Climate Worry May Fuel Later Distress in College Students, but Exercise Softens the Blow

Climate change is no longer just a matter of rising seas and shrinking glaciers; for many young people, it has become a persistent weight on the mind. A new three-wave longitudinal study published in BMC Psychology suggests that worry about climate change can quietly set the stage for later anxiety and depressive symptoms among Chinese college students, and that a specific psychological mechanism—general psychological inflexibility—helps explain how that happens. The research, led by Wenxing Bai of Universiti Malaya together with colleagues from institutions in China, South Korea, and Thailand, followed 1,630 students across three survey waves and offers one of the clearest temporal pictures yet of how eco-anxiety translates into psychological distress over time.

The study’s central question was deceptively simple: does worrying about climate change today predict psychological distress months later, and if so, through what psychological route? Cross-sectional surveys have long shown correlations between climate concern and poor mental health, but correlations alone cannot establish direction. By measuring climate change worry, psychological inflexibility, physical activity, and psychological distress at three separate time points, the researchers were able to build a statistical model in which each variable could predict later states while controlling for earlier ones—a design that substantially strengthens causal inference, even if it cannot prove causation outright.

Psychological inflexibility, the proposed mechanism at the heart of the study, is a well-established construct in clinical psychology. It refers to a rigid style of responding to difficult thoughts and feelings: entanglement with distressing cognitions, attempts to suppress or avoid unwanted inner experiences, and a loss of contact with personally valued actions. It is typically measured with the Acceptance and Action Questionnaire-II, the same instrument used in this study, and it is the primary therapeutic target of Acceptance and Commitment Therapy. People high in inflexibility do not necessarily experience more negative thoughts than others; rather, they struggle more with the thoughts they have, and that struggle itself becomes a source of suffering.

The numbers tell a coherent story. After the researchers matched responses across all three waves using unique participant identifiers and retained only students who completed every wave and passed quality checks, the final analytical sample stood at 1,630 participants. Climate change worry measured at Time 1 positively predicted psychological inflexibility at Time 2, with a standardized coefficient of 0.255, statistically significant at p < 0.001, even after controlling for baseline psychological inflexibility, baseline psychological distress, and demographic covariates. In plain terms, students who worried more intensely about climate change became, over time, more psychologically rigid—not merely more distressed, but less able to hold difficult inner experiences without being dominated by them.

That rigidity, in turn, carried the risk forward. Psychological inflexibility at Time 2 positively predicted psychological distress at Time 3 with a standardized coefficient of 0.252 in the fully adjusted model, again significant at p < 0.001. Meanwhile, the direct pathway from climate worry at Time 1 to distress at Time 3 shrank to a small but still significant coefficient of 0.098. This pattern is exactly what a mediation model predicts: much of the long-term effect of climate worry on distress travels through the intermediate state of inflexibility rather than acting directly. The indirect pathway was confirmed using regression-based longitudinal conditional process analysis, supported by simple slope tests, conditional indirect-effect estimation, collinearity diagnostics, and sensitivity analyses designed to probe the robustness of the findings.

But the study did not stop at identifying the mechanism. The researchers also asked whether anything might buffer the chain of risk, and their candidate was physical activity, assessed at Time 2 with the Physical Activity Rating Scale-3. Here the results were nuanced in a way that matters for how the findings are interpreted. Physical activity did not directly predict later psychological distress—its direct coefficient was a negligible 0.009, with p = 0.647, meaning that simply being more active conferred no standalone protective effect on distress in this model. Yet physical activity did something subtler: it significantly moderated the association between psychological inflexibility and later distress, with an interaction coefficient of -0.072, p < 0.001. The negative sign indicates that at higher levels of physical activity, the link between rigidity and subsequent distress was weaker.

The formal index of moderated mediation was significant, confirming that the entire indirect pathway—from climate worry through inflexibility to distress—was attenuated among students who were more physically active. Simple slope analyses illustrated the pattern: for sedentary students, psychological inflexibility was a strong predictor of later distress, whereas for highly active students the slope was visibly flatter. Importantly, the authors are careful about what this does and does not mean. Because physical activity showed no direct effect on distress, and because the conditional indirect association remained statistically significant even at high levels of activity, the findings are best described as attenuation rather than elimination of risk. Exercise, in this model, dampens the transmission of risk; it does not abolish it.

Why might physical activity weaken the inflexibility-to-distress link? The study itself does not test mechanisms, but several plausible explanations align with existing literature. Regular exercise is associated with improved emotion regulation, reduced rumination, and greater tolerance of physiological arousal—capacities that overlap conceptually with psychological flexibility. Physical activity also provides structured, value-consistent behavior, which is precisely what inflexibility erodes. A student who is caught in avoidance loops may nonetheless maintain a running or swimming routine, and that routine can function as an anchor of valued action even when inner experience is difficult. The moderation effect, though modest in magnitude, is consistent with the idea that behavioral activation creates alternative pathways through which distressing cognitions lose their grip.

The longitudinal design deserves particular emphasis, because it is what separates this study from the bulk of the eco-anxiety literature. Many surveys have documented that young people who worry about climate change also report more anxiety and depression, but such snapshots cannot distinguish whether worry causes distress, distress amplifies worry, or both are driven by a third factor. By measuring worry first, inflexibility second, and distress third—and by statistically controlling for baseline distress and baseline inflexibility—the researchers established temporal ordering. The attrition analysis and Harman’s single-factor diagnostic further addressed common longitudinal pitfalls, including differential dropout and common-method variance, lending additional credibility to the reported coefficients.

For university mental health services, the practical implications are twofold. First, the identification of psychological inflexibility as the mediating pathway points toward flexibility-oriented interventions. Acceptance and Commitment Therapy, which explicitly trains acceptance of difficult thoughts and recommitment to valued action, is a natural fit: if climate worry fuels distress by making students psychologically rigid, then teaching them to hold that worry without being consumed by it could interrupt the cascade. Second, the moderation findings suggest that accessible physical activity opportunities—campus sports programs, affordable facilities, low-barrier group exercise—may serve as a modest but meaningful protective layer, not as a replacement for psychological support but as a complement to it. The authors frame their conclusions cautiously, noting that the effects are statistical associations in a single national context and that the physical activity buffer attenuates rather than removes risk. Still, in a world where climate concern among the young is only intensifying, knowing both the mechanism and a partial counterweight is a genuinely useful piece of the puzzle.

Subject of Research: Longitudinal links among climate change worry, psychological inflexibility, physical activity, and psychological distress in Chinese college students

Article Title: Climate change worry and psychological distress among Chinese college students: a three-wave longitudinal moderated mediation study

Article References: Bai, W., Zuo, X., Cao, Y., Lei, Y., Pu, T., Wang, Y., & Wang, J. (2026). Climate change worry and psychological distress among Chinese college students: a three-wave longitudinal moderated mediation study. BMC Psychology. https://doi.org/10.1186/s40359-026-05576-1

Image Credits: AI Generated

DOI: 10.1186/s40359-026-05576-1

Keywords: climate change worry, psychological inflexibility, psychological distress, physical activity, college students, longitudinal study, moderated mediation, eco-anxiety, mental health, Acceptance and Commitment Therapy, China, anxiety