The longitudinal association between perceived stress and dry eye symptoms: A cross‐lagged panel study
Reut Ifrah, Liat Gantz, Avi BesserAbstract
Purpose
Dry eye disease is a multifactorial condition in which symptom burden may reflect ocular surface alterations together with broader psychophysiological processes. Previous studies have reported associations between dry eye symptoms and psychological distress, but most relied on cross‐sectional designs and could not clarify temporal directionality. This study examined the longitudinal association between perceived stress and dry eye symptoms using a two‐wave cross‐lagged panel design.
Methods
Seventy‐one young adults (60 females; mean age 22 ± 3 years) completed the Perceived Stress Scale‐14 (PSS‐14) and the Ocular Surface Disease Index (OSDI) at two time points approximately 7.5 months apart. The second assessment took place under conditions of heightened regional threat. Paired‐samples t tests, Pearson correlations, bootstrap confidence intervals, and cross‐lagged panel modeling were applied. The prespecified directional path from perceived stress at Time 1 to dry eye symptoms at Time 2 was evaluated using a one‐tailed test; all other significance tests were two‐tailed.
Results
Both perceived stress and dry eye symptoms increased significantly over the approximately 7.5‐month follow‐up period. Perceived stress increased by 4.2 points, t (70) = 6.20, p < 0.001, dz = 0.74, and dry eye symptoms increased by 7.5 points, t (70) = 3.46, p < 0.001, dz = 0.41. Perceived stress and dry eye symptoms were positively associated within each wave and showed substantial rank‐order consistency across waves. In the cross‐lagged model, the prospective path from perceived stress at Time 1 to dry eye symptoms at Time 2 was small and positive ( β = 0.18) but did not reach the conventional two‐tailed significance threshold ( p = 0.084); it reached statistical significance only under the prespecified directional one‐tailed test ( p = 0.042). The reverse cross‐lagged path was small and not statistically significant ( β = 0.09).
Conclusions
These findings provide suggestive, but not confirmatory, evidence of a small prospective association between earlier perceived stress and later dry eye symptom burden in a sample of predominantly young adult women. However, the finding should be interpreted cautiously given the modest sample size, the two‐wave design, and its dependence on a directional one‐tailed test. Further longitudinal studies are needed.