The "For You" Page Problem: Explicit Content Exposure and Mental Health Concerns in Short-Form Video Platforms

Background: Short-form video platforms (TikTok, Instagram Reels, YouTube Shorts) have experienced unprecedented growth globally, with particularly rapid adoption in South Asian developing economies like Nepal. However, concerns regarding algorithmic exposure to inappropriate content and mental health impacts remain understudied in these contexts, where digital literacy and regulatory frameworks are still evolving.Objective: This study provides the first comprehensive multi-platform comparison of public discourse surrounding explicit content exposure, age-related concerns, and mental health implications of short-form video platforms. We address the content-specific research gap identified by Nguyen et al. (2025) through large-scale secondary data analysis, with particular attention to the unique challenges facing developing digital economies like Nepal where English serves as the primary digital lingua franca.Methods: We collected 221,764 user-generated records from five sources: Google Play Store reviews (160,000), Apple App Store reviews (12,086), Reddit posts (48,606), Trustpilot reviews (800), and YouTube comments (18,539). After relevance filtering and deduplication, 8,435 records mentioning content concerns, age-related issues, or mental health were analyzed using VADER sentiment analysis and Latent Dirichlet Allocation (LDA) topic modeling.Results: Analysis revealed highly polarized sentiment (45.0% negative, 42.3% positive, 12.7% neutral) with platform-specific variations. Instagram exhibited the most negative sentiment (M = −0.096), while TikTok showed slightly positive sentiment (M = 0.025) despite receiving the most discussion (49.6% of relevant content). Five dominant themes emerged: (1) child safety and parental concerns, (2) algorithmic content recommendation issues, (3) personal usage experiences, (4) account moderation and banning, and (5) mental health and attention span concerns.Conclusions: Our multi-platform comparison reveals significant platform-specific variations in user sentiment, with Instagram exhibiting the most negative discourse (M = −0.096) despite TikTok dominating discussion volume. These findings align with and extend Nguyen et al.'s (2025) meta-analytic findings on short-form video's cognitive and mental health correlates. The results highlight the need for platform-specific content moderation strategies, transparent algorithmic practices, and age-appropriate safeguards—particularly critical in developing economies like Nepal where digital adoption is outpacing regulatory and educational infrastructure.