Caregiving-related experiences associated with depression severity and its symptomatology among caregivers of individuals with a severe mental disorder: an online cross-sectional study

Abstract

Caring for a relative with a severe mental disorder puts family caregivers to a great risk of depression. While overall caregiving burden is a strong predictor of depression, the contribution of the various dimensions of burden to caregivers' depression as well as their relationships with depressive symptoms has received little attention. 384 family caregivers completed a cross-sectional online survey including the Center for Epidemiological Studies Depression (CES-D) scale, the Zarit Burden Interview (ZBI), and the Brief Experience of Caregiving Inventory (BECI), measuring caregiving burden and experience. We estimated the structure of the relationships between caregiving experiences (i.e., ZBI and BECI subscales) and CES-D symptoms using a network approach. Negative Emotion/Consequences, (lack of) Positive Personal Experience, and Stigma/Effects on Family were the most connected caregiving dimensions to depression. To untangle the role of the Negative Emotion/Consequences component (by far the most central node in estimated networks), a secondary analysis incorporating its composing items was estimated. Losing control over life, feeling strained around the relative and impaired self-perceived health emerged as central nodes. Interestingly, these caregiving-related dimensions or experiences were differentially connected to depressive symptoms. We discuss how these findings might help future research and inform tailored psychoeducational interventions for family caregivers of people with a severe mental disorder.

Keywords: Burden; Caregiver; Depression; Mental illness; Network analysis; Psychiatry.

Fig. 1

a Network displaying the relationships between total CES-D score, ZBI and BECI dimensions (i.e., subscales). b Network displaying the relationships between CES-D symptoms, ZBI and BECI dimensions. Blue and red edges represent positive and negative partial correlations between nodes, respectively. The thickness of the line indicates the strength of the association (i.e., the edge weight). Note that only estimates for which bootstrap 95% confidence interval did not contain zero are reported. Centrality (Z-scores) of each node, ranked by importance, is shown below the corresponding network

Fig. 2

a Network displaying the relationships between total CES-D score and ZBI items from the Negative Emotion/Consequences dimension. b Network displaying the relationships between CES-D symptoms and ZBI items from the Negative Emotion/Consequences dimension. Edges represent positive partial correlations between nodes. The thickness of the line indicates the strength of the association. Note that only estimates for which 95% bootstrap confidence interval did not contain zero are reported. Centrality (Z-scores) of each node, ranked by importance, is shown below the corresponding network

Fig. 3

a Network displaying the relationships between psychoeducation, CES-D symptoms, ZBI and BECI dimensions (subscales). b Network displaying the relationships between psychoeducation, CES-D symptoms and ZBI items from the Negative Emotion/Consequences dimension. Blue and red edges represent positive and negative partial correlations between nodes, respectively. The thickness of the line indicates the strength of the relationship. Note that we only display and zoom in on relations between psychoeducation and other nodes for the sake of clarity