To the Editor:
Although early intervention with antiretrovirals (ARTs) can improve clinical outcomes and reduce transmission risk in youth with HIV (YWH),1 adequate adherence is essential to sustained, meaningful viral load reduction. Nonadherence is common in YWH, particularly in North America, where adolescents have the lowest average adherence rates globally (53%; 95% confidence interval [CI] 46–59%).2 Understanding treatment readiness, or the willingness to take health-promoting actions, may provide insight into nonadherence. Within the context of HIV care, higher readiness has been linked to medication adherence and viral suppression.3
Analyzing individual-level factors that affect treatment readiness, including executive functioning (e.g., planning, problem-solving, decision-making), is important for understanding behavioral change; however, the relationship between executive functioning and treatment compliance, particularly within YWH, is unclear. While it is well-known that HIV infection can lead to cognitive impairment4 and impact treatment compliance, a study of children and adolescents with perinatally-acquired HIV found no relationship between executive functioning and treatment adherence,5 a population which may be manifestly different from those with behaviorally acquired HIV. Conversely, in a study of youth (ages 18–24) with behaviorally acquired HIV, 64.7% had HIV-associated neurocognitive disorders,6 suggesting that YWH may experience problems with executive functioning.
Behavioral factors, including substance use and mental health problems, may impact executive functioning and influence treatment readiness and adherence in YWH. Compared with their otherwise healthy peers,7 YWH have higher rates of mental health problems and substance use, particularly marijuana and alcohol. Substance use may impact clinical outcomes by reducing decision-making capacity and engagement in health-seeking behaviors,6 while mental health problems, like anxiety, depression, and traumatic stress, may lead to poorer clinical outcomes.7
This study used structural equation modeling (SEM) and bivariate correlations to test the association between treatment readiness and substance use, mental health problems, and executive functioning. We hypothesized that substance use and mental health problems would negatively impact treatment readiness, a relationship that is mediated through decreased executive functioning. This hypothesis is supported if our model has a good fit, demonstrates that higher scores on behavioral health measures are associated with lower levels of treatment readiness, and that a reduction in readiness is mediated through decreased executive functioning.
This study used baseline data from the Motivational Enhancement System for Adherence (MESA) for Youth Starting ART study (Clinical Trial ID: NCT02761746), a multisite, randomized controlled trial conducted between Spring 2016 and Fall 2021 aimed at improving ART adherence in YWH. The MESA study was approved by the Institutional Review Board (IRB; 052115B3F) at Wayne State University and participating clinical sites. A full overview of the study, including inclusion and exclusion criteria, can be found by clicking on the clinical trial link. Recruitment occurred at seven US clinical sites (Wayne State University, Children’s Hospital Los Angeles, Children’s National Medical Center, Children’s Hospital of Philadelphia, University of Colorado, University of Miami, and Johns Hopkins University).
This study used four self-report questionnaires: the Behavior Rating Inventory of Executive Functioning-Adult Version (BRIEF-A), the Brief Symptom Inventory-18 (BSI-18), the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST), and the Readiness Ruler Importance Score (RRIS). We used three of the nine BRIEF-A subscales (Working Memory, Inhibit, and Plan/Organize) relevant to medication adherence and behavioral regulation to assess executive functioning (Cronbach’s α = 0.73–0.85).8 The BSI-18 has three subscales (Somatization, Depression, and Anxiety; Cronbach’s α = 0.71–0.85).9 The overall score is called the Global Severity Index. The ASSIST has eight items and assesses problematic substance use patterns across classes of substances. Given low prevalences of other substance use, we only reported outcomes for alcohol and marijuana subscales. The ASSIST has good construct validity and moderate to substantial test–retest kappa coefficients of agreement across the different classes of substances examined.10 The RRIS is a visual analogue scale, ranging from 0 (not at all important) to 100 (totally important).11
The average age of the 155 participants was 20.19 (SD = 2.18); 87.74% reported being biologically male at birth (87.74%). 83.23% of participants reported their gender identity as cis-male, 11.61% identified as cis-female, 3.87% as transgender male, and 0.65% as transgender female.53.55% identified as gay or lesbian; 18.06% as straight or heterosexual, 22.57% as bisexual, and 5.81% as “other.” Racially, 76.77% identified as Black/African American, 13.55% as Mixed Race/Other, 1.29% as Asian/Pacific Islander or Native American/Alaskan Native, and 6.45% as White. With regard to ethnicity, 16.13% identified as being Hispanic/Latino.
In total, 43.23% of participants had a viral load count between 10,001 and 100,000; 21.29% had viral load counts of >100,000; and 3.23% were undetectable (counts <50). In total, 39.07% of participants had CD4 counts of ≥500, 30.46% had counts between 350 and 499, 19.21% had counts between 200 and 349, and 11.26% had counts of <200.
Statistical analyses were performed using JASP (version 0.16.1), and SEM models used lavaan syntax. We first ran bivariate correlations: 17 of 21 correlation pairs were significant at the 0.05 level (Table 1). SEM analysis partially supported the hypothesized model. The initial model revealed statistically significant relationships between the latent factors for mental health problems and executive functioning (p < 0.01; estimate = 0.19; standard [std.] error = 0.06; 95% CI: 0.08, 0.31) and between executive functioning and treatment readiness (p = 0.03; estimate = −5.29; std. error = 2.40; 95% CI: −9.99, −0.58). Substance use was not significant in the model (p = 0.10). Overall fit of the model to the data is considered good (i.e., χ2 = 15.95, df = 12 p = 0.19; root mean squared error of approximation (RMSEA) = 0.05 [90% CI 0.00, 0.10], standardized root mean squared residual (SRMR) = 0.03; comparative fit index (CFI) = 0.99; incremental fit index (IFI) = 0.99, goodness of fit index (GFI) = 1.00). After removing substance use variables (Fig. 1), the refined final model retained good fit (χ2 = 10.37, df = 5 p = 0.07; RMSEA = 0.08 [90% CI 0.00, 0.16], SRMR = 0.03; CFI = 0.99; IFI = 0.99, GFI = 1.00) and was more parsimonious than the first model (AIC of 1927.55 vs. 3544.66).
Table 1.
Correlation Matrix for Behavioral Health, Executive Functioning, and Treatment Readiness Variables
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | |
|---|---|---|---|---|---|---|---|
| Variable/n | Alcohol | Marijuana | Memory | Inhibit | Plan | Readiness | |
| Mental health | — | ||||||
| Pearson’s r | — | ||||||
| p value | — | ||||||
| Alcohol | 117 | — | |||||
| Pearson’s r | 0.27 | — | |||||
| p value | <0.01 | — | |||||
| Marijuana | 125 | 100 | — | ||||
| Pearson’s r | 0.20 | 0.28 | — | ||||
| p value | 0.02 | 0.01 | — | ||||
| Memory | 155 | 117 | 125 | — | |||
| Pearsons r | 0.47 | 0.21 | 0.28 | — | |||
| p value | <0.01 | 0.02 | <0.01 | — | |||
| Inhibit | 155 | 117 | 125 | 155 | — | ||
| Pearsons r | 0.48 | 0.26 | 0.20 | 0.79 | — | ||
| p value | <0.01 | 0.01 | 0.02 | <0.01 | — | ||
| Plan | 155 | 117 | 125 | 155 | 155 | — | |
| Pearson’s r | 0.46 | 0.21 | 0.26 | 0.81 | 0.73 | — | |
| p value | <0.01 | 0.02 | <0.01 | <0.01 | <0.01 | — | |
| Readiness | 155 | 117 | 125 | 155 | 155 | 155 | — |
| Pearson’s r | −0.16 | –0.11 | –0.06 | −0.18 | −0.05 | −0.21 | — |
| p value | 0.05 | 0.22 | 0.54 | 0.03 | 0.54 | 0.01 | — |
Correlations in bold are significant at a 0.05 threshold.
FIG. 1.
Final, revised path diagram.
Consistent with previous research, we observed a positive relationship between mental health problems and executive functioning. Higher levels of anxiety, depression, and other mental health symptomatology led to decreased ability to plan future behavior, problem-solve, and inhibit impulsivity. Likewise, higher executive functioning was associated with higher treatment readiness. Notably, we did not find a statistical relationship between substance use and executive functioning, perhaps due to participants’ overall reliance on marijuana and alcohol instead of more cognitively and socially disruptive substances.
These results highlight the need for interventions targeting treatment readiness and adherence and YWH mental well-being. Cognitive therapies have been effective to improve mental health, executive functioning, and problem-solving skills.12 While the role of substance use in the overall SEM model is limited, its importance in youth health, HIV transmission and development, and course of HIV, should not be understated. Clinicians should continue screening for substance use and initiate treatment where appropriate.
The small, demographically homogenous sample size restricts generalizability to the broader YWH population. The applicability of the findings to youth transitioning into adulthood and their relevance to other factors affecting treatment readiness remain unclear. We suggest replication with larger, more diverse samples, using objective measures alongside self-reports. Other factors affecting treatment readiness, like unexpected pregnancy or criminal justice involvement, should also be explored.
Author Disclosure Statement
No competing financial interests exist.
Funding Information
This research was supported by the National Institute of Mental Health (NIMH) award # 5R01MH108442.
References
- 1. Mateo‐Urdiales A, Johnson S, Smith R, et al. Rapid initiation of antiretroviral therapy for people living with HIV. Cochrane Database Syst Rev 2019;6(6):CD012962; doi: 10.1002/14651858.CD012962.pub2/full [cited 2022 Aug 2] Available from: https://www.cochranelibrary.com/cdsr/ [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Kim SH, Gerver SM, Fidler S, et al. Adherence to antiretroviral therapy in adolescents living with HIV: Systematic review and meta-analysis. AIDS 2014;28(13):1945–1956. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. MacDonell KE, Naar-King S, Murphy DA, et al. Predictors of medication adherence in high risk youth of color living with HIV. J Pediatr Psychol 2010;35(6):593–601. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Watkins CC, Treisman GJ. Cognitive impairment in patients with AIDS: Prevalence and severity. HIV 2015;7:35–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Garvie PA, Brummel SS, Allison SM, et al. Roles of medication responsibility, executive and adaptive functioning in adherence for children and adolescents with perinatally acquired HIV. The Pediatric Infectious Disease Journal 2017;36(8):751–757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Nichols SL, Bethel J, Garvie PA, et al. Neurocognitive functioning in antiretroviral therapy-naïve youth with behaviorally acquired human immunodeficiency virus. J Adolesc Health 2013;53(6):763–771. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Starks TJ, MacDonell KK, Pennar AL, et al. Drug use among adolescents and young adults with unsuppressed HIV who use alcohol: Identifying patterns of comorbid drug use and associations with mental health. AIDS Behav 2020;24(10):2975–2983. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Gioia GA, Isquith PK. Behavior Rating Inventory for Executive Functions. In: Kreutzer JS, DeLuca J, Caplan B., editors. Encyclopedia of Clinical Neuropsychology [Internet]. New York, NY: Springer; 2011. [cited 2022 Aug 2]. p. 372–6. doi: 10.1007/978-0-387-79948-3_1881 [DOI] [Google Scholar]
- 9. Franke GH, Jaeger S, Glaesmer H, et al. Psychometric analysis of the brief symptom inventory 18 (BSI-18) in a representative German sample. BMC Med Res Methodol 2017;17(1):14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Humeniuk R, Ali R, Babor TF, et al. Validation of the alcohol, smoking and substance involvement screening test (ASSIST). Addiction 2008;103(6):1039–1047. [DOI] [PubMed] [Google Scholar]
- 11. Stott N, Rollnick S, Rees M, et al. Innovation in clinical method: Diabetes care and negotiating skills. Fam Pract 1995;12(4):413–418. [DOI] [PubMed] [Google Scholar]
- 12. Schonert-Reichl KA, Oberle E, Lawlor MS, et al. Enhancing cognitive and social–emotional development through a simple-to-administer mindfulness-based school program for elementary school children: A randomized controlled trial. Dev Psychol 2015;51(1):52–66. [DOI] [PMC free article] [PubMed] [Google Scholar]