respectively. When the subpopulation of each group that contracted their injury through fighting was examined, the corresponding percentages were 34% and 28%, 17%, and 15%, 1% and 6%, respectively—increased percentages of adverse outcomes in the TBI groups and an attenuation of the TBI effect in each “fighting” subgroup. Overall, TBI increased the risk of behavioral discharge 4 times and of criminal conviction 5 times compared with the normal group; it increased the risk of postinjury adverse action by 1.3 times and decreased the risk of medical discharge by 0.64 times compared with the orthopedic-injury group. “Adverse actions” increased in both the TBI and orthopedic-injury groups 1.75–3 times, but they increased more in the TBI group (3.0 times) than in the orthopedic-injury group (2.1 times). The differences between outcomes in the TBI (fighting) and the orthopedic-injury (fighting) subgroups were not statistically significant. The results of the study suggest that any person sustaining a TBI has a higher risk of later aggressive-behavior problems. Because the outcome measures in the study selected the more severe cases of behavioral impairment, the results may show only the peak of the full problem. Premorbid problems (such as fighting) are a risk factor for postinjury behavioral problems but did not produce worse outcomes in the TBI than in the orthopedic-injury group.

In another primary study, Tateno et al. (2003) assessed aggressive behaviors in 89 TBI cases and 26 multiple-trauma cases (without TBI) consecutively admitted into two Iowa hospitals. Severity of brain injury was measured according to the GCS and PTA and classified with the TCDB. Aggressivity was assessed with the Overt Aggression Scale (OAS) and premorbid aggressive behavior was estimated from premorbid police contact and legal actions. Psychiatric assessments were done by a psychiatrist using the PSE and the Structured Clinical Interview for DSM-IV, along with other standard psychiatric rating scales. Structural neuroimaging scans were also collected. Of the TBI group, 33.7% met the criteria for aggressive behavior in the 6 months after injury (called the aggressive group), compared with 11.5% of the non-TBI injured group; the remaining 66.3% of the TBI group showed low aggressive traits (the nonaggressive group). When the aggressive and nonaggressive TBI subjects were compared major depressive disorder was more frequent in the aggressive group (X2, 6.54; df, 1; p = 0.01), and the group had a higher Hamilton Depression Rating Score (t, -3.51; df, 87; p = 0.0007) and a higher Hamilton Anxiety Scale scores (t, -3.37; df, 87; p = 0.001). Focal frontal lobe lesions on a magnetic resonance scan occurred more frequently in the aggressive TBI group than in the nonaggressive group (X2, 8.05; df, 1; p = 0.005), whereas a more diffuse lesion was more frequent in the nonaggressive TBI group.

Virkkunen et al. (1977) examined 1,830 Finnish World Ward II veterans who had received penetrating head injuries and 500 noninjured veteran controls; both groups were followed for up to 37 years. Of the TBI group, 33.1% sustained frontal, 19.5% temporal, 37.7% parietal, and 9.7% occipital lesions. Most veterans in both groups had lifelong employment, despite their injuries. Criminal convictions were no more common in the TBI veterans (5.5%) than in the veteran controls (4.2%), and crimes of violence were not more common in the TBI veterans (0.9%) than in the noninjured veteran group (0.6%). Crimes did not tend to be recurrent in either group.

Secondary Study

Grafman et al. (1996) studied aggression and violence in Vietnam veterans who had TBI and non-brain-injured Vietnam veterans to show that mediofrontal and orbitofrontal lesions of

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement