Psychological Distress as a Predictor of Subsequent Loneliness Among Japanese University Students: A Cross-Lagged Analysis During the Post-COVID-19 Transition

Cureus. 2026 Apr 27;18(4):e107845. doi: 10.7759/cureus.107845. eCollection 2026 Apr.

ABSTRACT

Introduction The COVID-19 pandemic significantly affected university students’ mental health, and loneliness is frequently cited as a key contributor to psychological distress. However, whether loneliness drives distress or distress generates loneliness remains unresolved. This study examined this question during the post-COVID-19 transition, when social opportunities had been restored, but psychological effects may have persisted. Methods A longitudinal web-based survey was conducted with Japanese university students at two time points: October 2022 (T1, during COVID-19 restrictions) and February 2024 (T2, post-restrictions). From 3,000 respondents at T1 and 3,075 at T2, 230 (7.7%; 95 male respondents, 135 female respondents; mean age = 20.3 years) completed both surveys. Cross-lagged regression analysis was used to examine the bidirectional relationships between the Kessler Psychological Distress Scale (K6) and UCLA Loneliness Scale (three-item version) scores. Results Loneliness decreased significantly from T1 to T2 (p = 0.017, d = 0.16), whereas distress did not. K6 scores at T1 predicted loneliness at T2 (β = 0.16, 95% confidence interval (CI) = 0.04, 0.28, p = 0.011), but loneliness at T1 did not predict K6 scores at T2 (β = 0.05, 95% CI = -0.06, 0.16, p = 0.333). An equality constraint test indicated that the difference between the two cross-lagged paths was not statistically significant (Δχ²(1) = 1.26, p = 0.263); therefore, the asymmetric pattern is suggestive of temporal precedence rather than confirmatory evidence of a unidirectional effect. T1 distress also predicted increased passive media consumption at T2; however, T1 behavioral changes did not predict T2 psychological outcomes, except that exercise was associated with lower subsequent distress. Conclusions These findings provide preliminary evidence that, over the 16-month period from October 2022 to February 2024, psychological distress is temporally associated with subsequent loneliness rather than the reverse. In this context, loneliness may partly represent a cognitive manifestation of underlying distress. Interventions for lonely university students would benefit from addressing underlying psychological distress, rather than focusing solely on increasing social connections.

PMID:42220809 | PMC:PMC13215972 | DOI:10.7759/cureus.107845