Psychometric Network Analysis in Rehabilitation Research: A Methodological Demonstration in Depression Symptoms of Veterans and Service Members at 1 and 2 Years After Traumatic Brain Injury

Purpose/Objective: Psychometric network analysis (PNA) is an application of dynamic systems theory that can inform measurement of complex rehabilitation phenomena such as depressive symptom patterns in veterans and service members (V/SMs) after traumatic brain injury (TBI). This study applied PNA to the Patient Health Questionnaire-9 (PHQ-9), a common measure of depressive symptoms, in a sample of V/SMs with TBI at Years 1 and 2 (Y1–2) postinjury. Research Method/Design: A sample of 808 V/SMs with TBI participated, 594 contributing PHQ-9 data at Y1 and 585 at Y2. Participants were recruited while or after receiving inpatient postacute rehabilitation from one of five Veterans Affairs Polytrauma Rehabilitation Centers. Results: The networks were stable and invariant over time. At both times, network structure revealed the cardinal depressive symptom “feeling down, depressed, or hopeless,” as evidenced by its strength centrality. In the Y1 network, the suicidal ideation node was connected exclusively to the network through the guilt node, and in the Y2 network, the suicidal ideation node formed a second connection through the low mood node. The guilt node was the second most influential node at Y1 but was replaced by anhedonia node at Y2. Conclusions/Implications: This study demonstrated the potential of PNA in rehabilitation research and identified the primacy of feeling down, depressed, and hopeless after TBI at both Y1 and Y2, with guilt being the second most influential symptom at Y1, but replaced by anhedonia at Y2, providing supportive evidence that the relationships among depressive symptoms after TBI are dynamic over time.