Effect of digital mindfulness on perceived stress and anxious emotion among college students

The mental health of college students is a growing concern, with heightened social competition contributing to increased psychological pressure. Perceived stress—an individual's subjective appraisal of uncontrollability and difficulty of life events—is prevalent; a survey in China estimates that approximately 67.5% of university students experience medium to high levels of perceived stress. Higher perceived stress is associated with poorer sleep quality and is strongly linked to anxiety and depression; global estimates suggest that 12–50% of college students have anxiety. Mindfulness, defined as purposeful present-moment awareness without judgment, has been widely used to relieve stress and enhance mental health, with multiple studies indicating reductions in negative emotions and improvements in well-being following mindfulness training. Digital delivery of mindfulness via apps and web platforms increases accessibility, standardization, personalization, and effectiveness, and has shown benefits in reducing stress, anxiety, and depression among students. Hypothesis: A 4-week digital mindfulness intervention can reduce college students' perceived stress and anxiety and improve their mindfulness levels.

Prior research shows mindfulness training reduces negative emotions and improves well-being across diverse groups (e.g., Xu et al., Yu et al., Liu et al., Wu et al.). Digital mental health approaches, including smartphone apps and online platforms, offer advantages in accessibility and personalization. Randomized and controlled studies report that brief online mindfulness interventions alleviate depression, rumination, and trait anxiety (Barcaccia et al.), reduce perceived stress and improve resilience among medical students (Fazia et al.), and meta-analyses confirm significant effects of online mindfulness on stress, moderated by dosage (Spijkerman et al.). App-guided compassionate and loving-kindness meditation has improved depression and spirituality (Chiou et al., Liu et al.). Short-video app-guided mindfulness enhanced police officers' conflict resolution and communication skills and reduced emotional disorders and PTSD. Collectively, evidence supports the efficacy of digital mindfulness modalities for stress and anxiety reduction and mindfulness enhancement.

Design: Randomized controlled trial with pre- and post-test assessments. Participants: Convenience sample of college students from two universities in Chongqing, China. Inclusion criteria: (1) voluntary participation; (2) no prior systematic mindfulness training; (3) not using anxiolytics or undergoing other psychological interventions; (4) ability to complete all activities. Recruitment: 328 eligible students enrolled via campus advertisements; all completed baseline and were randomized to intervention (n=164) or control (n=164). Attrition: 18 did not complete post-test; final N=310 (intervention n=156; control n=154); age 18–22 years (M=19.26, SD=0.79). Informed consent obtained. Intervention: 4-week digital mindfulness training, 2–3 sessions/week, 15–20 minutes/session. Content based on Kabat-Zinn’s Mindfulness-Based Stress Reduction (MBSR) and tailored to college students’ cognitive and stress-related characteristics; 10 instructional audio sessions (mindful stretching, breathing, body scan). Delivery via social media; lead trainer released sessions on Mondays, Wednesdays, and Fridays. Participants completed practice same day and logged activity using an online mini-program. Adherence and guidance: Lead trainer responded to questions online, provided guidance, monitored adherence via online questionnaires and social media follow-ups, and adjusted training based on feedback. Control: No intervention during study; offered the same materials post-study with guidance. Measures: Chinese Perceived Stress Scale (CPSS; 14 items; 0–56; higher = more stress; Cronbach’s α baseline 0.74, post 0.82), Generalized Anxiety Disorder-7 (GAD-7; 0–21; higher = more severe anxiety; α baseline 0.79, post 0.85), Five Facet Mindfulness Questionnaire (FFMQ; 39–195; higher = more mindfulness; α baseline 0.70, post 0.77). Procedure: Baseline and post-intervention assessments of CPSS, GAD-7, FFMQ for both groups. Analysis: Data entry in Excel; SPSS 22.0 used for descriptive statistics, independent samples t-tests, χ² tests, and 2×2 repeated-measures ANOVAs (time: pre vs post; group: intervention vs control).

Baseline equivalence: No significant differences between groups in age, sex, CPSS, GAD-7, or FFMQ (e.g., CPSS t=0.10, p=.920; GAD-7 t≈−0.04, p=.969; FFMQ t=0.97, p=.334). Post-test between-group differences (Table 2): CPSS 21.94 ± 5.94 (intervention) vs 25.82 ± 7.03 (control), t=5.25, p<.001; GAD-7 3.51 ± 2.55 vs 5.40 ± 3.09, t=5.88, p<.001; FFMQ 123.36 ± 10.61 vs 120.06 ± 10.69, t=−2.72, p=.007. Perceived stress ANOVA (Table 3): Time main effect F[1,308]=38.90, p<.001, partial η²=.11; Group main effect F[1,308]=10.55, p=.001, partial η²=.03; Group×Time interaction F[1,308]=35.91, p<.001, partial η²=.10. Simple effects: Pre-test group difference ns (F=.10, p=.92); Post-test group difference significant (F=27.58, p<.001, η²=.08). Control pre vs post ns (F=.03, p=.86); Intervention pre vs post significant (F=75.26, p<.001, η²=.20). Anxiety ANOVA (Table 4): Time F=38.01, p<.001, η²=.11; Group F=12.87, p<.001, η²=.04; Interaction F=28.73, p<.001, η²=.09. Simple effects: Pre-test group difference ns (F=.002, p=.97); Post-test group difference significant (F=34.51, p<.001, η²=.10). Control pre vs post ns (F=.32, p=.57); Intervention pre vs post significant (F=66.85, p<.001, η²=.18). Mindfulness ANOVA (Table 5): Time F=10.22, p=.002, η²=.03; Group ns (F=1.18, p=.297); Interaction F=16.81, p<.001, η²=.05. Simple effects: Pre-test group difference ns (F=.94, p=.334); Post-test group difference significant (F=7.42, p=.007, η²=.02). Control pre vs post ns (F=.41, p=.524); Intervention pre vs post significant (F=26.79, p<.001, η²=.08). Overall, the intervention group showed significant reductions in perceived stress and anxiety and increases in mindfulness, whereas the control group showed no significant changes.

Findings support the hypothesis that a 4-week digital mindfulness program decreases perceived stress and anxiety and increases mindfulness among college students. Moderate interaction effect sizes (partial η²≈.10 for stress and .09 for anxiety) indicate meaningful psychological benefits. Potential mechanisms include altered appraisal of unpleasant emotions and sensations, reductions in physiological anxiety indicators, improved cognitive functioning and reduced mind-wandering that enhance present-moment focus, and stress-buffering via acceptance-based monitoring that builds resilience and coping capacity, thereby reducing the negative health impacts of stress. On mindfulness, although the overall group main effect was non-significant, the significant Group×Time interaction indicates that improvements are attributable to the intervention; trait-level mindfulness may require longer practice for stable changes. The digital format offers flexibility, accessibility, lower time and economic costs, and personalization that align with student preferences and constraints, suggesting strong practical utility for university mental health services.

A 4-week digitally delivered mindfulness practice effectively reduced perceived stress and anxiety and enhanced mindfulness among college students, providing evidence in favor of integrating digital mindfulness into campus mental health services. These results support the digital transformation of psychological interventions in higher education. Future work should elucidate the specific mechanisms of digital mindfulness, optimize personalization and guidance features, and explore best practices for embedding such interventions within college mental health education systems.

No long-term follow-up was conducted due to post-study attrition, limiting assessment of sustained effects. The control condition did not rule out placebo or other nonspecific influences (e.g., life events, individual traits). Future studies should include active control groups (e.g., psychoeducation or non-mindfulness relaxation) and multiple post-intervention follow-ups to evaluate specificity, persistence, and stability of effects.