Objective: To determine the contribution of demographics, injury type, pain, and psychological factors on postconcussive symptoms.
Setting and participants: Recently injured (n = 54) and noninjured (n = 184) adults were recruited from a hospital emergency department or the community. Thirty-eight individuals met the diagnostic criteria for a mild traumatic brain injury and 16 individuals received treatment for a minor traumatic non-brain injury.
Main measures: Standardized tests were administered to assess 4 postconcussion symptom types and theorized predictors including a "physiogenic" variable (injury type) and "psychogenic" variables (symptoms of anxiety, depression, and stress) within 1 month of the injury.
Results: In the injured sample, after controlling for injury type, demographics, and pain (chronic and current), a hierarchical regression analysis revealed that the combination of psychological symptoms predicted affective (F10,42 = 2.80, P = .009, Rchange = 0.27) but not other postconcussion symptoms types. Anxiety (β = .48), stress (β = .18), and depression (β = -.07) were not statistically significant individual predictors (P > .05). Cognitive and vestibular postconcussion symptoms were not predicted by the modeled factors, somatic sensory postconcussion symptoms were predicted by demographic factors only, and the pattern of predictors for the symptom types differed for the samples.
Conclusions: Traditional explanatory models do not account for these findings. The predictors are multifactorial, different for injured versus noninjured samples, and symptom specific.