| Psychosocial diagnoses (i.e. alcohol misuse, drug misuse, and psychiatric co-morbidity) were used to predict risk, severity, and outcome for head injury after trauma. A retrospective medical record review was conducted on all adult patients admitted to a South Carolina medical hospital as a result of an automobile accident in 1998. All inpatient and outpatient medical records from 1996 through 2000 were gathered to evaluate long-term patterns of hospital utilization. It was hypothesized that patients with acute diagnoses of psychosocial problems (i.e. at time of accident) would better predict head injury and increased severity of injury than chronic diagnoses (i.e. prior to injury). Additionally, patients with co-morbid diagnoses of head injury and psychosocial problems were predicted to have worse outcome after injury than other groups. Analyses included logistic regression (to predict diagnosis of head injury, injury severity, and changes in diagnostic status), analysis of covariance (to predict acute outcome), and repeated measures multivariate analysis of variance (to predict outcomes over time). Psychosocial diagnoses at time of injury better predicted risk and severity than chronic diagnoses. Long-term outcomes were related to chronic diagnoses; however the patterns were complex and, in some cases, contrary to hypotheses. |
