486 - Exploring the longitudinal relationship between future orientation and youth violence perpetration: A latent class analysis

Poster Number: 486
Publication Number: 486.100

Susheel K. Khetarpal, University of Pittsburgh School of Medicine, Pittsburgh, PA, United States; Kwonho Jeong, University of Pittsburgh School of Medicine, Pittsburgh, PA, United States; Kaleab Abebe, University of Pittsburgh School of Medicine, Pittsburgh, PA, United States; Elizabeth Miller, UPMC Childrens Hospital of Pittsburgh, Pittsburgh, PA, United States; Alison Culyba, University of Pittsburgh School of Medicine, Pittsburgh, PA, United States

Background: Future orientation, defined as hopes and aspirations for the future, is gaining promise as a cross-cutting protective factor against youth violence. Much of the research on future orientation and violence has used cross-sectional designs, assessed limited violence outcomes, and focused on youth in suburban settings.

Objective: To assess how baseline future orientation longitudinally predicted multiple forms of violence among a sample of male youth residing in neighborhoods with concentrated disadvantage.

Design/Methods: Data were drawn from a recently completed sexual violence prevention trial among 817 predominantly African American male adolescents, ages 13 to 19, residing in neighborhoods impacted by high levels of community violence in Pittsburgh, PA. We used latent class analysis (LCA) to create baseline future orientation profiles of participants. LCA models were chosen based on AIC/BIC fit indicates as well as 5% minimum membership threshold within a particular class. We then utilized mixed effects modeling to ascertain how these future orientation classes predicted multiple forms of violence (i.e., weapon violence, bullying, sexual harassment, non-partner sexual violence [NPSV], and intimate-partner sexual violence [SV]) over a 9-month follow-up period. Models were adjusted for age, intervention group, race/ethnicity, and parental education.

Results: Mean age was 15.5 +/- 1.6 years. Most participants identified as Black, Non-Hispanic (74%). Our LCA yielded four classes ranging from low to high future orientation, with nearly 80% of youth belonging to moderately-high and high future orientation classes (Table 1). Levels of violence perpetration among youth were high, ranging from 4.8% perpetrating intimate partner SV to 50.7% bullying someone else (Table 2). Future orientation latent class was significantly associated with all forms of violence studied (all p < 0.01) (Table 3). Compared to the low future orientation class, youth belonging to moderate-high and high future orientation classes were less likely to perpetrate weapon violence (Moderate-High: adjusted odds ratio [aOR]=0.45, 95% CI:[0.20, 1.13]; High: aOR=0.41, 95% CI:[0.18, 0.99]). Use of violence was consistently highest among youth in the low-moderate future orientation class.Conclusion(s): This LCA identified multiple, unique profiles of future orientation among male adolescents from disadvantaged neighborhoods and longitudinal associations with multiple forms of violence. Greater attention to nuanced patterns of future orientation may better inform interventions seeking to harness this protective factor to reduce youth violence.
Susheel Khetarpal - Curriculum VitaeCV_12-25-21 .pdf
Table 2. Prevalence of Violence Perpetration for each Latent Class at Baseline and Follow-Upa, NPSV = Non-Partner Sexual Violence (self-report of how often participant physically hurt, threatened, or pressured someone else they were not in a relationship with into unwanted sexual activity without their consent in the past 9 months).
b, IPSV = Intimate partner sexual violence (self-report of how often participant physically hurt, threatened, or pressured someone else they were in a relationship with into unwanted sexual activity without their consent in the past 9 months).