This study investigated the relationships between insomnia and select demog
raphic, injury and psychosocial variables in post-acute, traumatic brain in
jury. Clinical assessment of sleep and mood was undertaken via objective me
asures and a diagnostic interview among 91 consecutive brain injury admissi
ons to an outpatient neurorehabilitation clinic. No associations between in
somnia and gender, education, age, and time since injury were found. A logi
stic regression model of insomnia prediction based upon the Beck Depression
Inventory (BDI), sell-reported pain disturbance, litigation and Glasgow Co
ma Score (GCS) correctly classified 87% of the sample with respect to the p
resence or absence of insomnia; however, depression and injury severity wer
e the only variables that made a significant unique contribution to the pre
diction of insomnia. It is concluded that among post-acute traumatic brain
injury patients, insomnia is linked with both the presence of depression an
d a history of milder brain injuries. This suggests that the determinants o
f insomnia may differ from the acute to the post-acute phase, with neurolog
ical factors playing a primary role early in the recovery process and psych
osocial factors ascending later. Therefore, assessment and treatment of ins
omnia must give careful attention to the larger psychosocial context in whi
ch the sleep disorder emerges, particularly to role of emotional disturbanc
e.