Abstract
Purpose
Exposure to violence in childhood has been linked to adverse health outcomes. Little is known about the prevalence and relationship of youth and caregiver violence exposure to clinical outcomes among youth with perinatal HIV infection (PHIV). We evaluated associations of youth and caregiver violence exposure with unsuppressed viral load (VL) (HIV RNA>400 copies/ml) and CD4%<25% among 8-15 year-old participants with PHIV in the Pediatric HIV/AIDS Cohort Study (PHACS) Adolescent Master Protocol (AMP).
Methods
Annual clinical examination, record abstraction and interview data were collected, including youth report of recent exposure to violence and caregivers' self-report of being assaulted/abused in adulthood. Multivariable logistic regression methods were used to calculate adjusted odds ratios (OR) for unsuppressed VL and CD4%<25%, controlling for socio-demographic characteristics.
Results
Among 268 youth with PHIV (53% girls, mean age 12.8 years, 21% white, 42% with household income<$20,000/year), 34% reported past year violence exposure; 30% had a caregiver who reported being assaulted in adulthood. One quarter of youth (24%) had unsuppressed VL, and 22% had CD4%<25%. Youth who were exposed to violence in the past year vs. those who were not had elevated odds of unsuppressed VL. Youth with indirect exposure to violence in the past year vs. those without had elevated odds of unsuppressed VL and CD4%<25% in adjusted models.
Conclusion
Youth with PHIV report a high prevalence of recent violence exposure, which was associated with poor virologic and immunologic outcomes. Reducing violence and providing support to youth with violence exposure and PHIV may improve health outcomes.
Keywords: Violence, Adolescents, Gender, Virologic Suppression, Adherence
Violence during childhood and adolescence is a traumatic stressor, associated with a higher risk of adverse health outcomes and mortality [1-4]. It may worsen physical functioning through multiple mechanisms. Violence exposure may increase health-compromising behaviors, induce changes in immune activity and cortisol levels [5,6], contribute to telomere erosion, alter the trajectory of neurobehavioral development [7], and increase risk for mental health problems, including depression [6-9]. Many youth with perinatally acquired HIV (PHIV) and their families reside in communities with high rates of violence and poverty [10,11]. In one US study of 99 youth with PHIV, 19% reported being assaulted, 10% experienced sexual abuse, and 18% had witnessed a stabbing or shooting [11,12]. Stressful life events have been associated with diminished psychological and immune functioning in youth with PHIV[13-15]. However, no studies to date have determined the relationship of violence exposure to markers of HIV disease progression in youth.
In HIV-affected families, a high proportion of women and mothers living with HIV experience intimate partner violence and other traumas [16,17]. Exposure to intimate partner violence, for example, is associated with non-adherence to antiretroviral therapy (ART) [17] virologic failure [17,18] and poorer quality of life [19] among adults with HIV infection. Accumulating evidence demonstrates associations between caregiver exposure to violence, including intimate partner violence, and poorer mental and physical health outcomes in their children [20-24]. Caregivers' exposure to violence may affect children's health outcomes through disruptions in parenting and/or by eliciting fear and trauma in children who witness such violence.
Understanding the prevalence and correlates of violence exposure among youth with PHIV may be a critical step in preventing HIV disease progression, and, since virologic suppression lowers HIV transmission risk [25], reducing HIV transmission to others for the 3.3 million youth living with HIV globally. The objectives of this study were to determine the prevalence of direct violence victimization and indirect exposure to violence in the past year among US youth with PHIV and history of assault during adulthood among their caregivers, and to examine the relationship of youth and caregiver violence exposure to youths' virologic and immunologic outcomes. We hypothesized that youth reporting exposure to violence and those whose caregivers report exposure to violence would have poorer virologic and immunologic outcomes compared to those without violence exposure. We considered adherence to ART as a potential mechanism for these associations. We also investigated whether associations between violence and markers of HIV disease severity were similar in girls and boys with PHIV, since effects of violence exposure could vary by sex [23, 26-28].
Methods
Study Population
The Adolescent Master Protocol (AMP) of the Pediatric HIV/AIDS Cohort Study (PHACS) is a prospective study examining the impact of HIV infection and antiretroviral therapy on perinatally HIV-infected children and adolescents. Youth with PHIV and a comparison cohort of perinatally HIV-exposed, uninfected youth and their caregivers enrolled at one of 15 clinical sites in the US including Puerto Rico, between March 2007 and October 2009. Participants completed follow-up study visits every 6 months (2007-2010) or annually (2010-2015). Study visits included clinical examinations, medical record abstraction, and interviews. Inclusion criteria for AMP were: age 7-15 at entry, and, if PHIV, engaged in medical care, with available ART history data. Institutional Review Boards at all sites and at Harvard TH Chan School of Public Health approved the protocol. Parents or legal guardians provided written informed consent and youth provided assent per local IRB guidelines.
This analysis included only youth with PHIV ages 8-15 years, when youths' self-reported violence exposure was assessed, and used data from the earliest visit when youth and caregivers completed interviews about their exposure to violence. Violence exposure data for caregivers whose first interview occurred later than the child's were included if obtained within a year of the child's interview.
Outcome Measures
The primary outcomes were a) unsuppressed viral load (VL), defined as HIV-RNA >400 copies/ml, and b) CD4 percentage <25% (CD4%<25%). HIV RNA and CD4 lymphocyte data were abstracted from medical records at each visit. Analyses included those measures obtained concurrently with assessment of youth's exposure to violence (within +/- 3 months).
Primary Exposure Measures
Primary exposures of interest included: a) youth's self-reported exposure to violence in the past year (defined as indirect exposure to violence or direct violence victimization), and b) caregiver's self-reported history of having been physically or sexually assaulted in adulthood. Youth exposure to violence was assessed as part of the Life Events Checklist (LEC) interview, a measure assessing youth self-reported exposure to stressful life events within the past year [10,29]. Seven of the items from the LEC pertain to witnessed or experienced violence.
Three dichotomous composite indicators of youth violence exposure in the past year were created from the seven LEC items. 1) Indirect Exposure to Violence was defined as youth report of at least one of the following events in the past year: a) witnessed a fight in which a weapon was used, b) heard gunshots on their block, c) people in the neighborhood were hit by the police, or d) someone in the neighborhood was murdered; 2) Direct Violence Victimization was defined as youth report of having been a) physically attacked, b) raped or sexually assaulted, or c) robbed or burglarized in the past year; 3) Any Violence Exposure was defined as youth report of either indirect exposure to violence, or direct violence victimization in the past year, as indicated above.
Caregiver exposure to violence was assessed using the Client Diagnostic Questionnaire (CDQ) Short Form [30], a structured psychiatric interview conducted with caregivers that includes a list of traumatic events, including three items pertaining to caregiver's history of physical and/or sexual assault in adulthood. We created two composite indicators of caregiver exposure to violence from the three CDQ items related to violence: 1) Caregiver assault in adulthood was defined as caregiver-reported history of a) sexual assault or rape, b) physical assault or abuse by a partner or c) physical assault or abuse by someone other than a partner, in adulthood; 2) Caregiver assault by partner was defined as caregiver report of having experienced physical assault or abuse by a partner in adulthood.
Two different summary scores of exposure to violence were also calculated: 1) Youth Exposure to Violence score (0-7) - the sum of affirmative responses to the 7 LEC violence exposure items, and 2) Combined Youth/Caregiver Exposure to Violence score (0-10) - the sum of the 7 LEC violence exposure items and the three CDQ caregiver abuse/assault items. Youth missing data on all 7 youth LEC items or all 3 caregiver items were excluded from analyses of those variables.
Other Covariates
The following socio-demographic characteristics were obtained during caregiver interviews completed when the LEC was administered: child's age, sex, race/ethnicity, knowledge of HIV status and annual household income, as well as caregiver's highest level of education, relationship to child, marital status, and presence of caregiver's partner or spouse in the home. HIV clinical characteristics including Centers for Disease Control (CDC) C Classification and current antiretroviral regimen were obtained through medical chart abstraction. HAART was defined as three or more drugs from two or more classes. ART adherence was assessed during separate interviews with youth and their caregivers using an adherence questionnaire developed by the Adult AIDS Clinical Trials Group [31], adapted by the Pediatric AIDS Clinical Trials Group [32], and subsequently modified by the PHACS study [33]. Non-adherence was defined as youth or caregiver report of youth having missed one or more ART doses in the past 7 days (week) [33].
Statistical Analysis
We summarized and compared the frequency of demographic and HIV clinical characteristics by youth-reported exposure to violence in the past year, and caregiver-reported history of assault in adulthood. We evaluated bivariate associations of demographic and HIV clinical characteristics with unsuppressed VL (vs. suppressed), and with CD4% <25% (vs. ≥25%). Chi square or exact unconditional tests (as appropriate) were used to evaluate differences in the distribution of categorical variables and Wilcoxon rank sum and t-tests, as appropriate, were used to test differences in the distribution of continuous variables by VL and CD4% categories, and each type of violence exposure.
We investigated the association of exposure to violence with unsuppressed VL and with CD4%<25%. HIV clinical variables (knowledge of HIV status, and antiretroviral regimen) and socio-demographic variables were considered as potential confounders. We fit separate logistic regression models with exposure to each type of violence and used multivariable regression methods to adjust for potential confounders. Variables associated with both violence exposure and the outcome at p<=0.1, which were not considered to be potentially on the pathway between violence exposure and the outcome were included in multivariable models. When potential confounders were correlated (e.g. caregiver marital status and caregiver living with a partner, age and knowledge of HIV status) the variable with the larger effect estimate was selected for inclusion. Bivariate analyses examined whether adherence, considered to be on the causal pathway between violence exposure and the outcomes, was associated with both the violence exposures and the outcomes to evaluate whether it was a potential mediator. We conducted stratified analyses to explore whether associations between violence and VL and CD4 were similar for girls and boys, then fit models with a sex and violence exposure interaction term. The significance criterion was a p-value<0.05. Marginal associations (p-value<0.10) were also reported.
Results
Sample Characteristics
Among the 289 youth with PHIV who had exposure to violence data, 268 had concurrent VL and CD4% data and were included in these analyses. Characteristics of the 21 participants excluded were similar to those included in the analysis, but those with missing data were more likely than included participants to be white (52% vs. 21%, p=0.05) or Hispanic (52% vs. 21%, p= 0.001).
Of 268 youth participants, 53% were girls, mean age was 13 years, 71% were Black, 21% were Latino or Latina, and 42% lived in households with a household income under $20,000/year. Twenty-four percent of participants had unsuppressed VL and 22% had a CD4% <25%. Thirty-four percent of youth reported violence exposure in the past year and 30% had caregivers who reported a history of either physical or sexual assault in adulthood (Table 1). Sample characteristics did not vary by sex except that girls were less likely than boys to be on a protease inhibitor-based regimen (65% vs. 78%, p=0.02). Youth reporting violence exposure in the past year were more frequently older, living with a primary caregiver who was a biological relative other than their parent, or living with a caregiver who had not graduated from high school, or who was widowed and/or not living with a partner (Table 1). Youth exposed to violence were also more likely to have a CDC class C diagnosis, to report non-adherence to ART, or to have unsuppressed VL and CD4%<25%. Youth whose caregivers reported histories of assault in adulthood more frequently lived with their biological mother, had a CDC class C diagnosis, or were aware that they were HIV positive than those whose caregivers reported no adult experience of assault. Youth ART adherence was not associated with caregiver history of assault in adulthood.
Table 1. Sociodemographic, HIV disease, and Caregiver Characteristics of 268 8-15 year-old Youth with PHIV, by Exposure to Violence, Pediatric HIV/AIDS Cohort Study (2008-2013).
| Youth-reported violence exposure, past year | Caregiver assaulted in adulthood | Total (N=268) | |||||
|---|---|---|---|---|---|---|---|
|
|
|
||||||
| None (N=177) | Any1 (N=91) | p | No (N=186) | Yes (N=80) | p | ||
| Female | 94 (53%) | 48 (53%) | 0.95 | 94 (51%) | 47 (59%) | 0.22 | 142 (53%) |
| Race | 0.13 | 0.83 | |||||
| Asian | 1 (1%) | 0 (0%) | 1 (1%) | 0 (0%) | 1 (0%) | ||
| Black or African American | 118 (67%) | 71 (78%) | 129 (69%) | 58 (73%) | 189 (71%) | ||
| White | 41 (23%) | 15 (16%) | 40 (22%) | 16 (20%) | 56 (21%) | ||
| American Indian | 1 (1%) | 1 (1%) | 1 (1%) | 1 (1%) | 2 (1%) | ||
| More than One Race | 8 (5%) | 0 (0%) | 7 (4%) | 1 (1%) | 8 (3%) | ||
| Unknown | 8 (5%) | 4 (4%) | 8 (4%) | 4 (5%) | 12 (4%) | ||
| Latina or Latino Ethnicity | 39 (22%) | 18 (20%) | 0.67 | 40 (22%) | 17 (21%) | 0.96 | 57 (21%) |
| Age 12 years or older | 99 (56%) | 62 (68%) | 0.05 | 107 (58%) | 53 (66%) | 0.18 | 161 (60%) |
| Caregiver relationship | 0.02 | 0.02 | |||||
| Biological mother | 70 (40%) | 34 (38%) | 61 (33%) | 43 (54%) | 104 (39%) | ||
| Biological father | 10 (6%) | 7 (8%) | 15 (8%) | 2 (3%) | 17 (6%) | ||
| Other biological family member | 35 (20%) | 32 (36%) | 49 (26%) | 17 (21%) | 67 (25%) | ||
| Adoptive or foster parent | 55 (31%) | 15 (17%) | 54 (29%) | 15 (19%) | 70 (26%) | ||
| Step-parent or non-relative | 7 (4%) | 2 (2%) | 6 (3%) | 3 (4%) | 9 (3%) | ||
| Annual household income $20,000 or less | 73 (41%) | 39 (43%) | 0.79 | 72 (39%) | 40 (50%) | 0.23 | 112 (42%) |
| Caregiver graduated high school/GED | 136 (77%) | 57 (64%) | 0.02 | 138 (75%) | 53 (67%) | 0.19 | 193 (73%) |
| Caregiver marital status | 0.09 | 0.16 | |||||
| Married | 87 (49%) | 31 (35%) | 87 (47%) | 30 (38%) | 118 (45%) | ||
| Separated/divorced | 24 (14%) | 14 (16%) | 21 (11%) | 17 (22%) | 38 (14%) | ||
| Widowed | 15 (9%) | 14 (16%) | 19 (10%) | 9 (11%) | 29 (11%) | ||
| Single, never married | 50 (28%) | 30 (34%) | 57 (31%) | 23 (29%) | 80 (30%) | ||
| Caregiver living with partner/spouse | 98 (56%) | 37 (42%) | 0.03 | 99 (54%) | 35 (44%) | 0.34 | 135 (51%) |
| CDC Class C | 35 (20%) | 29 (32%) | 0.03 | 37 (20%) | 26 (33%) | 0.03 | 64 (24%) |
| ARV Regimen Grouping | 0.76 | 0.55 | |||||
| HAART with PI | 122 (70%) | 67 (74%) | 133 (72%) | 56 (71%) | 189 (71%) | ||
| HAART without PI | 33 (19%) | 14 (15%) | 34 (18%) | 12 (15%) | 47 (18%) | ||
| Non-HAART ARV | 11 (6%) | 6 (7%) | 11 (6%) | 6 (8%) | 17 (6%) | ||
| Not on ARV | 8 (5%) | 4 (4%) | 7 (4%) | 5 (6%) | 12 (5%) | ||
| Missed ≥1 ART dose in past 7 days | 37 (23%) | 35 (44%) | <0.001 | 49 (29%) | 22 (32%) | 0.57 | 72 (30%) |
| Child knows his/her HIV status | 118 (68%) | 61 (67%) | 0.89 | 117 (63%) | 62 (78%) | 0.03 | 179 (68%) |
| HIV RNA >400 copies/mL | 32 (18%) | 34 (37%) | <0.001 | 42 (23%) | 23 (29%) | 0.29 | 65 (24%) |
| CD4%<25% | 30 (17%) | 28 (31%) | 0.009 | 39 (21%) | 18 (23%) | 0.78 | 58 (22%) |
| Caregiver assaulted in adulthood | 51 (29%) | 29 (33%) | 0.53 | -- | -- | 80 (30%) | |
defined as youth report of at least one of the following in the past year: a) physically attacked, b) raped or sexually assaulted c) robbed or burglarized d) witnessed a fight in which a weapon was used, e) heard gunshots on their block, f) people in the neighborhood were hit by the police g) someone in the neighborhood was murdered in the past year; Missing observations: Caregiver relationship, n=1; race, n=12; annual household income, n=12; caregiver education, n=3; caregiver marital status, n=3; caregiver living with partner/spouse, n=3; CDC Class C, n=3; ARV regimen grouping, n=3; missed ≥1 ARV dose in past 7 days, n=27; child knows his/her HIV status, n=3. PHIV=Perinatally-acquired HIV infection
Prevalence of Specific Types of Violence
Table 2 summarizes the prevalence of specific types of violence exposure for all youth and by sex. Ten percent of youth with PHIV reported direct violence victimization in the past year, and 29% reported indirect exposure to violence; 24% had a caregiver with reported history of assault by a partner. In total, 53% of youth with PHIV reported any violence exposure in the past year or had a caregiver who reported having been assaulted in adulthood, of whom 36% had caregiver-reported exposure only, 34% reported past year indirect exposure only, 6% past year direct victimization only, and 24% more than one type of violence exposure. Boys were more likely than girls to have witnessed a fight involving a weapon, and marginally more likely to report having been physically attacked (Table 2).
Table 2. Prevalence of violence exposure among 268 8-15 year-old youth with PHIV, by sex, Pediatric HIV/AIDS Cohort Study (2008-2013).
| Characteristic | Girls (N=142) | Boys (N=126) | Total (N=268) | P-value |
|---|---|---|---|---|
| Youth-reported violence exposure, past year | 48 (33.8%) | 43 (34.1%) | 91 (34%) | 0.95 |
| Physically attacked | 5 (3.5%) | 11 (8.7%) | 16 (6%) | 0.07 |
| Sexually assaulted or raped | 1 (<1%) | 0 | 1 (<1%) | 0.53 |
| Robbed or burglarized | 7 (4.9%) | 5 (3.9%) | 12 (4%) | 0.70 |
| Witnessed fight in which weapon used | 4 (2.8%) | 11 (8.7%) | 15 (6%) | 0.03 |
| Heard gunshots on block | 26 (18.3%) | 27 (21.6%) | 53 (20%) | 0.50 |
| People in neighborhood hit by police | 8 (5.6%) | 12 (9.5%) | 20 (7%) | 0.23 |
| Someone in neighborhood murdered | 11 (7.7%) | 12 (9.5%) | 23 (9%) | 0.60 |
| Witnessed or indirect exposure to violence1 | 40 (28.2%) | 37 (29.4%) | 77 (29%) | 0.83 |
| Direct violence victimization2 | 12 (8.4%) | 15 (11.9%) | 27 (10%) | 0.35 |
| Youth exposure to violence score, Mean (s.d.) | 0.44 (0.69) | 0.62 (1.04) | 0.52 (0.88) | 0.09 |
| Median (Q1, Q3) | 0 (0, 1) | 0 (0, 1) | 0 (0, 1) | 0.55 |
| Caregiver physically or sexually assaulted in adulthood | 47 (33.3%) | 33 (26.4%) | 80 (30%) | 0.22 |
| Caregiver physically assaulted or abused by a partner | 38 (26.9%) | 26 (20.8%) | 64 (24%) | 0.24 |
| Caregiver physically assaulted or abused by non-partner | 12 (18.5%) | 13 (10.4%) | 25 (9%) | 0.60 |
| Caregiver sexually assaulted or raped | 17 (12.1%) | 11 (8.8%) | 28 (11%) | 0.39 |
| Youth and caregiver combined exposure to violence score, Mean (s.d.) | 0.91 (1.14) | 1.01 (1.32) | 0.95 (1.18) | 0.49 |
| Median (Q1, Q3) | 1 (0, 1) | 1 (0, 2) | 1 (0, 2) | 0.94 |
Witnessed fight in which weapon used, heard gunshots on block, people in neighborhood hit by police, or someone in neighborhood murdered in the past year;
Physically attacked, sexually assaulted or raped, or robbed or burglarized in the past year. PHIV=Perinatally-acquired HIV infection
Exposure to Violence, Unsuppressed VL and CD4%<25%
In separate unadjusted models, youth report of any violence exposure in the past year, indirect exposure to violence, and higher Youth Exposure to Violence scores were associated with unsuppressed VL and CD4%<25% (Table 3). Higher Combined Youth and Caregiver Exposure to Violence scores were associated with unsuppressed VL but not with CD4%<25%. Youth whose caregiver reported a history of assault by a partner had marginally greater odds of unsuppressed VL but not CD4%<25%. Direct youth violence victimization in the past year, and having a caregiver who reported having been assaulted in adulthood were not associated with unsuppressed VL or with CD4%<25%.
Table 3. Unadjusted and Adjusted Associations of Exposure to Violence with Virologic and Immunologic Outcomes Among 268 Youth with PHIV, Pediatric HIV/AIDS Cohort Study (2008-2013).
| HIV RNA>400 copies/mL | CD4%<25% | |||||||
|---|---|---|---|---|---|---|---|---|
|
|
|
|||||||
| OR (95% CI) | p | Adjusted OR (95% CI) | p | OR (95% CI) | p | Adjusted OR (95% CI) | p | |
| Any Youth-reported Violence Exposure, past year (referent: none) | 2.70 (1.53, 4.81) | <0.001 | 1.91 (1.02, 3.55) | 0.04 | 2.18 (1.20, 3.95) | 0.01 | 1.52 (0.79, 2.92) | 0.21 |
| Youth Direct Violence Victimization, past year (referent: none) | 1.94 (0.82, 4.42) | 0.12 | 1.69 (0.67, 4.24) | 0.27 | 0.81 (0.26, 2.07) | 0.68 | 0.63 (0.21, 1.88) | 0.40 |
| Youth Indirect Exposure to Violence, past year (referent: none) | 2.75 (1.53, 4.94) | <0.001 | 1.84 (0.97, 3.48) | 0.06 | 3.07 (1.67, 5.64) | <0.001 | 2.15 (1.11, 4.17) | 0.02 |
| Youth Exposure to Violence score | 1.58 (1.18, 2.13) | 0.002 | 1.29 (0.93, 1.78) | 0.13 | 1.42 (1.04, 1.91) | 0.02 | 1.17 (0.83, 1.64) | 0.37 |
| Caregiver History of Assault in Adulthood (referent: none) | 1.38 (0.76, 2.49) | 0.28 | 1.14 (0.60, 2.17) | 0.68 | 1.09 (0.57, 2.04) | 0.78 | 0.85 (0.43, 1.68) | 0.65 |
| Caregiver History of Assault by Partner (referent: none) | 1.75 (0.93, 3.24) | 0.08 | 1.36 (0.69, 2.66) | 0.37 | 1.31 (0.66, 2.50) | 0.43 | 0.95 (0.46, 1.94) | 0.88 |
| Combined Youth/Caregiver Exposure to Violence score | 1.40 (1.12, 1.76) | 0.004 | 1.19 (0.93, 1.52) | 0.18 | 1.21 (0.95, 1.52) | 0.12 | 1.00 (0.77, 1.30) | 0.99 |
Separate models were fit for each type of violence exposure. Adjusted models were adjusted for age, caregiver relationship, caregiver education level and caregiver living with partner/spouse. OR=Odds Ratio; CI=Confidence Interval; PHIV=Perinatally-acquired HIV infection
In multivariable models adjusting for youth age, caregiver education, caregiver-youth relationship, and caregiver living with a partner or spouse, any youth-reported violence exposure remained associated with unsuppressed VL but not with CD4% <25%. Indirect exposure to violence remained associated with CD4 %<25%, and marginally with unsuppressed VL (Table 3).
Non-Adherence and Unsuppressed VL and CD4%<25%
Youth with indirect exposure to violence were more likely to be non-adherent to ART (38% vs. 22%, p=0.002) than those without violence exposure. Youth who were non-adherent had elevated odds of unsuppressed VL (32% vs. 17%, p<0.0001) and CD4%<25% (29% vs. 17%, p=0.04) compared to adherent youth.
Sex, Violence Exposure and Unsuppressed VL and CD4%
Associations between some forms of violence and the two outcomes differed for boys versus girls (Figure 1a-b). Boys who reported indirect violence exposure in the past year were more likely to have unsuppressed VL vs. those without exposure, while for girls the difference in the prevalence of unsuppressed VL in those reporting indirect violence exposure vs. no exposure in the past year was smaller (Figure 1a). Among girls only, caregiver-reported adult history of assault by a partner was marginally associated with unsuppressed VL (p=0.08) (Figure 1b). We identified an interaction between indirect violence exposure in the past year and sex for unsuppressed VL (p=0.04) in unadjusted analyses (data not shown in tables). Boys who reported indirect violence exposure had elevated odds of unsuppressed VL relative to boys without exposure (OR=5.45, 95% CI: 2.24, 13.25, p=0.0002), which persisted after adjustment for age, caregiver education, caregiver relationship, and caregiver living with a partner or spouse (aOR=3.39, 95% CI: 1.30, 8.83). Among girls, indirect violence exposure was not associated with unsuppressed VL (OR=1.56, 95% CI: 0.7, 3.49). In adjusted models, the sex and indirect violence exposure interaction was not statistically significant (p=0.09). No statistically significant interactions were observed between sex and violence exposures in relation to CD4%, or between sex and other specific violence exposures and unsuppressed VL.
Figure 1.
a: Prevalence of unsuppressed viral load, by youth-reported exposure to violence in the past year among boys and girls with PHIV
b: Prevalence of unsuppressed viral load, by caregiver history of being assaulted in adulthood among boys and girls with PHIV
Discussion
We observed a high prevalence of violence exposure among youth with PHIV and their caregivers. Over half of youth reported violence exposure in the past year or had a caregiver who reported a history of assault during adulthood. Youth with PHIV and violence exposure in the past year had 2-fold elevated odds of having unsuppressed VL, and recent indirect exposure to violence was associated with CD4%<25%. Although earlier studies have shown associations of cumulative non-violent stressful life events with immunological outcomes [13-15], this is the first to examine violence exposure specifically, and to reveal associations with virologic and immunologic outcomes in youth with PHIV. These results highlight the need for strategies both to reduce violence exposure and address violence exposure experiences among youth with PHIV and their caregivers, which may reduce risks of HIV-associated complications and transmission of HIV, and promote youth well-being.
The proportion of youth with recent exposure to violence in our study is lower than that reported in the National Survey of Children's Exposure to Violence, which found that 37% of children 0-18 years of age experienced a physical assault in the past year. However, rates of some specific types of violence including witnessing violence in the community (18%), and experiencing sexual assault (1.5%) and robbery (6.5%) in the past year were similar [34]. Rates of youth-reported physical assault in our study were also similar to the 6% prior to age 13 previously reported by US youth with PHIV [11]. However, rates of other types of violence were lower than the previous study, perhaps in part reflecting the shorter recall period for violence exposure in our study (past year instead of lifetime), and lower mean age of youth with PHIV in our study. Comparison of results across studies is challenging due to differences in the types of violence measured, assessment tools, and sample demographic characteristics.
We found that youth reporting any violence exposure, direct or indirect, in the previous year had elevated odds of unsuppressed VL, but there was not significantly increased risk for reduced immunologic function, as measured by CD4%. In contrast, however, youth who were indirectly exposed to violence in the past year had elevated odds of both CD4%<25% and unsuppressed VL (the latter being marginally statistically significant). We did not observe associations of direct victimization with either clinical outcome, which may be due to the small number of youth reporting recent direct violence victimization.
Suboptimal ART adherence is one likely mechanism by which indirect violence exposure is associated with virologic and immunologic outcomes among youth with PHIV. In our study, youth reporting exposure to violence in the past year were more likely to have suboptimal adherence which was, in turn, associated with unsuppressed VL and CD4% <25%. Non-adherence may mediate the association between youth's exposure to violence and both unsuppressed VL and CD4% <25%. Longitudinal studies that account for the temporal ordering of violence exposure, subsequent non-adherence and later virologic and immunologic outcomes are needed to formally explore these mechanisms and evaluate mediation. Violence could also be an expression, in part, of HIV-related stigma and discrimination, both of which have been associated with lower levels of adherence and engagement in medical care [35]. Importantly, witnessing violence may be directly associated with physical alterations, including changes in cortisol levels, [36,37] and mental health sequelae, including PTSD and depression [7,8,38,39], all of which potentially influence health maintenance behaviors, virologic and immunologic functioning. Studies examining mechanisms of the violence-viral load relationship will be helpful to inform interventions.
We did not observe associations of caregiver exposure to violence with youth virologic and immunologic outcomes, or with youth ART adherence. The recall period for the assessment of caregiver violence exposure was longer than that for the youth violence exposure assessment, which may account in part for the lack of associations. Caregivers with trauma histories may have received counseling and support to cope with the effects of violence exposure which could mitigate its potentially negative impact on their health and parenting. In addition, caregivers' exposure to violence may have occurred prior to their child's birth, when their child was either not living with them or not present. Additionally, effects of caregivers' exposure to violence on children's health may manifest at other time points in children's lives, or in youths' developmental or psychological outcomes.
Our analyses suggest that recent indirect exposure to violence was related to unsuppressed VL in boys but not girls, but associations of indirect exposure to violence with CD4%, and of other forms of violence exposure with VL and CD4% did not vary by sex. Although the overall proportions of boys and girls reporting indirect exposure to violence was similar, boys reported higher rates of witnessing fights than girls. Since timing, type and intensity of different types of violence exposure during childhood and adolescence likely varies considerably by gender [23, 36, 34], as do cultural norms, coping skills and support systems, further study is necessary to clearly understand the ways in which exposure to violence may affect clinical outcomes differently in boys and girls.
Results from this study should be interpreted in light of some limitations. Because this study was cross-sectional, we are unable to determine the temporal ordering between violence exposure and clinical outcomes. We measured past year exposure but not frequency or cooccurrence of multiple types of violence exposure, and did not assess youths' lifetime history, which may have underestimated cumulative exposure and resulted in misclassification. Additionally, we did not use a standard measure of violence exposure and did not evaluate youth knowledge or experience of their caregiver's violence exposure.
These results have implications for clinical care, and for programs, policies and research to support youth with PHIV and their families. The high prevalence of recent violence exposure among youth with PHIV underscores the need in HIV clinical care settings for enhanced screening for violence exposure coupled with increasing awareness of and linkages to family and community resources, including integration of mental health services. Early identification of risk for violence and psychosocial support for youth and caregivers may mitigate negative HIV disease outcomes. Addressing violence as part of adherence support interventions in clinic and community settings is a likely mechanism to protect virologic and immunologic well-being and prevent HIV transmission. Multisectoral culturally relevant social, economic, recreational and educational programs that aim to reduce neighborhood violence, enhance security, foster resilience, and provide opportunities and resources for safe and healthy youth and family development may be appropriate interventions [7]. Future interdisciplinary longitudinal research should examine the applicability of current results globally where most youth with PHIV reside.
Implications and Contribution.
This study found high prevalence of violence exposure among youth with PHIV, which was associated with poor virologic and immunological outcomes. These results highlight the need to identify and address violence exposure experiences among youth with PHIV and their caregivers, which may reduce risks of HIV-associated complications and transmission of HIV.
Acknowledgments
We thank the children and families for their participation in PHACS, and the individuals and institutions involved in the conduct of PHACS. The study was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development with co-funding from the National Institute on Drug Abuse, the National Institute of Allergy and Infectious Diseases, the Office of AIDS Research, the National Institute of Mental Health, the National Institute of Neurological Disorders and Stroke, the National Institute on Deafness and Other Communication Disorders, the National Heart Lung and Blood Institute, the National Institute of Dental and Craniofacial Research, and the National Institute on Alcohol Abuse and Alcoholism, through cooperative agreements with the Harvard T.H. Chan School of Public Health (HD052102) (Principal Investigator: George Seage; Project Director: Julie Alperen) and the Tulane University School of Medicine (HD052104) (Principal Investigator: Russell Van Dyke; Co-Principal Investigator: Kenneth Rich; Project Director: Patrick Davis). Data management services were provided by Frontier Science and Technology Research Foundation (PI: Suzanne Siminski), and regulatory services and logistical support were provided by Westat, Inc (PI: Julie Davidson).
The following institutions, clinical site investigators and staff participated in conducting PHACS AMP in 2014, in alphabetical order: Ann & Robert H. Lurie Children's Hospital of Chicago: Ram Yogev, Margaret Ann Sanders, Kathleen Malee, Scott Hunter; Baylor College of Medicine: William Shearer, Mary Paul, Norma Cooper, Lynnette Harris; Bronx Lebanon Hospital Center: Murli Purswani, Mahboobullah Baig, Anna Cintron; Children's Diagnostic & Treatment Center: Ana Puga, Sandra Navarro, Patricia Garvie, James Blood; Children's Hospital, Boston: Sandra Burchett, Nancy Karthas, Betsy Kammerer; Jacobi Medical Center: Andrew Wiznia, Marlene Burey, Molly Nozyce; Rutgers - New Jersey Medical School: Arry Dieudonne, Linda Bettica, Susan Adubato; St. Christopher's Hospital for Children: Janet Chen, Maria Garcia Bulkley, Latreaca Ivey, Mitzie Grant; St. Jude Children's Research Hospital: Katherine Knapp, Kim Allison, Megan Wilkins; San Juan Hospital/Department of Pediatrics: Midnela Acevedo-Flores, Heida Rios, Vivian Olivera; Tulane University Health Sciences Center: Margarita Silio, Medea Jones, Patricia Sirois; University of California, San Diego: Stephen Spector, Kim Norris, Sharon Nichols; University of Colorado Denver Health Sciences Center: Elizabeth McFarland, Alisa Katai, Jennifer Dunn, Suzanne Paul; University of Miami: Gwendolyn Scott, Patricia Bryan, Elizabeth Willen.
Note: The conclusions and opinions expressed in this article are those of the authors and do not necessarily reflect those of the National Institutes of Health or U.S. Department of Health and Human Services.
Sources of Funding: The Pediatric HIV/AIDS Cohort Study (PHACS) was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development with co-funding from the National Institute on Drug Abuse, the National Institute of Allergy and Infectious Diseases, the Office of AIDS Research, the National Institute of Mental Health, the National Institute of Neurological Disorders and Stroke, the National Institute on Deafness and Other Communication Disorders, the National Heart Lung and Blood Institute, the National Institute of Dental and Craniofacial Research, and the National Institute on Alcohol Abuse and Alcoholism, through cooperative agreements with the Harvard T.H. Chan School of Public Health (HD052102) and the Tulane University School of Medicine (HD052104).
List of Abbreviations
- AMP
Adolescent Master Protocol
- aOR
Adjusted Odds Ratio
- ART
Antiretroviral therapy
- CDC
Centers for Disease Control and Prevention
- CDQ
Client Diagnostic Questionnaire
- HAART
Highly Active Antiretroviral Therapy
- HIV
Human Immunodeficiency Virus
- IRB
Institutional Review Board
- LEC
Life Events Checklist
- OR
Odds Ratio
- PHACS
Pediatric HIV/AIDS Cohort Study
- VL
Viral Load
Footnotes
Conflicts of Interest: None of the authors declare conflicts of interest.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Felitti FJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. Am J Preventive Med. 1998;14:245–258. doi: 10.1016/s0749-3797(98)00017-8. [DOI] [PubMed] [Google Scholar]
- 2.Foster H, Brooks-Gunn J. Effects of physical family and community violence on child development. Tremblay RE, topic. In: Tremblay RE, Boivin M, peters RDeV, editors. Encyclopedia on Early Childhood Development. Montreal, Quebec: Center of Excellence for Early Childhood Development and Strategic Knowledge Cluster on Early Child Development; 2011. [Accessed March 20, 2012]. pp. 1–7. online. Available at: http://www.child-encyclopedia.com/documents/Foster-Brooks-GunnANGxp1.pdf. [Google Scholar]
- 3.Boynton-Jarrett R, Ryan LM, Berkman LF, et al. Cumulative violence exposure and self-rated health: Longitudinal study of adolescents in the United States. Pediatrics. 2008;122:961–70. doi: 10.1542/peds.2007-3063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.MacMillan R. Violence and the life course: the consequences of victimization for personal and social development. Annu Rev Sociol. 2001;27:1–22. [Google Scholar]
- 5.McEwan BS, Stellar E. Stress and the individual: Mechanisms leading to disease. Arch Intern Med. 1993;153:2093–2101. [PubMed] [Google Scholar]
- 6.Shonkoff JP, Garner A the Committee on Psychosocial Aspects of Child and Family Health, Committee on Early Childhood, Adoption, and Dependent Care, and Section on Developmental and Behavioral Pediatrics. The lifelong effects of early childhood adversity and toxic stress. Pediatrics. 2012;129:e232–e246. doi: 10.1542/peds.2011-2663. [DOI] [PubMed] [Google Scholar]
- 7.Karatoreos IN, McEwen BS. Annual Research Review. The neurobiology and physiology of resilience and adaptation across the life course. Journal of Child Psychology and Psychiatry. 2013;54(4):337–347. doi: 10.1111/jcpp.12054. [DOI] [PubMed] [Google Scholar]
- 8.Moffitt TE the Klaus-Grawe 2012 Think Tank. Childhood exposure to violence and lifelong health: Clinical intervention science and stress biology research join forces. Dev Psychopathol. 2013;25(402) doi: 10.1017/S0954579413000801. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Pantalone DW, Hessler DM, Simoni JM. Pathways from interpersonal violence to health related outcomes in HIV positive sexual minority men. J Consult Clin Psych. 2010;38:387–97. doi: 10.1037/a0019307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kang E, Mellins CA, Dolezal C, et al. Disadvantaged neighborhood influences on depression and anxiety in youth with perinatally acquired human immunodeficiency virus: How life stressors matter. J Community Psychol. 2011;39:956–971. doi: 10.1002/jcop.20483. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lyon ME, Koenig LJ, Pals SL, et al. Prevalence and correlates of violence exposure among HIV-infected adolescents. J Assoc Nurses in AIDS Care. 2013;25(1 Suppl):S5–14. doi: 10.1016/j.jana.2013.02.004. Epub 2013 Jun 29. [DOI] [PubMed] [Google Scholar]
- 12.Koenig LJ, Pals SL, Chandwani S, et al. Sexual transmission risk behavior of adolescents with HIV acquired perinatally or through risky behaviors. J Acquir Immune Defic Syndr. 2010;55:380–90. doi: 10.1097/QAI.0b013e3181f0ccb6. [DOI] [PubMed] [Google Scholar]
- 13.Howland LC, Gortmaker SL, Mofenson LM, et al. Effects of negative life events on immune suppression in children and youth infected with Human Immunodeficiency Virus Type 1. Pediatrics. 2000;106:540–546. doi: 10.1542/peds.106.3.540. [DOI] [PubMed] [Google Scholar]
- 14.Howland LC, Storm DS, Crawford SL, et al. Negative life events: risk to health-related quality of life in children and youth with HIV infection. J Assoc Nurses in AIDS Care. 2007;18(1):3–11. doi: 10.1016/j.jana.2006.11.008. [DOI] [PubMed] [Google Scholar]
- 15.Elliot-DeSorbo DK, Martin S, Wolters PL. Stressful life events and their relationship to psychological and medical functioning in children and adolescents with HIV infection. J Acquir Immune Defic Syndr. 2009;52:364–370. doi: 10.1097/QAI.0b013e3181b73568. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Malee KM, Mellins CA, Huo Y, et al. Prevalence, incidence, and persistence of psychiatric and substance use disorders among mothers living with HIV. J Acquir Immune Defic Syndr. 2014;65:526–34. doi: 10.1097/QAI.0000000000000070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sullivan KA, Messer LC, Quinlivan EB. Substance abuse, violence and HIV/AIDS (SAVA) Syndemic effects on viral suppression among HIV positive women of color. AIDS Patient Care STDs. 2015;29:S42–48. doi: 10.1089/apc.2014.0278. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Mugavero MJ, Raper JL, Reif S, et al. Overload: Impact of incident stressful events on antiretroviral medication adherence and virologic failure in a longitudinal, multisite human immunodeficiency virus cohort study. Psychosom Med. 2009;71:920–26. doi: 10.1097/PSY.0b013e3181bfe8d2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.McDonnell KA, Gielen AC, O'Campo P, et al. Abuse, HIV status and health-related quality of life among a sample of HIV-positive and HIV negative low income women. Qual Life Res. 2005;14:945–57. doi: 10.1007/s11136-004-3709-z. [DOI] [PubMed] [Google Scholar]
- 20.Bauer NS, Gilbert AL, Carroll AE, et al. Associations between early exposure to intimate partner violence, parental depression and subsequent mental health outcomes. JAMA Pediatrics. 2013;167(4):341–47. doi: 10.1001/jamapediatrics.2013.780. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Carpenter GL, Stacks AM. Developmental effects of exposure to intimate partner violence in early childhood: A review of the literature. Child Youth Serv Rev. 2009;31:831–839. [Google Scholar]
- 22.Emery CR. Controlling for selection effects in the relationship between child behavior problems and exposure to intimate partner violence. J Interpers Violence. 2011;26:1541–58. doi: 10.1177/0886260510370597. [DOI] [PubMed] [Google Scholar]
- 23.Jun HJ, Corliss HL, Boynton-Jarrett R, et al. Growing up in a domestic violence environment: relationship with developmental trajectories of body mass index during adolescence into young adulthood. J Epidemiol Community Health. 2012;66:629–35. doi: 10.1136/jech.2010.110932. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Holt S, Buckley H, Whelan S. The impact of exposure to domestic violence on children and young people: a review of the literature. Child Abuse Negl. 2008;32:797–810. doi: 10.1016/j.chiabu.2008.02.004. [DOI] [PubMed] [Google Scholar]
- 25.Cohen MS, Chen YQ, McCauley M, et al. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med. 2011;365:493–505. doi: 10.1056/NEJMoa1105243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Zona K, Milan S. Gender differences in the longitudinal impact of exposure to violence on mental health in urban youth. J Youth Adolescence. 2011;40:1674–1690. doi: 10.1007/s10964-011-9649-3. [DOI] [PubMed] [Google Scholar]
- 27.Drury SS, Mabile E, Brett ZH, et al. The association of telomere length with family violence and disruption. Pediatrics. 2014;134(1):e128–137. doi: 10.1542/peds.2013-3415. Epub 2014 Jun 16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Gershon A, Minor K, Hayward C. Gender, victimization and psychiatric outcomes. Psychol Med. 2008;38(10):1377–91. doi: 10.1017/S0033291708003000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Mellins CA, Havens JF, Kang E. A child psychiatry service for children and families affected by the HIV epidemic. IX International AIDS Conference 1993; Berlin, Germany. 1993. Abstract PO-B35-2343. [Google Scholar]
- 30.Aidala A, Havens J, Mellins CA, et al. Development and validation of the Client Diagnostic Questionnaire (CDQ): a mental health screening tool for use in HIV/AIDS service settings. Psychol Health Med. 2004;9:362–80. [Google Scholar]
- 31.Chesney MA, Ickovics JR, Chambers DB, et al. Self-reported adherence to antiretroviral medications among participants in HIV clinical trials: The AACTG adherence instruments. AIDS Care. 2000;12:255–66. doi: 10.1080/09540120050042891. [DOI] [PubMed] [Google Scholar]
- 32.Van Dyke RB, Lee S, Johnson GM, et al. Reported adherence as a determinant of response to highly active antiretroviral therapy in children who have human immunodeficiency virus infection. Pediatrics. 2002;109:1–7. doi: 10.1542/peds.109.4.e61. [DOI] [PubMed] [Google Scholar]
- 33.Usitalo A, Leister E, Tassiopoulos K. Relationship between viral load and self-report measures of medication adherence among youth with perinatal HIV infection. AIDS Care. 2014;26:107–15. doi: 10.1080/09540121.2013.802280. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Finkelhor D, Turner HA, Shattuck A, et al. Violence, crime, and abuse exposure in a national sample of children and youth: An update. JAMA Pediatrics. 2015 doi: 10.1001/jamapediatrics.2013.42. Published online June 29, 2015. [DOI] [PubMed] [Google Scholar]
- 35.Katz IT, Ryu A, Onuegbu AG, et al. Impact of HIV-related stigma on treatment adherence: systematic review and meta-synthesis. J Int AIDS Soc. 2013;16:18640. doi: 10.7448/IAS.16.3.18640. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Sharkey PT, Tirado-Strayer N, Papchristos AV, et al. The effect of local violence on children's attention and impulse control. AJPH. 2012;102:2287–93. doi: 10.2105/AJPH.2012.300789. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Suglia SF, Stuadenmayer J, Cohen S, et al. Posttraumatic stress symptoms related to community violence and children's diurnal cortisol response in an urban community-dwelling sample. Int J Behav Med. 2010;17:43–50. doi: 10.1007/s12529-009-9044-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Kessler RC, McLaughlin KA, Green JG, et al. Childhood adversities and adult psychopathology in the WHO World Mental Health Surveys. Br J Psychiatry. 2010;197:378–385. doi: 10.1192/bjp.bp.110.080499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Green JG, McLaughlin KA, Berglund PA, et al. Childhood adversities and adult psychiatric disorders in the national comorbidity survey replication I: Associations with first onset of DSM-IV disorders. Arch Gen Psychiatry. 2010;67:113–123. doi: 10.1001/archgenpsychiatry.2009.186. [DOI] [PMC free article] [PubMed] [Google Scholar]