Symposia
Suicide and Self-Injury
Olivia C. Lawrence, M.A. (she/her/hers)
Lab Manager
Columbia University
New York, NY, United States
Ki Eun Shin, Ph.D. (she/her/hers)
Assistant Professor
Long Island University, Post
Long Island City, NY, United States
Da-Eun Lee, PhD (she/her/hers)
Postdoctoral Fellow
Yale School of Medicine
New Haven, CT, United States
Christine B. Cha, Ph.D. (she/her/hers)
Associate Professor
Yale University
New Haven, CT, United States
Many individuals at risk for suicide simultaneously experience a desire to live (DTL) and a desire to die (DTD), with notable within-person variability in daily life (Adler Mandel et al., 2024). Identifying distinct “phenotypes” of suicidal thinking may clarify who is at greatest risk and guide interventions (e.g., Kleiman et al., 2018; Czyz & King, 2015). Little is known about such patterns in adolescence, when suicidal thoughts often emerge (Gaylor et al., 2023). Using smartphone-based ecological momentary assessment (EMA), we examine whether distinct DTL–DTD profiles emerge among adolescents with and without suicidal ideation histories, and whether these profiles differ in lifetime suicidal behaviors.
123 community-based adolescents have been recruited (ages 15-17, M = 15.86, SD = .77). Latent profile analysis (LPA) identified distinct profiles of dynamics in DTD and DTL based on 14-day EMA data (5x/day). Each prompt included two single-item questions assessing momentary DTL and DTD (1=not at all, 7=very much). LPA indicators of DTD and DTL included: mean, maximum, combined variability of standard deviation (SD) and root mean squared successive difference (RMSSD) using principal component analysis (due to high correlation, rs=0.90- 0.92, ps < 0.00), and a within-person correlation between DTD and DTL.
A four-class solution fit best. Class 1 (“Moderate DTL/Mild DTD,” 4.1%) had moderate DTL (M=4.02) and mild DTD (M=2.10), moderate variability, and a modest negative correlation (r=−0.46). Class 2 (“High DTL/Low DTD,” 14.6%) showed high DTL (M=6.14) and low DTD (M=1.39) and a similar correlation to Class 1. Class 3 (“Stable Highest DTL/Minimal DTD,” 76.4%) showed a high DTL (M=6.85), minimal DTD (M=1.00), and minimal variability. Class 4 (“Ambivalent, Variable DTL/Elevated DTD,” 4.9%) exhibited moderate DTL (M=4.94), elevated DTD (M=2.47, Max = 6.50), high variability, and a strong inverse correlation (r=−0.70). After completing data collection (n=150), profiles will be compared on lifetime suicidal ideation and suicidal behavior using the three-step approach (Asparouhov & Muthén, 2014).
Preliminary findings reveal that adolescents differ in how they experience DTL and DTD across everyday life, with many showing stable, resilient profiles and small subgroups displaying more intense, variable, and/or elevated patterns. Once data collection is complete, exploring associations between profiles and suicidal ideation and behaviors may guide precision in risk detection and intervention for adolescents.