Traumatic Brain Injury and Depressive Symptoms in Community-Dwelling Older Adults: A Prospective Cohort Study

Mood symptoms are recognized short-term sequelae of traumatic brain injury (TBI). Research regarding depressive symptoms among older, community-based individuals with TBI is limited. Previous studies have largely focused on short-term and intermediate-term follow-up.

The Atherosclerosis Risk in Communities study is an ongoing prospective cohort study of community-dwelling adults in the United States, initially recruited from 1987 to 1989. TBI was defined using self-reported and International Classification of Diseases diagnostic codes. Participants were sampled from centers in Maryland, Minnesota, Mississippi, and North Carolina. Depressive symptoms and antidepressant prescription(s) were assessed beginning at visit 5 (2011-2013) and were analyzed as time-varying repeated outcome measures until visit 7 (2018-2019), representing a maximum follow-up interval of 9 years. We used generalized estimating equations (GEEs) with an autoregressive working correlation structure and binomial distribution with a log link to estimate risk ratios (RRs) for the association of TBI with time-varying depressive symptoms and antidepressant prescription(s), adjusting for sociodemographic characteristics including age, race and study center, income, educational attainment, and history of military service.

Our analysis included 6,607 individuals who attended at least 1 study visit between visits 5 and 7. The median age was 75 years (25th-75th percentiles = 72-80). More than half were female (59.0%), and 23.7% self-identified as Black. There were 2,113 participants (32.0%) with a lifetime history of TBI, most of which were classified as mild. There were 665 individuals (10.1%) who experienced depressive symptoms and 1,225 (18.5%) with antidepressant prescriptions recorded during follow-up. In fully adjusted GEE models, TBI was associated with an increased risk of depressive symptoms (RR 1.59, 95% CI 1.37-1.85) and antidepressant use (RR 1.32, 95% CI 1.20-1.45). These results persisted in analysis of subgroups defined by age, sex, race, and participant health characteristics, regardless of TBI frequency or severity.

We observed a robust and persistent association of TBI with depressive symptoms and antidepressant prescriptions in this cohort of older, community-dwelling adults regardless of participant characteristics. These results suggest that universal screening for and prompt treatment of depressive symptoms in among older adults with a history of TBI is warranted.