Transdiagnostic Ecological Momentary Intervention for Improving Self-Esteem in Youth Exposed to Childhood Adversity: The SELFIE Randomized Clinical Trial

IMPORTANCE: Targeting low self-esteem in youth exposed to childhood adversity is a promising strategy for preventing adult mental disorders. Ecological momentary interventions (EMIs) allow for the delivery of youth-friendly, adaptive interventions for improving self-esteem, but robust trial-based evidence is pending. OBJECTIVE: To examine the efficacy of SELFIE, a novel transdiagnostic, blended EMI for improving self-esteem plus care as usual (CAU) compared with CAU only. DESIGN, SETTING, AND PARTICIPANTS: This was a 2-arm, parallel-group, assessor-blinded, randomized clinical trial conducted from December 2018 to December 2022. The study took place at Dutch secondary mental health services and within the general population and included youth (aged 12-26 years) with low self-esteem (Rosenberg Self-Esteem Scale [RSES] <26) exposed to childhood adversity. INTERVENTIONS: A novel blended EMI (3 face-to-face sessions, email contacts, app-based, adaptive EMI) plus CAU or CAU only. MAIN OUTCOMES AND MEASURES: The primary outcome was RSES self-esteem at postintervention and 6-month follow-up. Secondary outcomes included positive and negative self-esteem, schematic self-beliefs, momentary self-esteem and affect, general psychopathology, quality of life, observer-rated symptoms, and functioning. RESULTS: A total of 174 participants (mean [SD] age, 20.7 [3.1] years; 154 female [89%]) were included in the intention-to-treat sample, who were primarily exposed to childhood emotional abuse or neglect, verbal or indirect bullying, and/or parental conflict. At postintervention, 153 participants (87.9%) and, at follow-up, 140 participants (80.5%), provided primary outcome data. RSES self-esteem was, on average, higher in the experimental condition (blended EMI + CAU) than in the control condition (CAU) across both postintervention and follow-up as a primary outcome (B?=?2.32; 95% CI, 1.14-3.50; P?<?.001; Cohen d-type effect size [hereafter, Cohen d] =?0.54). Small to moderate effect sizes were observed suggestive of beneficial effects on positive (B?=?3.85; 95% CI, 1.83-5.88; P?<?.001; Cohen d?=?0.53) and negative (B?=?-3.78; 95% CI, -6.59 to -0.98; P?=?.008; Cohen d?=?-0.38) self-esteem, positive (B?=?1.58; 95% CI, 0.41-2.75; P?=?.008; Cohen d?=?0.38) and negative (B?=?-1.71; 95% CI, -2.93 to -0.48; P?=?.006; Cohen d?=?-0.39) schematic self-beliefs, momentary self-esteem (B?=?0.29; 95% CI, 0.01-0.57; P?=?.04; Cohen d?=?0.24), momentary positive affect (B = 0.23; 95% CI, 0.01-0.45; P = .04; Cohen d = 0.20), momentary negative affect (B = -0.33; 95% CI, -0.59 to -0.03, P = .01, Cohen d = -0.27), general psychopathology (B?=?-17.62; 95% CI, -33.03 to -2.21; P?=?.03; Cohen d?=?-0.34), and quality of life (B?=?1.16; 95% CI, 0.18-2.13; P?=?.02; Cohen d?=?0.33) across postintervention and follow-up. No beneficial effects on symptoms and functioning were observed. CONCLUSIONS AND RELEVANCE: A transdiagnostic, blended EMI demonstrated efficacy on the primary outcome of self-esteem and signaled beneficial effects on several secondary outcomes. Further work should focus on implementing this novel EMI in routine public mental health provision. TRIAL REGISTRATION: Dutch Trial Register Identifier:NL7129(NTR7475).