Caregiving for a family member with dementia is a highly stressful situation that may last up to several years, and has been associated with symptoms of depression, anxiety, and sleep problems. These disorders frequently co-occur, with previous studies suggesting a bidirectional relationship between sleep and psychiatric disorders (i.e., anxiety and depression). Several models have highlighted the role of hyperarousal for understanding sleep disorders; however, there is little evidence about how sleep problems, depression and anxiety are linked together. Network analysis (NA) could help exploring the mechanisms underlying the associations between anxiety, depression, and sleep disturbances.

Participants were 368 family caregivers of a person with dementia. The depression-anxiety-sleep symptoms network was composed of 26 items. All the analyses were done using R (version 4.1.1).

Symptoms of tension, shakiness, restlessness, nervousness, and anxiety were strongly connected with the symptom feeling depressed. Symptoms of insomnia were connected to trouble focusing, which was linked to feeling that everything is an effort and apathy. The strongest nodes in the network were shakiness, tension, restlessness, nervousness, and restless sleep. Tension was the node with the most predictive power, while restless sleep was the node with the highest betweenness. Central stability coefficient showed adequate indices.

Hyperarousal symptoms (e.g., tension, restlessness) were the most strongly connected symptoms and showed close connection with symptoms of depression. The strength of these nodes suggests a prominent role of hyperarousal to maintaining, or even fueling, anxiety and depressive symptoms. Besides, our results invite the hypothesis that sleep problems may trigger symptoms specific to depression via fatigue or energy loss. Even though this study is limited by its cross-sectional design, it is the first to examine the network structure of the associations between symptoms of depression, anxiety, and sleep problems in a sample of informal caregivers, and to explore the role of hyperarousal in this network. Future studies should explore the temporal association between symptoms and the network dynamics, including response to the potential treatments.