Networks of episodic and persistent/recurrent depressive symptoms in older adults: Results from the survey of health, ageing, and retirement in Europe.

OBJECTIVE: The network theory of psychopathology proposes that dense symptom networks foster persistent symptoms. Studies testing this hypothesis in late-life depression (LLD) are lacking. This study examined whether, and how, cross-sectional networks of episodic and persistent/recurrent depressive symptoms in older adults differ. METHOD: Data of adults aged 60 + were drawn from the Survey of Health, Ageing and Retirement in Europe (SHARE). Depressive symptoms were assessed using the EURO-D scale. Participants were classified as episodic or persistent/recurrent based on their EURO-D score at wave 5 (2013) and wave 6 (2015) (EURO-D ≥4). For each group, separate cross-sectional symptom networks at wave 5 were estimated as Ising models and corrected for selection on the sum-score (Berkson's bias). The permutation-based Network Comparison Test was used to compare global strength and network structure. RESULTS: Of 8519 participants, 3746 (44.0%) participants showed episodic depressive symptoms, while 4773 (56.0%) showed persistent/recurrent depressive symptoms. Neither global network strength nor network structure differed statistically significantly between groups (p = .17 and p = .60, respectively). In both networks, lack of interest, depressive mood, and suicidality showed highest strength and expected influence. Findings were consistent across sensitivity analyses. CONCLUSION: Our results showed no difference in global strength and network structure between episodic and persistent/recurrent depressive symptoms in older adults, complementing mixed evidence from younger and middle-aged adults. Causal symptom interactions may play a smaller role in the persistence/recurrence of depressive symptoms in later life. Future studies should examine longitudinal networks, subtypes of LLD, and consider external factors.