Longitudinal associations between specific symptoms of depression: Network analysis in a prospective cohort study

Abstract Background Network perspective to mental disorders suggests that depression develops due to interrelated associations between individual symptoms rather than due to a common cause. However, it is unclear whether long-term longitudinal associations between specific symptoms of depression demonstrate coherent patterns. We examined the temporal sequences and changes in depressive symptoms over time, and whether some symptoms are more central than others in inducing changes in the rest of the symptoms over time. We also compared the network structure of depressive symptoms between people who were and were not taking medication for depression or anxiety. Methods Data were from the Survey of Health, Aging and Retirement in Europe, with five follow-ups conducted between 2004 and 2017. Participants who had data on depressive symptoms from at least two study waves were analyzed (n = 72,971). Depressive symptoms were self-reported using the 12-item EURO-D scale. Results All individual symptoms were longitudinally associated with each other. Changes in sad or depressed mood, diminished interest, and suicidal ideation were the most strongly associated with changes in other symptoms. There were no consistent differences in symptom associations between individuals taking versus not taking psychotropic medication. Limitations Depressive symptoms were self-reported and measured every two years, which may dilute some short-term temporal sequences of the symptoms. Conclusions Our findings demonstrate differences between depressive symptoms in their long-term associations with other depressive symptoms in the general population. Changes in sad or depressed mood, diminished interest, and suicidal ideation have the strongest associations with changes in the rest of the symptoms.