A Meta-Analysis of the Relationship Between Worry and Rumination

Clinical scientists disagree about whether worry and rumination are distinct or represent a unitary construct. To inform this debate, we performed a series of meta-analyses evaluating the relationship between worry and different forms of rumination. A total of 719 effect sizes (N = 69,305) were analyzed. Worry showed a large association with global rumination and with the brooding and emotion-focused subtypes of rumination (rs = .51–.53). However, even when corrected for measurement error, the correlations did not approach unity (ρs = .57–.62). Worry showed a smaller, though still significant, association with the reflection subtype of rumination (r = .28, ρ = .34). Characteristics of the study, sample, and measures moderated the worry–rumination relationship. Worry and rumination, as indexed by current self-report measures, reflect closely related but nonredundant constructs. Given that these constructs have both common and distinct features, researchers should select between them carefully and, when possible, study them together.