Back

Symptoms network analysis of serious mental illness: A cross disasters comparison

Levin, Y.; Bachem, R.; Goodwin, R.; Ben-Ezra, M.

2021-04-13 epidemiology
10.1101/2021.04.10.21255242 medRxiv
Show abstract

BackgroundThe Kessler Psychological Distress Scale (K-6) has been used worldwide in community epidemiological surveys and served as a screening measure for serious mental illness in the general population. We take a novel approach by examining the symptoms network of the K-6 and the exploration of differences between three types of disasters: Nature related, Terror attacks, and COVID-19. AimsTo explore the K-6 symptoms network and its structure replication across the three types of disasters. MethodsA network analysis of psychological distress symptoms as assessed by the K-6 was conducted using data from 9,271 participants from different disaster samples: Terror (n = 5842), COVID-19 (n = 2428), and Nature related (n = 1001). ResultsWhile there were extensive connections between items across all disaster samples, network structure differed across the disaster types. While after a nature related disaster and the COVID-19 pandemic depression- and anxiety-items were interconnected, a terror attack resulted in more separated manifestations of anxiety and depression. Centrality analysis showed "depressed/no cheering up" to be the node with the highest strength centrality in all networks; in the Nature-related network, "restless or fidgety" was also highly central. ConclusionsResults provide evidence of different psychological distress structures in different disasters. Depending on the type of disaster, trauma-focused interventions may need to be augmented, with specific components directed at depression and/or anxiety.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.