Interrelations of resilience factors and their incremental impact for mental health: insights from network modeling using a prospective study across seven timepoints.

Resilience can be viewed as trajectory of stable good mental health or the quick recovery of mental health during or after stressor exposure. Resilience factors (RFs) are psychological resources that buffer the potentially negative effects of stress on mental health. A problem of resilience research is the large number of conceptually overlapping RFs complicating their understanding. The current study sheds light on the interrelations of RFs in the face of the COVID-19 pandemic as a use case for major disruptions. The non-preregistered prospective study assessed a sample of 1275 German-speaking people from February 2020 to March 2021 at seven timepoints. We measured coping, hardiness, control beliefs, optimism, self-efficacy, sense of coherence (SOC), sense of mastery, social support and dispositional resilience as RFs in February 2020, and mental health (i.e., psychopathological symptoms, COVID-19-related rumination, stress-related growth) at all timepoints. Analyses used partial correlation network models and latent growth mixture modeling (LGMM). Pre-pandemic RFs were strongly interrelated, with SOC being the most central node. The strongest associations emerged between coping using emotional support and social support, SOC and sense of mastery, and dispositional resilience and self-efficacy. SOC and active coping were negatively linked. When we examined RFs as predictors of mental health trajectories, SOC was the strongest predictor of psychopathological symptoms and rumination, while trajectories of stress-related growth were predicted by optimism. Subsequent network analyses, including individual intercepts and slopes from LGMM, showed that RFs had small to moderate associations with intercepts but were unrelated to slopes. Our findings provide evidence for SOC playing an important role in mental distress and suggest further examining SOC's incremental validity. However, our results also propose that RFs might be more important for stable levels of mental health than for adaptation processes over time. The differential associations for negative and positive outcomes support the use of multidimensional outcomes in resilience research.