Trends in Mental Health Outcomes of College Students Amid the Pandemic (Roadmap mHealth App): Longitudinal Observational Study

Abstract

Background: The mental health crisis among college students intensified amid the COVID-19 pandemic, suggesting an urgent need for innovative solutions to support them. Previous efforts to address mental health concerns have been constrained, often due to the underuse or shortage of services. Mobile health (mHealth) technology holds significant potential for providing resilience-building support and enhancing access to mental health care.

Objective: This study aimed to examine the trends in mental health and well-being outcomes over 3 years among college students, with an exploratory aim to assess the potential impact of the Roadmap mHealth app on these outcomes.

Methods: A fully automated longitudinal observational study was conducted remotely from a large public academic institution in the Midwestern United States, evaluating mental health and well-being outcomes among college students using the Roadmap mHealth app over 3 fall semesters from 2020 to 2022. The study enrolled 2164 college students in Year I, with 1128 and 1033 students returning in Years II and III, respectively. Participants completed various self-reported measures, including the Patient Health Questionnaire-9 for depression, Generalized Anxiety Disorder-7 for anxiety, and additional metrics for coping, flourishing, and loneliness.

Results: The findings indicated an evolving trajectory in students' mental health. In Year I, depression and anxiety levels were higher compared with levels reported between 2014 and 2019, remaining stable into Year II. However, significant decreases were noted by Year III for both depression (Year I mean 7.78, SD 5.65 vs Year III mean 6.21, SD 4.68; t₁₀₈=-2.90; P=.01) and anxiety (Year I mean 6.61, SD 4.91 vs Year III mean 5.62, SD 4.58; t₁₁₆=-2.02; P=.046). Problem-focused coping decreased initially from Year I (mean 2.46, SD 0.58) to Year II (mean 2.36, SD 0.60; t₁₀₇₃=-5.87; P<.001), then increased by Year III (mean 2.40, SD 0.63; t₇₀₆=2.26; P=.02). Emotion-focused (Year I mean 2.33, SD 0.41 vs Year III mean 2.22, SD 0.47; t₉₉₄=-7.47; P<.001) and avoidant coping (Year I mean 1.76, SD 0.37 vs Year III mean 1.65, SD 0.38; t₉₉₇=-8.53; P=.02) consistently decreased. Loneliness significantly decreased from Year I (mean 5.79, SD 1.74) to Year III (mean 5.17, SD 1.78; t₁₀₁₃=-10.74; P<.001), accompanied by an increase in flourishing from Year I (mean 63.78, SD 14.76) to Year III (mean 66.98, SD 15.06; t₉₉₄=7.22; P<.001). Analysis of app usage indicated that the positive piggy bank and gratitude journal were the favored activities. Greater engagement with the app was positively correlated with enhanced flourishing, even after adjusting for demographic and sociobehavioral factors (β=.04, SE .016; t₃₉₇₄=2.17; P=.03).

Conclusions: In this study, students' mental health and well-being improved, with notable reductions in depression, anxiety, and loneliness, associated with an increase in flourishing. The app did not appear to worsen students' mental health. Based on the usage pattern, it is possible the app enhanced positive psychology-based practices. Future research should explore the efficacy of mHealth interventions through randomized controlled trials to further understand their impact on college students' mental health outcomes.

Trial registration: ClinicalTrials.gov NCT04766788; https://clinicaltrials.gov/ct2/show/NCT04766788.

International registered report identifier (irrid): RR2-10.2196/29561.

Keywords: COVID; COVID-19; SARS-CoV-2; anxiety; app; college; coronavirus; depression; digital health; eHealth; flourishing; higher education; loneliness; longitudinal; mHealth; mental health; mobile phone; pandemic; positive psychology; smartphone; student; telehealth; telemedicine; wellbeing.

Figure 1

Study schema across all 3 years of the study.

Figure 2

CONSORT (Consolidated Standards of Reporting Trials) flow diagram for participant recruitment and enrollment, accompanied by a Venn diagram illustrating student participation across the 3-year study period. Year I: n=2164; Year II: n=1128; and Year III: n=1033. N=755 students participated in all 3 phases; n=373 were unique to Year I and Year II; and n=278 participants were unique to Year I and Year III.

Figure 3

Coping skills across the 3-year study period. The boxes in each plot depict differences between pairs of phases. Panels (A), (B), and (C) illustrate problem-focused, emotion-focused, and avoidant coping, respectively.

Figure 4

Depression (PHQ-9) and anxiety (GAD-7) across the 3-year study period. GAD-7: Generalized Anxiety Disorder-7; PHQ-9: Patient Health Questionnaire-9.

Figure 5

Flourishing and loneliness across the 3-year study period.

Figure 6

Potential pathways influencing student flourishing. We summarize the multilevel models, where the endpoints of the arrows represent the outcomes, and the starting points indicate the predictors. All predictors in the models are lagged by 1 year, allowing them to forecast the outcomes in the subsequent year. Solid lines with asterisks denote significant P values, while dashed lines indicate nonsignificant P values. The β symbol represents the estimated value in the respective model.

Figure 7

UpSet plot for activity engagement patterns in Year I and Year II.