- Internal entrapment predicts rapid increases in suicidal ideation within 30-120 minutes; external entrapment exerts delayed effects across 60-360 minutes.
- Both internal and external entrapment are cross-sectionally associated with suicidal ideation, explaining 25.3% of within-person variance.
- High-frequency EMA (10 daily prompts) uncovers minute-level dynamics and suggests real-time interventions should prioritise targeting internal entrapment.
J Med Internet Res. 2026 Aug 4;28:e81095. doi: 10.2196/81095.
ABSTRACT
BACKGROUND: The Integrated Motivational-Volitional Model posits entrapment as a central driver of suicidal ideation (SI), yet little is known about how internal (IE) vs external entrapment (EE) unfolds in real time. Prior ecological momentary assessment (EMA) studies have generally treated entrapment as unidimensional and used sampling intervals of several hours, potentially obscuring rapid risk processes.
OBJECTIVE: This study examined (1) cross-sectional and longitudinal temporal associations between IE, EE, and SI across lags of 1-3 prompts; (2) differences in immediacy of IE vs EE effects; and (3) whether the exact time interval (in minutes) between assessments moderates these associations.
METHODS: A total of 91 (62.2% female; Mage=31.6, SD 10.6) adults initiating outpatient psychotherapy (convenience sample) completed a 7-day EMA protocol with 10 daily prompts (30-120 min intervals). Participants provided informed consent, and ethical approval was obtained. A total of 5414 (85% compliance) observations were analyzed using multilevel models with maximum likelihood estimation (assuming missing at random). SI (4 items) and entrapment (IE, EE; 1 item each) were rated on 5-point scales. Lagged effects (up to ≈6 h) controlled for prior SI, and time intervals (in minutes) were tested as moderators.
RESULTS: SI varied primarily within persons (intraclass correlation=0.15). Cross-sectionally, both IE (est=0.16, 95% CI 0.08 to 0.24; P<.001) and EE (est=0.19, 95% CI 0.09 to 0.27; P<.001) were significantly associated with SI, explaining 25.3% of within-person variance (Quasi R²). Lagged analyses revealed distinct temporal patterns. At Lag 1 (30-120 min), IE predicted increases in SI (est=0.13, 95% CI 0.09 to 0.18; P<.001), whereas EE did not (est=0.01, 95% CI -0.03 to 0.06; P=.60). At lag 2 (60-240 min), EE predicted SI (est=0.05, 95% CI 0.01 to 0.10; P<.05), whereas IE was not significant. At lag 3 (90-360 min), IE negatively predicted SI (est=-0.06, 95% CI -0.11 to -0.01; P<.05), while EE positively predicted SI (est=0.05, 95% CI 0.00 to 0.10; P<.05). Prior SI consistently predicted subsequent SI (est range=0.38-0.42; P<.001). Time intervals did not moderate effects (P values>.12).
CONCLUSIONS: Entrapment is a dynamic correlate of SI, with IE and EE showing distinct temporal patterns. IE has rapid, within-minute effects on SI, whereas EE shows delayed associations. This study is innovative in modeling minute-level dynamics using high-frequency EMA. Unlike prior unidimensional, lower-frequency designs, it captures ultra-short-term risk processes and clarifies temporally distinct pathways within the Integrated Motivational-Volitional Model and enhances theoretical precision regarding proximal vs distal drivers of SI. Clinically, results suggest that interventions may benefit from targeting IE in real time.
PMID:42550981 | DOI:10.2196/81095
Share Evidence Blueprint
Save to Google Notes
Search Google Scholar
Save as PDF
⭐ My Revision List
