关键词: 应激, 心理健康, 曲线关系, 三水平元分析

Abstract: Stress is a pervasive experience that affects individuals across all aspects of daily life. A Lancet survey reported that more than 500 million individuals worldwide suffer from stress-related mental health problems each year. Against this backdrop, public concern for mental health has grown substantially. Subsequently, identifying strategies to enhance mental health and resilience under stress has become an urgent priority. The dose–response theory posits a linear relationship in which stress negatively impacts mental health—a perspective widely held by both the public and academia. However, over the past 15 years, evidence from positive psychology suggests that moderate stress may foster resilience and adaptive psychological outcomes. According to the stress inoculation theory, stress and mental health follow an inverted U-shaped trajectory. Thus, there is a lack of consensus between the linear and curvilinear models. A systematic search of empirical studies reporting quadratic effects of stress on mental health was conducted across multiple databases: Web of Science, ProQuest, EBSCO, and PubMed, as well as the Chinese databases CNKI and WanFang Data. Analyses were performed in R using the “metafor” packages. A three-level meta-analysis was used to synthesize linear (r) and semi-partial (sr) effect sizes and to test moderators. Publication bias was assessed using funnel plots, Egger-MLMA regression tests, and the trim-and-fill method. A total of 23 studies (59 linear and 59 quadratic effect sizes) with 31,422 participants were synthesized to directly compare the two hypotheses. The results indicated a moderate negative correlation between linear stress terms and mental health, and a small and negative partial correlation between quadratic stress terms and mental health. Research evidence indicates that the linear association between stress and mental health remains the dominant pattern. Considering statistical constraints, these findings also supported the curvilinear hypothesis, suggesting a potential curvilinear relationship between stress and mental health. Meta-regression analyses further revealed that the method of stress measurement significantly moderated the curvilinear association, while population characteristics and types of mental health outcomes moderated the linear relationship. Among healthy populations, stress was negatively associated with mental health; in trauma-exposed populations, the linear effect was non-significant. When the outcome was negative mental health, stress consistently predicted worse outcomes, whereas for positive mental health outcomes, the linear effect was not significant. By contrast, no statistically significant moderation was detected in the association between the quadratic component of stress and mental health. However, substantial heterogeneity is observed, indicating that future research should pay greater attention to the nonlinear patterns linking stress and mental health as well as the conditions under which such patterns emerge. In particular, more attention should be given to the potential for individuals to realize their capacities and experience growth under moderate levels of stress. These findings demonstrate a small and negative curvilinear relationship between stress and mental health, indicating that moderate levels of stress may promote better psychological outcomes. The variability of the linear relationship across populations underscores the importance of considering sample characteristics and outcome measures in future research. A more precise understanding of these mechanisms and boundary conditions is critical to clarifying the complex role of stress on mental health.

Key words: stress, mental health, curvilinear relationship, three-level meta-analysis