A Cross-Sectional Study of Short-Video Viewing and Young Adult Health: Dose–Response Relationships with Sleep, Dream Anxiety, Ocular Surface, and Quality of Life

Background: Short-video platforms (e.g., Douyin, TikTok) have gained immense popularity among young adults, yet their addictive potential and health consequences remain understudied. Unlike gaming addiction, which has been linked to myopia and mental distress, short-video viewing may pose distinct risks due to its fragmented, high-frequency, and pre-sleep usage patterns. This study aimed to investigate the relationships between short-video viewing and multifaceted health outcomes in Chinese young adults, with a focus on dose–response relationships. Methods: A cross-sectional survey was conducted among 328 young adults aged 20–33 years. Participants were categorized into a non-addicted group (daily short-video viewing < 0.5 h, n = 164) and a short-video viewing group (daily viewing ≥ 1 h, n = 164), further divided by duration: 1–2 h (n = 96), 3–4 h (n = 29), 5–6 h (n = 22), and >6 h (n = 17). Outcome measures included the Hospital Anxiety and Depression Scale (HADS), Van Dream Anxiety Scale (VDAS), 36-Item Short-Form Health Survey (SF-36), Ocular Surface Disease Index (OSDI), Internet Addiction Test (IAT), Chinese Internet Addiction Scale—Revised (CIAS-R), and Mobile Phone Addiction Index (MAPI). Linear regression, one-way ANOVA with Tukey post hoc comparisons (or Kruskal–Wallis tests with Dunn’s test for non-normal variables), and dose–response curve fitting were applied. Segmented regression was used to explore potential dose–response patterns. Results: Compared with the non-addicted group, the short-video viewing group had significantly shorter sleep duration, higher HADS, VDAS, OSDI, IAT, CIAS-R, and MAPI scores, and lower SF-36 scores (all p < 0.0001). Clear dose–response relationships were observed: as daily viewing duration increased, sleep duration decreased linearly (r = −0.73, p < 0.001), while OSDI (r = 0.89, p < 0.001) and VDAS (r = 0.68, p < 0.001) increased. Among participants viewing >6 h/day, all (100%) had moderate-to-severe dry eye (OSDI ≥ 23), and 71% (12/17) had probable anxiety/depression (HADS ≥ 11). Notably, MAPI showed the strongest correlation with VDAS (r = 0.77, p < 0.001), indicating that mobile phone addiction was closely associated with nightmare-related anxiety. The SF-36 physical component summary (PCS) was negatively correlated with daily short-video viewing duration (r = −0.62, p < 0.001). Exploratory threshold analyses suggested that viewing > 3 h/day was associated with greater severity of sleep and ocular surface complaints. The dose–response thresholds identified in this study are exploratory, as the number of participants in the higher-duration categories was limited (3–4 h: n = 29; 5–6 h: n = 22; >6 h: n = 17). These breakpoints should be interpreted with caution and validated in future large-scale studies. Conclusions: Short-video viewing is associated in a dose-dependent manner with sleep quality, ocular surface health, and psychological well-being, with effect sizes generally comparable to, and in some domains (e.g., dry eye and dream anxiety) somewhat larger than, those reported for gaming addiction. The strong association between mobile phone addiction (MAPI) and dream anxiety (VDAS) highlights the distinctive association pattern of pre-sleep short-video use. Future interventional studies are warranted to determine the optimal duration limit. Our exploratory thresholds (approximately 3.2 h/day for sleep and 4.0 h/day for OSDI) provide preliminary reference points, but the specific target duration remains to be established through adequately powered studies designed to compare multiple cut-off levels.