Abstract
Background
Adolescents and young adults with HIV (AYAWH) represent vulnerable populations, with increased risk of virologic failure, loss to follow-up, and death. Depression and substance use in AYAWH can lead to worse outcomes, yet this overlap is not well understood.
Methods
This cross-sectional study included adolescents (10–17 years) and young adults (18–24 years) with HIV in the Caribbean, Central and South America network for HIV epidemiology (CCASAnet). Participants were administered surveys to assess for depression, substance use, and antiretroviral therapy (ART) adherence. Risk factors for depression; alcohol, tobacco, and substance use; missing ART doses; viral suppression; and 1-year retention were assessed.
Results
Six hundred twenty-five AYAWH were included. Depression prevelance was 16%. Males (adjusted odds ratio [aOR], 0.26; 95% CI, 0.16–0.44) and younger youth (15-year-olds vs 18-year-olds: aOR, 0.61; 95% CI, 0.40–0.95) were less likely to have depression. Fifty-eight percent reported using alcohol, 28% reported tobacco use, 17% reported cannabis use, and 4% reported cocaine use. Forty-one percent missed 1 or more doses of ART in the past week. Forty percent had detectable viral loads at the time of survey completion. Those who acquired HIV perinatally were more likely to have an unsuppressed viral load (aOR, 2.4; 95% CI, 1.24–4.62; P = .009). Only 73% of participants were retained in care following the survey; there was no statistical association between retention and age, sex, education, probable route of HIV acquisition, depression, and needing intervention for substance use.
Conclusions
Substance use and depression were prevalent in AYAWH, as were missed doses of ART and detectable viral loads.
Keywords: adolescents, alcohol use, depression, HIV/AIDS, substance use
Substance use and depression were common in adolescents and young adults living with HIV, as were missed doses of ART and detectable viral loads. Understanding these dynamics locally can lead to better inventions to improve viral suppression and retention in care.
According to the Joint United Nations Programme on HIV/AIDS, ∼39.9 million people globally were living with HIV in 2023, with Latin America and the Caribbean accounting for 2.3 million and 340 000 people, respectively [1]. Adolescents and young adults, particularly girls and young women, were disproportionately affected by HIV [1].
Adolescents and young adults with HIV (AYAWH) experience intricate challenges in their physical, social, emotional development, and health care transitions, which complicate health care delivery. This population is particularly vulnerable within the context of HIV, as they are at an increased risk of virologic failure, loss to follow-up, and shorter life expectancy [2–5]. In a cohort of adolescents and young adults with perinatally aquired HIV infection, mortality was 7%, with a median age of death of 20.8 years, underscoring the need to address barriers to care in this vulnerable population [6].
Additionally, depression is a common issue among AYAWH, impacting treatment adherence. In a study of adolescents with HIV in Mozambique, anxiety, depression, and post-traumatic stress disorder (PTSD) were common, and two-thirds had antiretroviral therapy (ART) adherence <90% [7]. In China, high suicide rates were seen in persons with HIV aged 15–24 [8]. In another cohort of AYAWH with perinatally acquired HIV, 42% were diagnosed with a mental health condition, and yet just over half were referred for mental health services [9]. In 1 study, nearly a quarter of AYAWH screened positive for moderate depression [10].
Misuse of alcohol, tobacco, and other substances is also common in these groups. Substance use often overlaps with mental health challenges and can lead to worse HIV outcomes: poor ART adherence or unsuppressed viral loads [11–13]. In a Thai cohort of young adults, 14% met criteria for harmful alcohol use and 17% were smokers. Both harmful alcohol use and lifetime suicide attempts were associated with lack of HIV viral load suppression [14]. In another group of AYAWH in Uganda, 42% reported substance use in the preceding 12 months, with the most commonly used substances being alcohol, benzodiazepines, and marijuana [15].
Data from Latin America and the Caribbean on the impact of depression and substance use on ART adherence, viral suppression, and retention in care among AYAWH are scarce. To address this gap, we evaluated the prevalence of and risk factors associated with depression and use of alcohol, tobacco, and other substances in a cohort of AYAWH from Latin America and the Caribbean. Furthermore, we investigated the association with self-reported adherence to ART and viral suppression at the time of survey administration as well as retention in care during the subsequent year following survey administration.
METHODS
Study Population
This cross-sectional study included adolescents (10 to 17 years) and young adults (18 to 24 years) with HIV on ART for ≥6 months in the Caribbean, Central and South America network for HIV epidemiology (CCASAnet). CCASAnet (www.ccasanet.org) is a member of the International Epidemiology Databases to Evaluate AIDS and has previously been described [16]. In this study, 7 clinic sites from CCASAnet were included: Instituto Nacional de Infectologia Evandro Chagas, Rio de Janeiro, Brazil; Universidade Federal de São Paulo, São Paulo, Brazil; Universidade Federal de Minas Gerais, Belo Horizonte, Brazil; Fundación Arriarán, Santiago, Chile; Le Groupe Haitien d’Etude du Sarcome de Kaposi et des Infections Opportunistes (GHESKIO), Port-au-Prince, Haiti; Hospital Escuela, Tegucigalpa, Honduras; Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán, Mexico City, Mexico; and Instituto de Medicina Tropical Alexander von Humboldt, Universidad Peruana Cayetano Heredia, Lima, Peru.
Each site received approval from local institutional ethics review boards, and individual participant and/or parental informed consent was obtained. All study activities were approved by the Vanderbilt University Medical Center (VUMC) Institutional Review Board. Data from contributing sites were collected and managed using Research Electronic Data Capture (REDCap), hosted at VUMC [17, 18].
Measures
Three surveys were administered to enrolled participants: the Patient Health Questionnaire (PHQ-2 and PHQ-9/PHQ-A) for depression [19, 20], the Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST-WHO) for substance use [21], and the Pediatric HIV/AIDS Cohort Study Adolescent Master Protocol (PHACS-AMP) for ART adherence [22]. Data collection took place from 2/13/2018 to 5/31/2019.
The PHQ-2 was administered to all enrolled participants, with the PHQ-9 used for adults and the PHQ-A used for those age <18 who screened positive on the PHQ-2. A score of ≥3 on the PHQ-2 suggests a likelihood of major depressive disorder, prompting the use of the PHQ-9 or PHQ-A for further assessment. The 9 items on the PHQ-9 correspond to the 9 diagnostic criteria for major depressive disorder in the Diagnosis and Statistical Manual of Mental Disorders (DSM-IV). The ASSIST-WHO survey assesses lifetime substance use of tobacco, alcohol, cannabis, cocaine, amphetamine-type stimulants, sedatives, hallucinogens, inhalants, opioids, and other drugs. Each response option on questions 2 to 7 of the ASSIST-WHO is assigned a numerical score. The ASSIST risk score is calculated by adding individual substance scores, categorized as “lower,” “moderate,” or “high.” A score of ≤3 is considered lower risk and requires no intervention. A score ranging from 4 to 26 is classified as moderate risk and calls for a brief intervention. A score of ≥27 indicates high risk of dependence, necessitating intensive intervention. The ASSIST survey has been validated only in adults, so a sensitivity analysis was conducted that only included young adults. The Pediatric HIV/AIDS Cohort Study (PHACS) Adolescent Master Protocol (AMP; PHACS-AMP) questionnaire asked the number of antiretroviral medication doses missed over the previous 7 days [23].
Sociodemographic characteristics were collected by each investigation site. We included age, sex, years of education, country of origin, probable route of HIV acquisition, and CD4+ cell count within 6 months of the survey. Education levels were defined as primary (years 1–6 in school), lower secondary (years 7–8 in school), and upper secondary (years 9–12 in school). Viral suppression was defined as having a viral load <200 copies/mL within +/− 7 days of survey administration. Participants were considered retained in care if they had 2 or more clinic visits separated by >90 days during the 365 days following survey administration.
Statistical Analysis
Descriptive statistics were presented using frequencies (proportions) for categorical variables and medians (interquartile ranges [IQRs]) for continuous variables. P values comparing adolescents and young adults were calculated using chi-square statistics for categorical variables and Wilcoxon rank-sum tests for continuous variables.
The association between depression and sociodemographic characteristics was estimated using ordered logistic regression models after categorizing scores based on severity cut-points (minimal [0–4], mild [5–9], moderate [10–14], moderately severe [15–19], and severe [20–27]; primary) and keeping the raw PHQ scores as a continuous variable (secondary). Secondary analyses were also performed categorizing PHQ scores as a binary variable (no depression [0–4], depression [5–27]) and fitting a logistic regression model. All models were controlled for age, years since ART initiation, sex at birth, probable route of HIV acquisition (perinatal vs nonperinatal), study site, and years of education.
Intervention scores (none, brief, and intensive intervention) derived from the ASSIST substance use risk scores were analyzed using ordered logistic regression. We performed a secondary analysis with the risk categories dichotomized to identify the need for intervention scores (intensive intervention vs other) and fit binary logistic regression models. Risk scores for alcohol, tobacco, and a combined score of all substances using the highest score obtained from any substance used were analyzed. Because some adolescents in the study reported substance use, binary logistic regression predicting the odds of substance use was also conducted. Ordered logistic regression models were used to examine associations between patient characteristics and the self-reported number of ART doses missed in the past 7 days. These models controlled for the aforementioned covariates, along with depression and whether an intervention was needed for alcohol, tobacco, or substance use.
Logistic regression models were used to examine associations between patient characteristics and unsuppressed viral load and clinical retention; the same covariates were included in these models, as well as an indicator that the participant self-reported missing ≥1 dose of ART in the past 7 days. Individuals with missing data were excluded from analyses, as the rates of missingness were low (<1% of participants were missing years since ART, education, and viral load data; ∼6% were missing ART adherence data). To relax linearity assumptions, the covariates age and years since ART initiation were expanded in all regression models using restricted cubic splines with 3 knots [24]. Reference values presented in the results (18 years of age and 12 years on ART) were chosen as values close to the median and of scientific interest; comparator values (15 and 21 years of age and 6 and 18 years for time on ART) were chosen to be equidistant from the reference and were supported by the data (eg, 15 and 21 represent the approxiate median ages among study adolescents and adults, respectively).
RESULTS
Characteristics of Study Population
A total of 249 adolescents (aged 10–17 years) and 376 young adults (aged 18–24 years) were included (n = 625); 321 (51%) were female, and the median age (range) was 18.9 (10.1–24.9) years. Most participants were from Brazil (32%), Haiti (32%), and Honduras (26%). Education level at the time of the survey was predominantly secondary, and 78% acquired HIV via perinatal transmission. Participants had been on ART for a median (IQR) of 12 (6–16) years. The median CD4 count within 6 months of the survey (IQR) was 618 (380–864) (Table 1).
Table 1.
Demographic and Clinical Characteristics of Adolescents and Young Adults From CCASAnet Sites According to Age Group
| Adolescent | Young Adult | Overall | |||
|---|---|---|---|---|---|
| No. | (n = 249), No. (%) | (n = 376), No. (%) | (n = 625), No. (%) | P Value | |
| Country | <.0001 | ||||
| Brazil | 51/249 (21) | 151/376 (40) | 202/625 (32) | ||
| Chile | 0/249 (0) | 11/376 (3) | 11/623 (2) | ||
| Honduras | 82/249 (33) | 80/376 (21) | 162/625 (26) | ||
| Haiti | 109/249 (44) | 91/376 (24) | 200/625 (32) | ||
| Mexico | 0/249 (0) | 30/376 (8) | 30/625 (5) | ||
| Peru | 7/249 (3) | 13/376 (4) | 20/625 (3) | ||
| Median age (IQR), y | 625 | 14.9 (13.3–16.6) | 20.9 (19.3–22.7) | 18.9 (16.0–21.5) | <.0001 |
| Median age at enrollment (IQR), y | 625 | 3.1 (1.5–6.5) | 7.8 (2.8–17.1) | 5.3 (2.0–12.2) | <.0001 |
| Sex: Female | 625 | 132/249 (53) | 189/376 (50) | 321/625 (51) | .50 |
| Detectable viral load | 622 | 102/248 (41.1) | 148/374 (39.57) | 250/622 (40.2) | .70 |
| Education level | 622 | <.0001 | |||
| None or primary | 109/249 (44) | 32/373 (9) | 141/622 (23) | ||
| Lower secondary | 73/249 (29) | 51/373 (14) | 124/620 (20) | ||
| Upper secondary | 55/249 (22) | 201/373 (54) | 256/622 (41) | ||
| University or postgraduate | 12/249 (5) | 89/373 (24) | 101/622 (16) | ||
| Perinatal route of HIV acquisition | 625 | 224/249 (89.9) | 261/376 (69.4) | 485/625 (77.6) | <.0001 |
| Missed medication in last wk | 612 | 100/244 (41) | 152/368 (41) | 252/612 (41.2) | .94 |
| Median years since ART initiation (IQR) | 612 | 10.0 (7–13) | 14 (6–18) | 12 (6–16) | <.0001 |
| Median CD4 count within 6 mo of survey (IQR), cells/mm3 | 260 | 728 (498–1013) | 543 (354–817) | 618 (380–864) | .0013 |
| Median ART doses missed in last wk (each medication; IQR) | 588 | 0 (0–1) 0–14 (range) |
0 (0–1) 0–35 (range) |
0 (0–1) 0–35 (range) |
.23 |
P values comparing adolescents and adults were calculated using Pearson chi-square statistics for categorical variables or Wilcoxon rank-sum tests for continuous variables.
Abbreviations: ART, antiretroviral therapy; CCASAnet, Caribbean, Central and South America network for HIV epidemiology; IQR, interquartile range.
Depression Screening
The prevalence of depression was 16% (101/625); 29 (5%) participants had mild depression, 38 (6%) had moderate depression, 22 (4%) had moderately severe depression, and 12 (2%) had severe depression. Young adults were more likely than adolescents to have all levels of depression (P = .002), and all those who had severe depression were young adults (12/625) (Table 2). Figure 1 shows associations between patient characteristics and the odds of having more severe depression after adjusting for study site and other variables. (Unadjusted results for this and all other statistical analyses are reported in the Supplementary Data.) Participants with older ages had greater odds of having a higher level of depression (21-year-olds vs 18-year-olds [reference]: adjusted odds ratio [aOR], 1.45; 95% CI, 1.10–1.92; P = .013). Males (aOR, 0.26; 95% CI, 0.16–0.44; P < .0001) were more likely to score in a lower depression category. Participants who had perinatally acquired HIV tended to have slightly lower levels of depression (aOR, 0.57; 95% CI, 0.27–1.20; P = .14), although this was not statistically significant. Results were similar when PHQ-9 scores were kept continuous or dichotomized as depression yes/no; age and gender were predictive in both multivariable models (data not shown).
Table 2.
PHQ-9/A Scores From CCASAnet Sites According to Age Group
| No. | Adolescent (n = 249), No. (%) | Adult (n = 376), No. (%) | Overall (n = 625), No. (%) | P | |
|---|---|---|---|---|---|
| Total PHQ-9 score | 625 | .0001 | |||
| Median (IQR) | 0 (0–1) | 1 (0–2) | 0 (0–2) | ||
| Range | 0–19 | 0–26 | 0–26 | ||
| Depression cut-points from PHQ-9 | 625 | .002 | |||
| Minimal | 225/249 (90.36) | 299/376 (79.52) | 524/625 (83.84) | ||
| Mild | 9/249 (3.61) | 20/376 (5.32) | 29/625 (4.64) | ||
| Moderate | 10/249 (4.02) | 28/376 (7.45) | 38/625 (6.08) | ||
| Moderately severe | 5/249 (2.01) | 17/376 (4.52) | 22/625 (3.52) | ||
| Severe | 0/249 (0.00) | 12/376 (3.19) | 12/625 (1.92) |
P values comparing PHQ-9 scores were calculated using Pearson chi-square statistics for categorical variables or Wilcoxon rank-sum tests for continuous variables.
Abbreviations: CCASAnet, Caribbean, Central and South America network for HIV epidemiology; IQR, interquartile range; PHQ-9, Patient Health Questionnaire.
Figure 1.
Adjusted odds ratios for being in higher depression categories (minimal, mild, moderate, moderately severe, and severe) based on patient characteristics. The ordinal logistic regression model also adjusted for study site. Abbreviation: ART, antiretroviral therapy.
Substance Use Screening
Three hundred sixty participants (58%) reported using alcohol at any time, with 240 (38%) reporting use in the previous 3 months. One hundred seventy-six (28%) reported any tobacco use, with 104 (17%) reporting tobacco use in the previous 3 months; 105 (17%) reported any cannabis use, with 50 (8%) in the previous 3 months; and 27 (4%) reported any cocaine use, with 5 (1%) in the previous 3 months. The need for intervention based on ASSIST-WHO results is shown in Table 3.
Table 3.
Type of Intervention Needed for Tobacco, Alcohol, or Substance Use Based on ASSIST-WHO
| No. | Adolescent (n = 249), No. (%) | Adult (n = 376), No. (%) | Overall (n = 625), No. (%) | P | |
|---|---|---|---|---|---|
| Tobacco | 625 | <.001 | |||
| No intervention | 233/249 (93.57) | 292/376 (77.66) | 525/625 (84.00) | ||
| Brief intervention | 15/249 (6.02) | 77/376 (20.48) | 92/625 (14.72) | ||
| More intensive intervention | 1/249 (0.40) | 7/376 (1.86) | 8/625 (1.28) | ||
| Alcohol | 625 | <.001 | |||
| No intervention | 236/249 (94.78) | 319/376 (84.84) | 555/625 (88.80) | ||
| Brief intervention | 11/249 (4.42) | 49/376 (13.03) | 60/625 (9.60) | ||
| More intensive intervention | 2/249 (0.80) | 8/376 (2.13) | 10/625 (1.60) | ||
| Substance use (all types combined) | 625 | .012 | |||
| No intervention | 229/249 (91.97) | 229/249 (91.97) | 544/625 (87.04) | ||
| Brief intervention | 18/249 (7.23) | 55/376 (14.63) | 73/625 (11.68) | ||
| More intensive intervention | 2/249 (0.80) | 6/376 (1.60) | 8/625 (1.28) |
P values comparing tobacco, alcohol, and substance use scores were calculated using Pearson chi-square statistics.
Abbreviations: ASSIST-WHO, Alcohol, Smoking, and Substance Involvement Screening Test–World Health Organization; PHQ-9, Patient Health Questionnaire.
The associations between needing interventions and baseline demographics are shown in Figure 2.
Figure 2.
Adjusted odds ratios for need of higher intervention (none, brief, or intensive) to treat for (A) alcohol, (B) tobacco, and (C) substance use based on the ASSIST-WHO survey. The ordinal logistic regression model also adjusted for study site. Abbreviations: ART, antiretroviral therapy; ASSIST-WHO, Alcohol, Smoking, and Substance Involvement Screening Test–World Health Organization.
Alcohol
Younger participants were less likely to need intervention for alcohol (15-year-olds vs 18-year-olds: aOR, 0.32; 95% CI, 0.17–0.60; P = .002). The effect of age was no longer significant when the analysis was limited to young adults only (P = .81; data not shown).
Tobacco
Participants who were older (21-year-olds vs 18-year-olds: aOR, 1.52; 95% CI, 1.13–2.05; P = .008) and male (aOR, 1.91; 95% CI, 1.16−3.16; P = .01) were more likely to need intervention for tobacco use. Male sex was associated with higher odds of needing a more intensive intervention for tobacco when the analysis was limited to young adults only (data not shown).
Substance Use
In the multivariable analysis, there were no significant associations between patient characteristics and the need for more intervention for substance use. In analyses limited to young adults, older age was associated with a slightly lower risk of needing intervention (24 years vs 21 years: aOR, 0.40; 95% CI, 0.17–0.96; P = .05, data not shown).
Self-reported Adherence to Antiretroviral Therapy
Though 41.2% of participants reported missing ≥1 dose of medication in the past week, the median number of ART doses missed (IQR, range) was 0 (0–1, 0–35). Figure 3A shows the association between patient characteristics and the odds of missing more ART doses in the past 7 days. Those who were determined to need intervention for tobacco use were more likely to report missing ART doses (intensive intervention vs no intervention: aOR, 6.3; 95% CI, 1.48–27.2; P = .03). Need for alcohol or substance use interventions was not statistically associated with missing ART doses, although confidence intervals were wide, suggesting that the study may have been underpowered to detect an association. Similarly, there was no statistical association detected between age, sex, route of transmission, or depression and adherence to ART. Results were similar in a sensitivity analysis that dichotomized the outcome as ≥1 vs 0 doses missed in the past 7 days (Supplementary Figure 1).
Figure 3.
Adjusted odds ratios with 95% CIs for (A) self-reported number of ART doses missed in the last 7 days, (B) unsuppressed viral load, and (C) 1-year retention. Logistic regression models were also adjusted for study site. Abbreviation: ART, antiretroviral therapy.
Unsuppressed Viral Load
A total of 250 (40%) participants—41% of adolescents and 40% of young adults—had a detectable viral load at survey completion. Figure 3B shows the association between unsuppressed viral load and participant characteristics. Those who acquired HIV perinatally were more likely to have a detectable viral load (aOR, 2.40; 95% CI, 1.24–4.62; P = .009). Those who reported missing ≥1 ART dose in the past week had higher odds of having a detectable viral load (aOR, 2.33; 95% CI, 1.58–3.44; P < .001). There was no statistical association between having an unsuppressed viral load and age, years since ART initiation, gender, education, depression, or needing intervention for alcohol, tobacco, or other substance use (P > .05 for all). Results were similar when alcohol, tobacco, and other substance use variables were included in the model as continuous variables.
One-Year Retention
During the year after survey administration, 457 (73%) participants were retained in care, 77% of adolescents and 70% of young adults. Figure 3C shows the association between clinical retention and participant characteristics. Participants who had been on ART for a shorter length of time at survey administration were more likely to be retained (6 vs 12 years on ART: aOR, 1.79; 95% CI, 1.20–2.66; P = .015). There was no statistical association between retention and age, sex at birth, education, probable route of HIV acquisition, depression, or needing intervention for alcohol, tobacco, or substance use (P > .05 for all). Results were similar when alcohol, tobacco, and other substance use variables were included in the model as continuous variables. Results were also similar when adjusted for whether the 1-year follow-up period overlapped with the coronavirus disease 2019 pandemic (ie, the participant was surveyed on or after March 1, 2019, which was the case for 34% of our participants).
DISCUSSION
Our study of 625 AYAWH in Latin America and the Caribbean revealed several notable findings. The prevalence of depression was 16%, with higher rates oberved among females and older young adults, particularly those with non–perinatally acquired HIV. This pattern aligns with a systematic review of >2600 AYAWH [25], which notably included no studies from Latin America or the Caribbean region, making our study an important contributor to the literature on this vulnerable population. While our depression rate (16%) was slightly lower than that found in the systematic review (26%), females were still more affected. Factors such as adverse experiences in childhood, the sociocultural roles of adverse events, sexual abuse, and variations in coping skills may explain the gender differences [25]. This explanation could be true in our region as well. Female respondents and those with depression have been more likely to report HIV-related stigma [26], and Latin America is known to have large regions of inequity, particularly among women, children, and adolescents [27].
Substance use was common in this population, with more than half reporting alcohol use and over a quarter reporting tobacco use. Missed doses of medication were reported by 41% of participants, and those who needed intervention for tobacco use were more likely to miss ART doses. The association between tobacco and adherence still held in analyses adjusted for education, clinical site, and HIV acquisition route. Of note, however, only 8 (1.3%) participants needed the more intensive tobacco intervention, so these results are driven by a very small number of participants, which is seen in the very wide confidence intervals for this association. In a previous study in Brazil among people with HIV aged 12–65, prevalence of tobacco use was 45% and alcohol use was 49.5%, and these individuals reported worse self-rated health when compared with those without HIV [28]. Other data on 1221 children with perinatally acquired HIV showed that these children had lower self-esteem and lower future orientation [29, 30], which has been shown to lead to increased risk for substance use [31]. This supports the notion of a complex interplay between substance use, depression, and health outcomes in AYAWH.
A total of 40% of participants had detectable viral loads at the time of the survey; those who acquired HIV perinatally were more likely to have unsuppressed viral loads. Further, only 73% of participants were retained in care during the year following administration of the survey, which falls significantly short of the World Health Organization goals for retention. There was no statistical association between retention and age, gender, education, probable route of HIV acquisition, depression, or needing treatment for alcohol, tobacco, or substance use, though there may have been reduced power to detect true associations. Our findings are similar to other regional studies; in 1 study on the HIV care cascade in Brazil, children and adolescents aged 5 to 17 years had rates of viral suppression <60% [32]. In a study of >35 000 adults with HIV and 2601 children with HIV across the IeDEA region (including CCASAnet), viral suppression was less likely among children compared with adults [33]. Further study is needed to examine the continuum of care outcomes, particularly retention in care and viral suppression, among AYAWH, recognizing that barriers may differ for those with perinatal vs nonperinatal acquisition.
Our study reveals several opportunities to improve outcomes for AYAWH. First, addressing depression and substance use early is crucial in providing a multifaceted approach to caring for these vulnerable populations. Screening for depression and substance use to identify the mental health needs of AYAWH may not be consistently performed in routine clinical practice, leading to missed opportunities for linkage to appropriate care and tailored services. Studies that investigated depression and the possible associated factors show a great deal of variability in the prevalence of the depressive symptoms, in part because mental health problems are often assessed together as “psychiatric problems,” “mood disorders,” “emotional/behavioral problems,” or “mental health diagnoses.” More specific screening in AYAWH, for depression, anxiety, internalized stigma, post-traumatic stress disorder, and other conditions, could facilitate targeted interventions with greater likelihood of success. Until we more clearly understand the complex role mental health plays in this population (and its interplay with environment and genetics), it will be challenging to improve outcomes.
Integrating substance use screening into routine care for AYAWH is essential. Such integration should include questions during care visits on specific substance use patterns, substance of choice, and frequency of use in order to structure tailored interventions that have the greatest chance of success. With increased data availability on behavioral interventions and medication-assisted treatments for substance use disorders, we need to first understand when and where to deploy these efficacious tools.
In our cohort, 41.3% of the participants missed antiretroviral doses in the preceding week, highlighting the need to further understand the care continuum in AYAWH. Factors such as medication fatigue, depression, substance use, access to medication, and environmental and social chaos could all play a role. If so, AYAWH may benefit greatly from long-acting injectable antiretroviral therapy. Understanding these dynamics locally can lead to better inventions to improve viral suppression and retention in care.
Our study had several limitations. Our data were cross-sectional, and hence, we were unable to determine causation. Key data on substance use and adherence were self-reported and may reflect reporting biases. We did not have denominator information for self-reported numbers of doses missed over the past 7 days. In addition, there may also have been reduced power to detect true associations. That said, our study reflected several strengths. It is one of the first to describe depression, substance use, and adherence among AYAWH in Latin America, including those with and without perinatally acquired HIV. Analyses were stratified for each outcome of interest by study site to limit the influence of intersite differences on our inferences. Our results, using data from a large cohort from multiple countries, emphasize the need to monitor and assess substance use and mental health problems in this population, with special attention to women and those in late adolescence or early adulthood. Programs focused on early recognition and treatment of depression and substance use should be broadly available and accessible, and public policies should be implemented that are directed at improving ART adherence and better comprehensive care for ALWH in Latin America.
CONCLUSIONS
In conclusion, we present critical data on AYAWH in Latin America and the Caribbean. Substance use and depression were common in this population, as were missed doses of ART and detectable viral loads. More studies are needed to better understand the complex interactions between mental health, substance use, and the continuum of care for adolescents and young adults with HIV.
Supplementary Material
Acknowledgments
The authors would like to thank all participants for their contributions to this study.
Author contributions. D.M.M., S.D., B.E.S., and C.C.M. conceived of the study and helped with manuscript writing and data analysis. A.K. and B.S. provided statistical analyses. R.C.M.S., V.R., B.C.R., M.T.L., F.M., F.R., J.P., F.M., S.W.C., and A.K.P. all helped with manuscript writing and data analysis.
Data availability. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to international data privacy regulations and ethical restrictions.
Financial support. This work was supported by the National Institutes of Health (NIH). The Caribbean, Central and South America network for HIV epidemiology (CCASAnet) (U01AI069923), a member cohort of the International Epidemiologic Databases to Evaluate AIDS (leDEA). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. This award is funded by the following institutes: the National Institute of Allergy and Infectious Diseases (NIAID), the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), the National Cancer Institute (NCI), the National Institute of Mental Health (NIMH), the National Institute on Drug Abuse (NIDA), the National Heart, Lung, and Blood Institute (NHLBI), the National Institute on Alcohol Abuse and Alcoholism (NIAAA), the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), the Fogarty International Center (FIC), and the National Library of Medicine (NLM).
Contributor Information
Daisy Maria Machado, Department of Pediatrics, Escola Paulista de Medicina, Universidade Federal de São Paulo, São Paulo, Brazil.
Stephany N Duda, Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee, USA.
Regina Célia de Menezes Succi, Department of Pediatrics, Escola Paulista de Medicina, Universidade Federal de São Paulo, São Paulo, Brazil.
Ahra Kim, Department of Biostatistics, Vanderbilt University School of Medicine, Nashville, Tennessee, USA.
Paridhi Ranadive, Department of Biostatistics, Vanderbilt University School of Medicine, Nashville, Tennessee, USA.
Vanessa Rouzier, Haitian Group for the Study of Kaposi's Sarcoma and Opportunistic Infections (GHESKIO), Port-au-Prince, Haiti; Weill Cornell Medicine, New York, New York, USA.
Brenda Crabtree-Ramírez, Departamento de Infectología, Instituto Nacional de Ciencias Médicas y Nutrición, Salvador Zubirán, México City, México.
Marco T Luque, Department of Pediatrics, Instituto Hondureño de Seguridad Social and Hospital Escuela Universitario, Tegucigalpa, Honduras.
Fernando Mejia, Instituto de Medicina Tropical Alexander von Humboldt, Universidad Peruana Cayetano Heredia, Lima, Peru.
Fernanda Rodríguez, Fundación Arriarán, Santiago, Chile.
Jorge Pinto, School of Medicine, Federal University of Minas Gerais, Belo Horizonte, Brazil.
Sandra Wagner Cardoso, SDT/AIDS Clinical Research Laboratory, Instituto de Pesquisa Clínica Evandro Chagas-Fundação Oswaldo Cruz, Rio de Janeiro, Brazil.
Fernanda Maruri, Division of Infectious Diseases, Vanderbilt University School of Medicine, Nashville, Tennessee, USA.
Bryan E Shepherd, Department of Biostatistics, Vanderbilt University School of Medicine, Nashville, Tennessee, USA.
Catherine C McGowan, Division of Infectious Diseases, Vanderbilt University School of Medicine, Nashville, Tennessee, USA.
Anna K Person, Division of Infectious Diseases, Vanderbilt University School of Medicine, Nashville, Tennessee, USA.
Supplementary Data
Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
References
- 1. UNAIDS . Global HIV & AIDS statistics—fact sheet. Available at: https://www.unaids.org/en/resources/fact-sheet. Accessed September 4, 2024.
- 2. Neilan AM, Ufio OL, Brenner IR, et al. Projected life expectancy for adolescents with HIV in the US. JAMA Health Forum 2024; 5:e240816. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Lamb MR, Fayorsey R, Nuwagaba-Biribonwoha H, et al. High attrition before and after ART initiation among youth (15–24 years of age) enrolled in HIV care. AIDS 2014; 28:559–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Mbébi EPJ, Penda CI, Ngondi G, et al. High risk of virologic failure among HIV-infected children and adolescents routinely followed-up in littoral region of Cameroon. PLoS One 2023; 18:e0289426. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Somerville K, Jenkins CA, Carlucci JG, et al. Outcomes after second-line antiretroviral therapy in children living with HIV in Latin America. J Acquir Immune Defic Syndr 2021; 87:993–1001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Ounchanum P, Aurpibul L, Teeraananchai S, et al. High mortality in adolescents and young adults with perinatally-acquired HIV in Thailand during the transition to adulthood. AIDS Care 2024; 36:964–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Nguyen N, Lovero KL, Falcao J, et al. Mental health and ART adherence among adolescents living with HIV in Mozambique. AIDS Care 2023; 35:182–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Zhang H, Li Z, Li P, et al. Life expectancy and mental related burden of disease among people living with HIV/AIDS. AIDS Care 2024; 36:15–23. [DOI] [PubMed] [Google Scholar]
- 9. Vázquez-Pérez Á, Velo C, Escosa L, et al. Mental health in children, adolescents, and youths living with perinatally acquired HIV: at the crossroads of psychosocial determinants of health. Children (Basel) 2023; 10:405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Sohn AH, Singtoroj T, Chokephaibulkit K, et al. Long-term post-transition outcomes of adolescents and young adults living with perinatally and non-perinatally acquired HIV in Southeast Asia. J Adolesc Health 2023; 72:471–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Olashore AA, Chiliza B, Paruk S. Antiretroviral therapy non-adherence and its relationship with cognitive impairment, alcohol use disorder, and depression in adolescents living with HIV. BMC Psychiatry 2023; 23:532. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Morawej Z, Nyundo A, Kinyaga A, et al. Prevalence and factors associated with substance use among HIV positive youth attending HIV care and treatment centers in Dodoma, Tanzania. AIDS Res Ther 2022; 19:65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Haas AD, Technau KG, Pahad S, et al. Mental health, substance use and viral suppression in adolescents receiving ART at a paediatric HIV clinic in South Africa. J Int AIDS Soc 2020; 23:e25644. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Aurpibul L, Kosalaraksa P, Kawichai S, et al. Alcohol use, suicidality and virologic non-suppression among young adults with perinatally acquired HIV in Thailand: a cross-sectional study. J Int AIDS Soc 2023; 26:e26064. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Brathwaite R, Mutumba M, Nannono S, Ssewamala FM, Filiatreau LM, Namatovu P. Prevalence and correlates of substance use among youth living with HIV in fishing communities in Uganda. AIDS Behav 2024; 28:2414–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. McGowan CC, Cahn P, Gotuzzo E, et al. Cohort profile: Caribbean, Central and South America network for HIV research (CCASAnet) collaboration within the International epidemiologic Databases to Evaluate AIDS (IeDEA) programme. Int J Epidemiol 2007; 36:969–76. [DOI] [PubMed] [Google Scholar]
- 17. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research Electronic Data Capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009; 42:377–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Harris PA, Taylor R, Minor BL, et al. The REDCap Consortium: building an international community of software platform partners. J Biomed Inform 2019; 95:103208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Kroenke K, Spitzer RL, Williams JBW. The Patient Health Questionnaire-2: validity of a two-item depression screener. Med Care 2003; 41:1284–92. [DOI] [PubMed] [Google Scholar]
- 20. Kroenke K, Spitzer RL, Williams JBW. The PHQ-9. J Gen Intern Med 2001; 16:606–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. World Health Organization . The Alcohol, Smoking and Substance Involvement Screening Test (ASSIST). Available at: https://www.who.int/publications/i/item/978924159938-2. Accessed September 4, 2024.
- 22. Usitalo A, Leister E, Tassiopoulos K, et al. Relationship between viral load and self-report measures of medication adherence among youth with perinatal HIV infection. AIDS Care 2014; 26:107–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Zhang ZJ, Patel K, Kovac A, Jacobson D, Tassiopoulos K, Williams P. Pediatric HIV/AIDS Cohort Study (PHACS) Adolescent Master Protocol (AMP / PH200) and AMP Up (PH300) annual administrative report. Available at: https://phacsstudy.org/cms_uploads/Latest%20Documents/AMP_AAR_APR_2021_web_updated.pdf. Accessed May 21, 2025.
- 24. Shepherd BE, Rebeiro PF. Caribbean, Central and South America network for HIV epidemiology. Brief report: assessing and interpreting the association between continuous covariates and outcomes in observational studies of HIV using splines. J Acquir Immune Defic Syndr 2017; 74:e60–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Ayano G, Demelash S, Abraha M, Tsegay L. The prevalence of depression among adolescent with HIV/AIDS: a systematic review and meta-analysis. AIDS Res Ther 2021; 18:23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Dessie ZG, Zewotir T. HIV-related stigma and associated factors: a systematic review and meta-analysis. Front Public Health 2024; 12:1356430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Sanhueza A, Carvajal L, Cueva DA, et al. The Every Woman Every Child Initiative: supporting countries in Latin America and the Caribbean to reduce social inequalities in health. Int J Equity Health 2022; 21:83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. De Boni RB, de Vasconcellos MTL, Luis NSP, et al. Substance use, self-rated health and HIV status in Brazil. AIDS Care 2021; 33:1358–62. [DOI] [PubMed] [Google Scholar]
- 29. Gamarel KE, Finer Z, Resnicow K, et al. Associations between internalized HIV stigma and tobacco smoking among adolescents and young adults living with HIV: the moderating role of future orientations. AIDS Behav 2020; 24:165–72. [DOI] [PubMed] [Google Scholar]
- 30. Lin X, Fang X, Chi P, Heath MA, Li X, Chen W. Social ecological factors associated with future orientation of children affected by parental HIV infection and AIDS. J Health Psychol 2016; 21:1404–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Earnshaw VA, Eaton LA, Collier ZK, et al. HIV stigma, depressive symptoms, and substance use. AIDS Patient Care STDS 2020; 34:275–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Ferreira AACM, Pinho RGG, de Aquino LM, et al. Disparities in HIV continuum of care in the paediatric population: a real-life study in Brazil. HIV Med 2023; 24:411–21. [DOI] [PubMed] [Google Scholar]
- 33. Jiamsakul A, Kariminia A, Althoff KN, et al. HIV viral load suppression in adults and children receiving antiretroviral therapy—results from the IeDEA collaboration. J Acquir Immune Defic Syndr 2017; 76:319–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.