# Bridging mental health, cognition and the brain in mild traumatic brain injury: A multilayer network analysis of the TRACK-TBI study

## TL;DR

This study uses a multilayer network approach to identify key factors linking brain structure, mental health, and cognition in mild traumatic brain injury.

## Contribution

The study introduces a novel multilayer network analysis framework to identify bridging variables in mild traumatic brain injury.

## Key findings

- Insomnia severity, immediate verbal memory, somatisation, and processing speed are key bridging variables in mild traumatic brain injury.
- Multilayer network analysis reveals relationships between brain structure, mental health, and cognition in mTBI.
- Bridging variables could inform future training targets to improve outcomes in mTBI patients.

## Abstract

•Multilayer-network analysis allows investigation of brain-behaviour relationships within the same framework.•Bridge strength centrality can identify variables that ‘bridge’ the brain and behavioural domains in such networks.•Insomnia severity, immediate verbal memory, somatisation and processing speed are bridges in mild traumatic brain injury.•These bridges can in the future help pinpoint potential training targets and strategies to improve outcomes in this cohort.

Multilayer-network analysis allows investigation of brain-behaviour relationships within the same framework.

Bridge strength centrality can identify variables that ‘bridge’ the brain and behavioural domains in such networks.

Insomnia severity, immediate verbal memory, somatisation and processing speed are bridges in mild traumatic brain injury.

These bridges can in the future help pinpoint potential training targets and strategies to improve outcomes in this cohort.

People with mild traumatic brain injury (mTBI) suffer from several mental health symptoms (e.g., anxiety, depressive symptoms) and cognitive deficits (e.g., attentional deficits, slowed processing speed). However, symptoms in TBI are largely investigated in isolation, using univariate approaches, ignoring interactions between symptoms and the underlying large-scale brain networks. We constructed the first multilayer network in mTBI to examine relationships between networks of cognition, mental health and structural brain measures and to identify key variables bridging relationships across these networks.

Chronic phase cross-sectional data (6-month follow-up) from 457 mTBI participants was extracted from the TRACK-TBI Longitudinal study. We selected four variables from self-report mental health questionnaires (affective layer), eight cognitive test scores from the NIH toolbox (cognitive layer), and gray matter volumes from eight brain regions of the central executive and salience networks from anatomical MRI scans (brain layer). We used a multilayer network approach to examine the relationships (edges) between all variables (nodes) across layers. We then used the bridge strength centrality metric to identify nodes that ‘bridge’ the affective, cognitive, and brain layers.

In this sample of mTBI participants, across all affective and cognitive layer nodes, only impairments in insomnia were noted. Multilayer network analysis revealed insomnia severity, immediate verbal memory, somatisation and processing speed nodes exceeded an a priori 80th percentile threshold on the bridge strength scores and may therefore be regarded as key nodes potentially bridging relationships across affective, cognitive and brain layers.

The bridging nodes identified in our multilayer network analyses may suggest targets for future studies to develop more customized, efficient, and efficacious treatments to alleviate mental health symptoms and cognitive deficits in mTBI.