Impacts of the Mindfulness Meditation Mobile App Calm on Undergraduate Students' Sleep and Emotional State: Pilot Randomized Controlled Trial

Abstract

Background: Undergraduate students frequently experience negative emotional states and sleep quality, which is believed to have worsened following the COVID-19 pandemic.

Objective: This study piloted the use of a popular mobile mindfulness app (Calm) as a potential intervention to improve state depression, anxiety, stress, and sleep quality in undergraduate students attending a Canadian university, following the COVID-19 pandemic.

Methods: Undergraduate students were randomly assigned to a control or treatment group and completed a series of 3 questionnaires to evaluate baseline state emotional health (Depression Anxiety Stress Scale 42-Item Version [DASS-42], Perceived Stress Scale 10-Item Version [PSS-10], and Pittsburgh Sleep Quality Index). Treatment group participants were instructed to engage with the fully-automated Calm app's sleep section for 30 days: 20 minutes daily, 5 days a week, along with an additional 30 minutes of interaction with other app sections each week, resulting in a goal of 130 minutes per week. The control participants were instructed to continue with everyday life and refrain from the use of mindfulness-based apps for 30 days. Following the 30-day treatment period, all participants repeated the 3 questionnaires. The impact of the treatment on all outcomes was examined using linear mixed model analyses. Independent samples t tests were used to determine if psychosocial health or sleep scores differed between baseline and follow-up and if differences in such scores were present between the groups.

Results: A total of 80 students met the inclusion criteria and were randomly assigned to the control (n=40) or treatment (n=40) group. One control participant was lost to follow-up and 3 treatment participants discontinued engaging with the Calm app. Both control (n=39) and treatment (n=37) groups began with similar demographic, emotional state, and sleep characteristics. Treatment participants engaged with the Calm app's sleep section for an average of 234 minutes per week; however, 54% (20/37) met the minimum prescribed interaction time across all 4 weeks. Following the 30-day treatment period, compared to the control group, the treatment group's state anxiety (mean 14, SD 7.4 vs mean 12, SD 7.8; P=.002), state stress (DASS-42: mean 20, SD 8.8 vs mean 15, SD 8.5; P<.001; PSS-10: mean 22, SD 5.9 vs mean 19, SD 5.9; P=.02), and sleep quality (mean 7.7, SD 2.7 vs mean 6.4, SD 3.5; P<.001) improved. Posttreatment, state stress and perceived stress severity was lower in the treatment versus control group (DASS-42: P=.02; PSS-10: P=.03, respectively).

Conclusions: These pilot findings indicate that a mindfulness app may be an effective tool for reducing state anxiety and stress, as well as enhancing sleep quality among undergraduate university students. A larger, randomized controlled trial should confirm these findings.

Keywords: anxiety; depression; emotional state; mHealth; mindfulness; mobile apps; mobile health; mobile phone; pilot study; sleep; stress; students; university.

Figure 1.. Study design line schematic including recruitment, baseline, intervention, and follow-up.

Figure 2.. CONSORT flow diagram of participants’ eligibility, enrollment, group allocation, treatment, and data analyzed in this pilot study. CONSORT: Consolidated Standards of Reporting Trials.

Figure 3.. Depression, anxiety, and stress scores among the control (n=39) and treatment (n=37) groups pre- and posttreatment with the Calm app. Peaks of the bar columns represent the mean score, and error bars represent the SD. Statistical analysis by linear mixed model. (A) Mean DASS-42 state depression. (B) Mean DASS-42 state anxiety. (C) Mean DASS-42 state stress. (D) Mean PSS-10 state stress. Mean state anxiety and state stress measured by the DASS-42 and PSS-10 decreased significantly in the treatment group following 30 days of treatment. There were no changes in any of the state emotional health outcomes in the control group between baseline and 30 days later. DASS-42: Depression Anxiety Stress Scale-42; PSS-10: Perceived Stress Scale-10. *P<.05, **P<.01, ***P<.001.

Figure 4.. Mean PSQI scores among the control (n=39) and treatment (n=37) groups pre- or posttreatment with the Calm app. Peaks of the bar columns represent the mean score, and error bars represent the SD. Statistical analysis by linear mixed mode. Mean PSQI score decreased (improved) significantly in the treatment group following 30 days of treatment with the Calm app. There were no changes in PSQI score in the control group between baseline and 30 days later. PSQI: Pittsburgh Sleep Quality Index. ***P<.001.