Stress generation theory suggests that individuals with psychiatric disorders have characteristics or engage in behaviors that increase the amount of dependent (self-generated), but not independent (fateful), stress they are exposed to. Our recent comprehensive meta-analysis amalgamating over 30 years of stress generation research (Rnic et al., 2023) documented stress generation effects broadly across various forms of psychopathology. Since the publication of these findings, Dang and Xiao (2025) reanalyzed a subset of studies from the original meta-analysis by controlling for baseline stressful life events (SLEs). We discuss theoretical and statistical concerns with controlling for baseline SLEs when predicting subsequent stressors. First, dependent and independent stress are both composite formative constructs comprised of completely different SLEs aggregated at each assessment wave, such that the formative latent construct is fundamentally different at each time point. Second, temporal precedence of psychopathology relative to SLEs is already established during assessment via careful dating of stressor onsets. Third, given that SLEs are discrete, time-limited experiences, temporal continuity of SLEs cannot be assumed or, in the case of independent stressors, is actively precluded. Fourth, partialing out variance shared among SLEs over time is problematic because shared variance potentially underlies the direct stress generation effect and moderating effects that are critical for examining group differences. With limited exceptions, controlling for baseline SLEs is not recommended. Future stress and stress generation research will benefit from the use of statistical approaches that are aligned with this more precise conceptualization of dependent and independent stress as formative variables. (PsycInfo Database Record (c) 2026 APA, all rights reserved).
Rnic et al. (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: