App-based sleep-procrastination intervention shows mixed objective benefits in university students
A pilot randomized trial of 46 Chinese university students tested an app delivering the I-PACE (Interaction of Person-Affect-Cognition-Execution) model to reduce sleep procrastination over 8 weeks. The online group showed a numerically earlier sleep onset trajectory on objective wearable measures, though gains were modest and did not survive statistical correction.
Sleep procrastination—intentionally delaying sleep despite wanting to sleep—is common among university students and linked to poor academic and health outcomes. This pilot randomized trial enrolled 46 Chinese university students with high sleep procrastination and assigned them to an 8-week online app intervention (n=16), an equivalent offline intervention (n=16), or a control group (n=14).
Both interventions targeted the I-PACE model, which addresses the person's affect, cognition, and execution around sleep behaviors. Researchers measured sleep timing using both wearable devices (objective) and student self-report (subjective) at baseline, after 8 weeks, and again 2 months later. On subjective measures, all three groups improved similarly over time with no significant difference between groups. On objective wearable data, the online group showed a statistically significant different sleep timing trajectory compared to offline and control groups (Wald χ²=28.79, P=.03), with the app group's mean sleep onset approaching zero hours difference from target sleep time by week 7 (−0.05 h). However, this advantage did not survive statistical correction for multiple comparisons, and sleep onset timing rebounded by week 8, suggesting the effect was not sustained. The authors note that adequately powered trials with formal testing of proposed mechanisms are needed to determine whether app-based I-PACE interventions offer genuine clinical benefit.
The online intervention's apparent advantage on objective wearable measures is intriguing but fragile: the group's mean sleep onset time approached the target by week 7, yet the improvement vanished by week 8, suggesting no sustained benefit. Critically, no single week showed a significant between-group difference after Bonferroni correction for multiple testing—a standard safeguard against false positives in pilot trials. The fact that subjective measures (self-reported sleep procrastination) improved equally across all three groups, including control, hints at a placebo or regression-to-the-mean effect; students recruited for high sleep procrastination naturally regress toward baseline over time. The I-PACE model is theoretically sound—addressing emotion, thoughts, and behavioral execution—but this pilot does not yet prove it outperforms simpler approaches. The study is small (46 total), conducted in one population (Chinese university students), and lacks formal mediation analysis to test whether the intervention actually changed the proposed mechanisms. Future work needs larger sample sizes and direct measurement of emotion regulation, cognition, and behavioral activation to justify the I-PACE approach's added complexity.
The deeper explanation, the effect sizes and caveats, and what this means for your own sleep, exercise and reading — read it inside Cadence.
Open in Cadence →References
- Interaction of Person-Affect-Cognition-Execution-Based Digital Interventions for Sleep Procrastination in Chinese University Students: Pilot Randomized Study. — JMIR formative research (Read the original)