The current study applied survival analysis to examine factors associated with nondisclosure of human immunodeficiency virus (HIV) serostatus among mothers living with HIV (MLH) who had participated in a cognitive-behavioral intervention to disclose their HIV status to their children. Data were utilized from MLH in the intervention arm of the teaching, raising, and communicating with kids (TRACK; Schulte et al., 2021) trial focusing on serostatus disclosure/nondisclosure across four time points (baseline, 3, 9, and 15 months). MLH (Mage = 39.4) resided in California or Georgia and identified as Latina (33%), Black (54%), White (5%), or multiracial (8%). Physical, mental health, and psychosocial measures from the 3-month assessment were used to predict nondisclosure applying Cox regression survival analysis. Nondisclosure was associated with MLH reporting better physical and mental health, less conflict with their child, better cohesion within their families, and less perceived stigma. MLH reporting better physical functioning were 58% less likely to disclose compared to those reporting physical limitations (hazard ratio [HR] = 0.42). Those reporting lower levels of disclosure self-efficacy were 59% less likely to disclose than those reporting higher levels (HR = 2.47); by 67 weeks into the study, the nondisclosure rate was 56% for those reporting lower self-efficacy compared to 24% for those reporting higher self-efficacy. Addressing the inclination not to disclose when the MLH is feeling healthy may be an aspect to incorporate into future interventions. Furthermore, improving disclosure self-efficacy to a high level appears to be a critical component to intervention success. (PsycInfo Database Record (c) 2024 APA, all rights reserved).