options were Male/Man, Female/Woman, and Something Else (Specify). Only responses indicating Male or Female were included in this analysis; the one response received of Something Else was excluded from analyses that included the sex variable. Race and ethnicity were measured through two questions: (1) “Do you consider yourself to be Hispanic or Latino?” with response options Yes or No; and (2) “What race do you consider yourself? (Please check at least one and all that apply)” with response options White, Black or African American, Asian, Native American/Alaska Native, Pacific Islander, Multiracial, or Other (Specify). For this analysis, the variable race was categorized into Non-Hispanic Black (Black), Non-Hispanic White (White), Hispanic, or Non-Hispanic Other (Other).
Sexual orientation was measured by the question, “Which of the following labels best describes your sexual orientation?” The response options were Straight, Heterosexual, Gay or Lesbian, Bisexual, Something Else, or Still Undecided/Questioning. Responses were coded as heterosexual, sexual minority (lesbian, gay, bisexual), and other (something else or still undecided/questioning).
HIV Testing.
HIV testing behavior was measured by two questions: (1) “Have you ever been tested for HIV/AIDS?” with response options Yes, No, or Don’t Know; and (2) “Approximately, when did you take your last test?” with an open response option for the test date. For this analysis, the variable for ever having been tested for HIV was dichotomized so that Yes was coded as tested, and No and Don’t Know were coded as not tested. The variable for HIV testing within the previous year was calculated by subtracting the interview date from the indicated test date, and dichotomized into a variable indicating whether they were tested or not tested in the previous year. Additionally, participants were asked, “Have you ever experienced the following health problem?” in which a list of problems included A Sexually Transmitted Infection (STI) Other Than HIV, with response options of Yes or No.
Sexual Risk Factors.
Questions related to sexual risk factors included having had sex during the previous year, condom use during the previous year, and number of sex partners during the previous year. To examine having had sex during the previous year and condom use, participants were asked, “Have you had sex with anyone in the past year?” and “Have you had vaginal or anal sex without a condom in the past year?” Response options were Yes or No for both questions. To assess the number of sex partners during the previous year, participants were asked, “In the past year, with how many different partners have you had any sex with?” and the open response answers were coded into fewer than four partners or four or more partners.
Experiences of Violence.
The violence factors included three questions related to sexual abuse or assault. The overarching question was phrased, “I would like to know if the situation happened to you when you were 17 years old or younger, (or if over 18) since you turned 18, and during the time you have been homeless this time.” The response options for each period were Yes or No, and the situations were as follows: (1) “been in a sexual relationship in which you were physically abused?”; (2) “been in a sexual relationship in which you were sexually abused?”; and (3) “been sexually assaulted or raped?” If a participant indicated Yes to any period (aged < 17 years, ≥ 18 years, or while homeless), the response was coded as yes for ever having experienced the corresponding situation.
Transactional Sex.
Four questions were related to exchanging sex for money or favors. Again, the overarching question was phrased, “I would like to know if the situation happened to you when you were 17 years old or younger, (or if over 18) since you turned 18, and during the time you have been homeless this time.” The response options for each period were Yes or No, and the transactional sex scenarios were as follows: (1) “had sex with someone to get money or cash?”; (2) “had sex for drugs, food, a ride, or a place to stay?”; (3) “had a friend, mentor, or family member who was involved with your having sex for money?”; and (4) “had a ‘street daddy,’ ‘boyfriend,’ or ‘pimp’ who was involved in your having sex for money?” If a participant indicated Yes to any period (aged < 17 years, ≥ 18 years, or while homeless), the response was coded as yes for ever having experienced the corresponding situation.
ANALYSIS
Cross-tabulations were conducted for having ever been tested for HIV and for HIV testing within the past year stratified by each of the demographic and other factors. Pearson’s chi-square (χ2) tests were used to determine statistically significant differences in HIV testing by each variable of interest at the p < .05 level. All analyses were conducted by using IBM SPSS Statistics 21 (SPSS Incorporated, Chicago, IL).
RESULTS
OVERALL PREVALENCE OF STUDY VARIABLES
A total of 693 youth participated in the 2015 AYCNA. The majority of AYCNA participants were male (66.5%), aged 20–25 years (70.9%), heterosexual (73.1%), and Black (70.0%). Participants had varying degrees of sexual risk as indicated in Table 1, including 85.0% having been sexually active within the previous year, 63.0% having had vaginal or anal sex without a condom during the previous year, and 38.4% having had four or more sex partners during their lifetime. Additionally, at the time of the survey, 8.5% had ever had an STI (Table 1). The prevalence of ever testing for HIV was 88.4% (n = 597 of 675), while the prevalence of HIV testing within the previous year was 74.6% (n = 516 of 692) (Tables 2 and 3).
TABLE 1.
Sample Characteristics for 2015 Atlanta Youth Count and Needs Assessment (AYCNA)
| Variable | n (%) |
|---|---|
| Age group (years) (n = 693) | |
| 15–19 | 202 (29.1) |
| 20–25 | 491 (70.9) |
| Race/ethnicity (n = 691) | |
| Non-Hispanic Black | 484 (70.0) |
| Hispanic | 56 (8.1) |
| Non-Hispanic White | 37 (5.4) |
| Non-Hispanic other | 114 (16.5) |
| Sex (n = 683) | |
| Female | 229 (33.5) |
| Male | 454 (66.5) |
| Sexual orientation (n = 689) | |
| Heterosexual | 504 (73.1) |
| Sexual minoritya | 170 (24.7) |
| Othera | 15 (2.2) |
| Sexual risk factorsb | |
| Sexually active within the previous year (n = 693) | 589 (85.0) |
| Sex without condom within the previous year (n = 579) | 365 (63.0) |
| ≥ 4 sex partners (n = 571) | 219 (38.4) |
| Experiences of violenceb | |
| Physical abuse in a relationship (n = 663) | 141 (21.3) |
| Sexual abuse in a sexual relationship (n = 663) | 83 (12.5) |
| Sexual assault or rape (n = 663) | 171 (25.8) |
| Transactional sexb | |
| Sex exchanged for drugs, food, rides, or shelter (n = 663) | 98 (14.8) |
| Sex exchanged for money (n = 663) | 156 (23.5) |
| Sex exchanged for money—family, friend, or mentor involved (n = 662) | 64 (9.7) |
| Sex exchanged for money—“street daddy,” “boyfriend,” or “pimp” involved (n = 663) | 45 (6.8) |
| Sexually transmitted infection (STI)b | |
| Ever had an STI (non-HIV) (n = 669) | 57 (8.5) |
Note. All numbers might not sum to 693 because of missing data, and percentages might not add to 100 because of rounding.
Sexual minority includes youth who responded that they were lesbian, gay, or bisexual. Other includes youth who responded “something else” or “still undecided/questioning.”
Frequencies are reported for those who said Yes to each risk factor, experience, or infection; data for No responses are not reported in Table 1.
TABLE 2.
Prevalence of Ever Having Been Tested for HIV Infection, by Demographic and Other Factors, 2015 Atlanta Youth Count and Needs Assessment (AYCNA)
| Variable | n (%) | 95% CI | p valuea |
|---|---|---|---|
| Ever having been tested for HIV infection (n = 675) | 597 (88.4) | 86.0, 90.8 | — |
| Age group (years) | < .001 | ||
| 15–19 (n = 199) | 161 (80.9) | 75.4, 86.4 | |
| 20–25 (n = 476) | 436 (91.6) | 89.1, 94.1 | |
| Race/ethnicity | .109 | ||
| Non-Hispanic Black (n = 471) | 415 (88.1) | 85.2, 91.0 | |
| Hispanic (n = 56) | 45 (80.4) | 70.0, 90.8 | |
| Non-Hispanic White (n = 36) | 33 (91.7) | 82.7, 100.7 | |
| Non-Hispanic Other (n = 111) | 103 (92.8) | 88.0, 97.6 | |
| Sex | .752 | ||
| Female (n = 221) | 197 (89.1) | 85.0, 93.2 | |
| Male (n = 445) | 393 (88.3) | 85.3, 91.3 | |
| Sexual orientation | .081 | ||
| Heterosexual (n = 491) | 428 (87.2) | 84.2, 90.2 | |
| Sexual minorityb (n = 168) | 156 (92.9) | 89.0, 96.8 | |
| Otherc (n = 15) | 12 (80.0) | 59.8, 100.2 | |
| Sexual risk factors | |||
| Sexually active within the previous year (n = 583) | 523 (89.7) | 87.2, 92.2 | .010 |
| Not sexually active within the previous year (n = 92) | 74 (80.4) | 72.3, 88.5 | |
| Sex without a condom within the previous year (n = 360) | 328 (91.1) | 88.2, 94.0 | .271 |
| Sex with a condom within the previous year (n = 213) | 188 (88.3) | 84.0, 92.6 | |
| ≥ 4 sex partners (n = 218) | 198 (90.8) | 87.0, 94.6 | .435 |
| < 4 sex partners (n = 347) | 308 (88.8) | 85.5, 92.1 | |
| Experiences of violence | |||
| Physical abuse in a relationship (n = 139) | 131 (94.2) | 90.3, 98.1 | .018 |
| No physical abuse in a relationship (n = 518) | 451 (87.1) | 84.2, 90.0 | |
| Sexual abuse in a sexual relationship (n = 82) | 80 (97.6) | 94.3, 100.9 | .006 |
| No sexual abuse in a sexual relationship (n = 575) | 502 (87.3) | 84.6, 90.0 | |
| Sexual assault or rape (n = 169) | 159 (94.1) | 90.5, 97.7 | .009 |
| No sexual assault or rape (n = 488) | 423 (86.7) | 83.7, 89.7 | |
| Transactional sex | |||
| Sex exchanged for drugs, food, rides, or shelter (n = 97) | 93 (95.9) | 92.0, 99.8 | .014 |
| No sex exchanged for drugs, food, rides, or shelter (n = 560) | 489 (87.3) | 84.5, 90.1 | |
| Sex exchanged for money (n = 156) | 147 (94.2) | 90.5, 97.9 | .011 |
| No sex exchanged for money (n = 501) | 435 (86.8) | 83.8, 89.8 | |
| Sex exchanged for money—family, friend, or mentor involved (n = 64) | 60 (93.8) | 87.9, 99.7 | .170 |
| No sex exchanged for money—family, friend, or mentor involved (n = 592) | 521 (88.0) | 85.4, 90.6 | |
| Sex exchanged for money—“street daddy,” “boyfriend,” or “pimp” involved (n = 45) | 40 (88.9) | 79.7, 98.1 | .947 |
| No sex exchanged for money—“street daddy,” “boyfriend,” or “pimp” involved (n = 612) | 542 (88.6) | 86.1, 91.1 | |
| Sexually transmitted infections (STI) | |||
| Ever had an STI (non-HIV) (n = 57) | 55 (96.5) | 91.7, 101.3 | .045 |
| Never had an STI (n = 611) | 535 (87.6) | 85.0, 90.2 |
Note. CI: confidence interval.
p values were derived by using Pearson’s chi-square test. 2 × 2 comparisons were conducted for all analyses with the exception of race/ethnicity (4 × 2) and sexual orientation (3 × 2).
Sexual minorityincludes youth who responded that they were lesbian, gay, or bisexual. Other includes youth who responded “something else” or “still undecided/questioning.”
TABLE 3.
Prevalence of Having Been Tested for HIV During the Previous Year, by Demographic and Other Factors, 2015 Atlanta Youth Count and Needs Assessment (AYCNA)
| Variable | n (%) | 95% CI | p valuea |
|---|---|---|---|
| Tested for HIV during the previous year (n = 692) | 516 (74.6) | 71.4, 77.8 | — |
| Age group (years) | .829 | ||
| 15–19 (n = 201) | 151 (75.1) | 69.1, 81.1 | |
| 20–25 (n = 491) | 365 (74.3) | 70.4, 78.2 | |
| Race/ethnicity | .088 | ||
| Non-Hispanic Black (n = 483) | 355 (73.5) | 69.6, 77.4 | |
| Hispanic (n = 56) | 38 (67.9) | 55.7, 80.1 | |
| Non-Hispanic White (n = 37) | 33 (89.2) | 79.2, 99.2 | |
| Non-Hispanic Other (n = 114) | 89 (78.1) | 70.5, 85.7 | |
| Sex | .290 | ||
| Female (n = 229) | 177 (77.3) | 71.9, 82.7 | |
| Male (n = 454) | 334 (73.6) | 69.5, 77.7 | |
| Sexual orientation | .320 | ||
| Heterosexual (n = 503) | 371 (73.8) | 70.0, 77.6 | |
| Sexual minorityb (n = 170) | 134 (78.8) | 72.7, 84.9 | |
| Otherc (n = 15) | 10 (66.7) | 42.8, 90.6 | |
| Sexual risk factors | |||
| Sexually active within the previous year (n = 588) | 452 (76.9) | 73.5, 80.3 | .001 |
| Not sexually active within the previous year (n = 104) | 64 (61.5) | 52.1, 70.9 | |
| Sex without a condom within the previous year (n = 364) | 285 (78.3) | 74.1, 82.5 | .472 |
| Sex with a condom within the previous year (n = 214) | 162 (75.7) | 70.0, 81.4 | |
| ≥ 4 sex partners (n = 219) | 176 (80.4) | 75.1, 85.7 | .115 |
| < 4 sex partners (n = 351) | 262 (74.6) | 70.0, 79.2 | |
| Experiences of violence | |||
| Physical abuse in a relationship (n = 140) | 116 (82.9) | 76.7, 89.1 | .036 |
| No physical abuse in a relationship (n = 522) | 388 (74.3) | 70.6, 78.0 | |
| Sexual abuse in a sexual relationship (n = 82) | 68 (82.9) | 74.8, 91.0 | .123 |
| No sexual abuse in a sexual relationship (n = 580) | 436 (75.2) | 71.7, 78.7 | |
| Sexual assault or rape (n = 170) | 138 (81.2) | 75.3, 87.1 | .074 |
| No sexual assault or rape (n = 492) | 366 (74.4) | 70.5, 78.3 | |
| Transactional sex | |||
| Sex exchanged for drugs, food, rides, or shelter (n = 97) | 79 (81.4) | 73.7, 89.1 | .184 |
| No sex exchanged for drugs, food, rides, or shelter (n = 565) | 425 (75.2) | 71.6, 78.8 | |
| Sex exchanged for money (n = 155) | 128 (82.6) | 76.6, 88.6 | .031 |
| No sex exchanged for money (n = 507) | 376 (74.2) | 70.4, 78.0 | |
| Sex exchanged for money—family, friend, or mentor involved (n = 63) | 54 (85.7) | 77.1, 94.3 | .060 |
| No sex exchanged for money—family, friend, or mentor involved (n = 598) | 449 (75.1) | 71.6, 78.6 | |
| Sex exchanged for money—“street daddy,” “boyfriend,” or “pimp” involved (n = 44) | 35 (79.5) | 67.6, 91.4 | .583 |
| No sex exchanged for money—“street daddy,” “boyfriend,” or “pimp” involved (n = 618) | 469 (75.9) | 72.5, 79.3 | |
| Sexually transmitted infections (STI) | |||
| Ever had an STI (non-HIV) (n = 57) | 50 (87.7) | 79.2, 96.2 | .039 |
| Never had an STI (n = 611) | 462 (75.6) | 72.2, 79.0 |
Note. CI: confidence interval.
p values were derived by using Pearson’s chi-square test. 2 × 2 comparisons were conducted for all analyses with the exception of race/ethnicity (4 × 2) and sexual orientation (3 × 2).
Sexual minorityincludes youth who responded that they were lesbian, gay, or bisexual. Other includes youth who responded “something else” or “still undecided/questioning.”
DIFFERENCES IN HIV TESTING BY DEMOGRAPHIC CHARACTERISTICS
Prevalence of ever testing for HIV was significantly higher among persons who were aged 20–25 years (91.6%, n = 436 of 476), compared with those aged 15–19 years (80.9%, n = 161 of 199) (χ2 = 15.70; p < .001). HIV testing within the previous year did not differ significantly by age (χ2 = 0.05; p = .829). Neither ever testing for HIV nor HIV testing within the previous year differed by race/ethnicity, sex, or sexual orientation (Tables 2 and 3).
DIFFERENCES IN HIV TESTING BY SEXUAL RISK, VIOLENCE, AND TRANSACTIONAL SEX
Prevalence of ever testing for HIV was higher among homeless youth who had been sexually active during the previous year (89.7%, n = 523 of 583), compared with those who had not been sexually active (80.4%, n = 74 of 92) (χ2 = 6.69; p < .05). Ever testing for HIV was higher among homeless youth who had experienced physical abuse in a sexual relationship (94.2%, n = 131 of 139), compared with youth who had not (87.1%, n = 451 of 518) (χ2 = 5.59; p < .05); youth who had experienced sexual abuse in a sexual relationship (97.6%, n = 80 of 82), compared with youth who had not (87.3%, n = 502 of 575) (χ2 = 7.47; p < .01); and youth who had experienced sexual assault or rape (94.1%, n = 159 of 169), compared with youth who had not (86.7%, n = 423 of 488) (χ2 = 6.80; p < .01). Testing prevalence was higher among those who indicated having exchanged sex for drugs, food, rides, or shelter (95.9%, n = 93 of 97), compared with those who had not (87.3%, n = 489 of 560) (χ2 = 5.98; p < .05), and those who had exchanged sex for money (94.2%, n = 147 of 156), compared with those who had not (86.8%, n = 435 of 501) (χ2 = 6.45; p < .05). Ever testing for HIV did not vary on the basis of condom use; number of sex partners; whether they had a friend, mentor, or family member involved with their having exchanged sex for money; or whether participants had a street daddy, boyfriend, or pimp involved in their having exchanged sex for money (Table 2).
Prevalence of HIV testing within the previous year was higher among youth who had been sexually active during the previous year (76.9%, n = 452 of 588), compared with those who had not been sexually active (61.5%, n = 64 of 104) (χ2 = 10.95; p < .01). Testing within the previous year was significantly higher for youth who had experienced physical abuse in a sexual relationship (82.9%, n = 116 of 140), compared with youth who had not (74.3%, n = 388 of 522) (χ2 = 4.42; p < .05). Prevalence of HIV testing within the previous year was higher for those who had exchanged sex for money (82.6%, n = 128 of 155), compared with those who had not (74.2%, n = 376 of 507) (χ2 = 4.63; p < .05). HIV testing within the previous year did not differ on the basis of condom use; number of sex partners; experience of sexual abuse in a sexual relationship; experience of sexual assault or rape; whether they exchanged sex for drugs, food, rides, or shelter; whether participants had a friend, mentor, or family member involved with their having exchanged sex for money; or whether they had a street daddy, boyfriend, or pimp involved in their having exchanged sex for money (Table 3).
DIFFERENCES IN HIV TESTING BY STI HISTORY
Participants who had ever had an STI (non-HIV) had a higher prevalence of ever testing for HIV (96.5%, n = 55 of 57) than those who had not (87.6%, n = 535 of 611) (χ2 = 4.03; p < .05). Similarly, those who had ever had an STI (non-HIV) had a higher prevalence of testing for HIV within the previous year (87.7%, n = 50 of 57) than those who had not (75.6%, n = 462 of 611) (χ2 = 4.27; p < .05) (Tables 2 and 3).
DISCUSSION
In this study of HIV testing history, sexual risk, and other factors among homeless youth aged 14–25 years in Atlanta, GA, we found that the prevalence of ever testing for HIV among the homeless youth was 88.4%, which is high when compared with 9.3% among a nationally representative sample of U.S. high school students in 2017 (Kann et al., 2018). Our findings are consistent with the previous research showing high HIV testing prevalence among urban homeless youth in Los Angeles (85%) and Midwestern cities (67%). This increased testing prevalence among AYCNA respondents might be because either homeless youth engage in high-risk sexual activity or have increased access to testing through service providers focusing on homeless youth as indicated from previous studies of homeless youth (Cochran et al., 2002; Ferguson et al., 2011; Martino et al., 2011; Mastro et al., 2012; Solorio et al., 2006; Tucker, Ryan, & Golinelli, 2012; Tyler, 2013; Tyler et al., 2012). Also, from 2012 to 2015, there was an intervention in Atlanta to improve diagnosis, linkage, and retention in care of youth ages 18–24 at high-risk for contracting HIV. The Metropolitan Atlanta Community Adolescent Rapid Testing Initiative (MACARTI) included nontraditional venue testing (night clubs, street testing, Pride events, etc.), motivational interviewing, and case management (Camacho-Gonzalez et al., 2017). Some of the nontraditional testing venues from this initiative may have been in place at the time of the AYCNA interview and participants might have beneiftied from these testing venues and taken HIV tests.
Demographically, statistically significant differences in HIV testing prevalence existed only by age, but not for sex, race/ethnicity, or sexual orientation. Older youth had a significantly higher prevalence of HIV testing (ever testing for HIV and HIV testing within the previous year) for those aged 20–25 years, compared with those aged 15–19 years. These findings are consistent with a study in Los Angeles that reported that older youth had a higher prevalence of testing for HIV, compared with younger youth (Solorio et al., 2006). It is intuitive that HIV testing prevalence would increase with age among youth, because older youth have had more sexual experiences, and have had more opportunity for testing. Likewise, a study of homeless youth in the Midwest reported that older youth are more likely to get tested because of greater knowledge of health service locations, and a greater knowledge of HIV and STIs, and their association with sexual risk behaviors (Tyler et al., 2012).
Unlike our study, previous studies have determined substantial differences in homeless youth testing for HIV on the basis of race/ethnicity and sexual orientation. A study of homeless youth in Los Angeles and San Diego reported that Black and White youth were more likely to be tested than Latino youth, and gay/bisexual youth were more likely to be tested than heterosexual youth (De Rosa et al., 2001). Another study reported that gay/bisexual homeless youth report higher levels of behaviors that put them at risk for HIV infection, which is why they were tested at higher rates (Solorio et al., 2006). Differences by these demographics might not have been observed for the AYCNA data because of targeted HIV testing efforts in Atlanta during that time that included a large portion of AYCNA participant demographic. During 2015, the CDC-funded High-Impact Prevention Program at the Fulton County Department of Health and Wellness had its best year with a 40.0% increase in HIV testing at community-based organizations and health departments, compared with the previous year (Fulton County, GA, 2011). The HIV testing component of this program enhanced testing for HIV in Atlanta and focused on youth, homeless populations, LGBTQ youth, Blacks, and other difficult to reach populations.
Our study had certain limitations. AYCNA was conducted in Atlanta and participants might not be representative of the national homeless youth population. The survey instrument relied on self-reported data, which introduces possible reporter bias. The data collection protocol was such that participants were unable to take the survey independently. A researcher read the questions, and the youth verbally indicated their responses, which might be limiting because youth might not have felt comfortable revealing certain information (e.g., sexual orientation, sexual abuse, or age). Additionally, youth may have been more likely to say they had been tested due to social desirability. Lastly, during consent procedures participants were informed that if they reported for themselves or another minor (aged <18 years) that they had been abused, neglected, or exploited, the researchers were required by Georgia State law to make a report to child protective services (Forge, Hartinger-Saunders, Wright, & Ruel, 2018; State of Georgia, Office of the Child Advocate, 2016). Therefore, despite assuring the participants that their identity would be protected and kept confidential, participants who were aged <18 years might have inflated their age.
CONCLUSIONS
Results from this study provide a detailed examination of the HIV testing prevalence of Atlanta’s homeless youth, which has not previously been reported. Additionally, for improved accuracy in describing the current population of homeless youth in metro Atlanta, the study utilized sophisticated systematic capture-recapture field sampling methods to locate members of this hard-to-reach population (Wright et al., 2016). The high rates of HIV testing among homeless youth in metro Atlanta, relative to the national YRBS rate, are most likely indicative of public health efforts to decrease infection rates among vulnerable populations, and is a public health success for HIV prevention in this southern hotspot. The AYCNA HIV testing prevalence may also reflect the fact that urban areas tend to be focal points for HIV prevention interventions, STD testing, and social marketing campaigns. Still, a number of respondents (25.3%) had not been tested for HIV in the past 12 months. The CDC recommends annual testing for those at higher risk, for example men who have sex with men, > 1 sex partner since last HIV test, sex in exchange for drugs or money, and those who have been diagnosed with an STI (CDC, 2019). The recently announced Ending the HIV Epidemic: A Plan for America initiative proposes to end the HIV epidemic in the United States within 10 years (Department of Health and Human Services, 2019). In order for this initiative to be successful, we must ensure that all (especially youth) who are at any risk of contracting HIV, are tested, linked to care, and treated effectively. Opportunities exist for further research on effective strategies to increase rates of HIV testing and examine the linkages to care and outcomes for Atlanta’s homeless youth living with HIV infection.
Acknowledgments
We thank the many youth who contributed their time to participating in this study. We also express our appreciation to the many university students and community volunteers who served as field research assistants and survey administrators or helped in other ways to implement the parent study. Finally, from the Centers for Disease Control and Prevention we thank Deron Burton, MD, JD, MPH for providing feedback on draft versions of this manuscript, and Cheryll Kay Smith for technical writing and copyediting support.
Contributor Information
Ranell L. Myles, Centers for Disease Control and Prevention, Atlanta, Georgia.
Janae Best, University of Michigan, Ann Arbor, Michigan..
Greg Bautista, Centers for Disease Control and Prevention, Atlanta, Georgia..
Eric R. Wright, Georgia State University, Atlanta, Georgia.
Ana LaBoy, Georgia State University, Atlanta, Georgia..
Zewditu Demissie, Centers for Disease Control and Prevention, Atlanta, Georgia.; U.S. Public Health Service Commissioned Corps, Rockville, Maryland.
Hazel D. Dean, Centers for Disease Control and Prevention, Atlanta, Georgia.
REFERENCES
- Beharry MS (2012). Health issues in the homeless youth population. Pediatric Annals, 41, 154–156. [DOI] [PubMed] [Google Scholar]
- Caccamo A, Kachur R, & Williams SP (2017). Narrative review: Sexually transmitted diseases and homeless youth––What do we know about sexually transmitted disease prevalence and risk? Sexually Transmitted Diseases, 44, 466–476. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Camacho-Gonzalez AF, Gillespie SE, Thomas-Seaton L, Frieson K, Hussen SA, Murray A, . . . Chakraborty R (2017). The metropolitan Atlanta commumity adolescent rapid testing initiative study: Closing the gaps in HIV care among youth in Atlanta, Georgia, USA. AIDS, 31, S267–S275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Camacho-Gonzalez AF, Wallins A, Toledo L, Murray A, Gaul Z, Sutton MY, . . . Chakraborty R (2016). Risk factors for HIV transmission and barriers to HIV disclosure: Metropolitan Atlanta youth perspectives. AIDS Patient Care and STDs, 30, 18–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Carmona J, Slesnick N, Guo X, & Letcher A (2014). Reducing high risk behaviors among street living youth: Outcomes of an integrated prevention intervention. Children and Youth Services Review, 43, 118–123. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Centers for Disease Control and Prevention. (2016). HIV surveillance supplemental report: Diagnoses of HIV infection among adolescents and young adults in the United States and 6 dependent areas 2010–2014. Retrieved December 4, 2018, from https://www.cdc.gov/hiv/pdf/library/reports/surveillance/cdc-hiv-surveillance-supplemental-report-vol-21-3.pdf
- Centers for Disease Control and Prevention. (2017). HIV Surveillance Report, 2016. Retrieved December 4, 2018, from https://www.cdc.gov/hiv/pdf/library/reports/surveillance/cdc-hiv-surveillance-report-2016-vol-28.pdf
- Centers for Disease Control and Prevention. (2018a). HIV among youth. Retrieved December 4, 2018, from https://www.cdc.gov/hiv/group/age/youth/index.html
- Centers for Disease Control and Prevention. (2018b). Diagnoses of HIV infection among adults and adolescents in metropolitan statistical areas—United States and Puerto Rico, 2016. Retrieved December 4, 2018, from https://www.cdc.gov/hiv/pdf/library/reports/surveillance/cdc-hiv-surveillance-supplemental-report-vol-23-2.pdf
- Centers for Disease Control and Prevention. (2019). HIV/AIDS, HIV basics, testing. Retrieved February 8, 2019, from https://www.cdc.gov/hiv/basics/testing.html
- Cochran BN, Stewart AJ, Ginzler JA, & Cauce AM (2002). Challenges faced by homeless sexual minorities: Comparison of gay, lesbian, bisexual, and transgender homeless adolescents with their heterosexual counterparts. American Journal of Public Health, 92, 773–777. [DOI] [PMC free article] [PubMed] [Google Scholar]
- De Rosa CJ, Montgomery SB, Hyde J, Iverson E, & Kipke MD (2001). HIV risk behavior and HIV testing: A comparison of rates and associated factors among homeless and runaway adolescents in two cities. AIDS Education and Prevention, 13, 131–148. [DOI] [PubMed] [Google Scholar]
- Department of Health and Human Services. (2019). Ending the HIV epidemic: A plan for America. Retrieved July 15, 2019, from https://www.hhs.gov/blog/2019/02/05/ending-the-hiv-epidemic-a-plan-for-america.html
- Edidin J, Ganim Z, Hunter SJ, & Karnik NS (2012). The mental and physical health of homeless youth. Child Psychiatry & Human Development, 43, 354–375. [DOI] [PubMed] [Google Scholar]
- Farrow J, Deisher R, Brown R, Kulig JW, & Kipke MD (1992). Health and health needs of homeless and runaway youth. Journal of Adolescent Health, 13, 717–726. [DOI] [PubMed] [Google Scholar]
- Ferguson KM[dummy_check_author_name], Bender K, Thompson S, Xie B, & Pollio D (2011). Correlates of street-survival behaviors in homeless young adults in four U.S. cities. American Journal of Ortho psychiatry, 81, 401–409. [DOI] [PubMed] [Google Scholar]
- Forge N, Hartinger-Saunders R, Wright E, & Ruel E (2018). Out of the system and onto the streets: LGBTQ-identified youth experiencing homelessness with past child welfare system involvement. Child Welfare, 96, 47–74. [Google Scholar]
- Fulton County, Georgia. (2011). HIV testing. Retrieved December 4, 2018, from http://www.fultoncountyga.gov/component/content/article/696-hiv-task-force/8061-hiv-testing
- Hsu H, Fulginiti A, Rice E, Rhoades H, Winetrobe H, & Danforth L (2018). But everyone is doing it (sort of)! Perceived sexual risks in the social environment and the impact on homeless youth engagement in concurrent sexual relationships. AIDS and Behavior, 22, 3508–3518. [DOI] [PubMed] [Google Scholar]
- Kann L, McManus T, Harris WA, Shanklin SL, Flint KH, Queen B, . . . Ethier KA (2018). Youth risk behavior surveillance—United States, 2017. MMWR Surveillance Summaries, 67, 1–114. Retrieved December 4, 2018, from https://www.cdc.gov/healthyyouth/data/yrbs/pdf/2017/ss6708.pdf [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lally MA, van den Berg JJ, Westfall AO, Rudy BJ, Hosek SG, Fortenberry JD, . . . Adolescent Medicine Trials Network for HIV/AIDS Interventions. (2018). HIV continuum of care for youth in the United States. Journal of Acquired Immune Deficency Syndromes, 77, 110–117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Marshall BD, Kerr T, Shoveller JA, Montaner JS, & Wood E (2009). Structural factors associated with an increased risk of HIV and sexually transmitted infection transmission among street-involved youth. BMC Public Health, 9, 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Martino SC, Tucker JS, Ryan G, Wenzel SL, Golinelli D, & Munjas B (2011). Increased substance use and risky sexual behavior among migratory homeless youth: Exploring the role of social network composition. Journal of Adolescent Health, 40, 1634–1648. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mastro TD, Cunningham J, Medrano T, & van Dam J (2012). Youth and HIV: The intersection of homelessness, orphaned status, injection drug use and sexual risk. AIDS, 26, 111–113. [DOI] [PubMed] [Google Scholar]
- Ober AJ, Martino SC, Ewing B, & Tucker JS (2012). If you provide the test, they will take it: Factors associated with HIV/STI testing in a representative sample of homeless youth in Los Angeles. AIDS Education and Prevention, 24, 350–362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pfeifer RW, & Oliver J (1997). A study of HIV seroprevalence in a group of homeless youth in Hollywood, California. Journal of Adolescent Health, 20, 339–342. [DOI] [PubMed] [Google Scholar]
- Rotheram-Borus MJ, Song J, Gwadz M, Lee M, Van Rossem R, & Koopman C (2003). Reductions in HIV risk among runaway youth. Prevention Science, 4, 173–187. [DOI] [PubMed] [Google Scholar]
- Santa Maria D, Flash CA, Narendorf S, Barman-Adhikari A, Petering R, Hsu H, . . . Ferguson K (2018). Knowledge and attitudes about pre-exposure prophylaxis among young adults experiencing homelessness in seven U.S. cities. Journal of Adolescent Health, 64, 574–580. [DOI] [PubMed] [Google Scholar]
- Solorio MR, Milburn NG, Weiss RE, & Batterham PJ (2006). Newly homeless youth STD testing patterns over time. Journal of Adolescent Health, 39, 443e9–443e16. [DOI] [PubMed] [Google Scholar]
- State of Georgia. Office of the Child Advocate. (2016). Mandated reporting. Retrieved April 2, 2019, from https://oca.georgia.gov/mandated-reporting
- Stricof RL, Kennedy JT, Nattell TC, Weisfuse IB, & Novick LF (1991). HIV sereo prevalence in a facility for runaway and homeless adolescents. American Journal of Public Health, 81(Suppl), 50–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tucker JS, Hu J, Golinelli D, Kennedy DP, Green HD Jr., & Wenzel SL (2012). Social network and individual correlates of sexual risk behavior among homeless young men who have sex with men. Journal of Adolescent Health, 51, 386–392. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tucker JS, Ryan GW, & Golinelli D (2012). Substance use and other risk factors for unprotected sex: Results from an event-based study of homeless youth. AIDS and Behavior, 16, 1699–1707. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tyler KA (2013). Homeless youths’ HIV risk behaviors with strangers: Investigating the importance of social networks. Archives of Sexual Behavior, 42, 1583–1591. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tyler KA, Akinyemi SL, & Kort-Butler LA (2012). Correlates of service utilization among homeless youth. Children and Youth Services Review, 34, 1344–1350. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tyler KA, & Melander L (2010). The effect of drug and sexual risk behaviours with social network and non-network members on homeless youths’ sexually transmissible infections and HIV testing. Sexual Health, 7, 434–440. [DOI] [PubMed] [Google Scholar]
- Wright E, Ruel E, Fuoco M, Trouteaud A, Sanchez T LaBoy A, . . . Hartinger-Saunders R (2016). Atlanta youth count! 2015 homeless youth count and needs assessment. Retrieved December 4, 2018, from https://atlantayouthcount.weebly.com/uploads/7/9/0/5/79053356/aycna_final_report_may_2016_final.pdf
- Young SD, & Rice E (2011). Online social networking technologies, HIV knowledge, and sexual risk and testing behaviors among homeless youth. AIDS and Behavior, 15, 253–260. [DOI] [PMC free article] [PubMed] [Google Scholar]