Objective: To compare the costs of immediate computed tomography during triage for admission with those of observation in hospital in patients with mild head injury.
Design: Prospective cost effectiveness analysis within a multicentre, pragmatic randomised trial.
Setting: 39 acute hospitals in Sweden
Participants: 2602 patients (aged > or = 6) with mild head injury.
Interventions: Immediate computed tomography or admission for observation.
Main outcome measures: Direct and indirect costs related to the mild head injury during the acute and three month follow-up period.
Results: Outcome after three months was similar for both strategies (non-significantly in favour of computed tomography). For the acute stage and complications, the cost was 461 euros (314 pounds sterling, 582 dollars) per patient in the computed tomography group and 677 euros (462 pounds sterling, 854 dollars) in the observation group; an average of 32% less in the computed tomography group (216 euros, 95% confidence interval -272 to -164; P < 0.001). Sensitivity analysis showed that computed tomography was the most cost effective strategy under a broad range of assumptions. After three months, total costs were 718 euros and 914 euros per patient-that is, 196 euros less in the computed tomography group (- 281 to - 114; P < 0.001). The lower cost of the computed tomography strategy at the acute stage thus remained unchanged during follow-up.
Conclusion: Patients with mild head injury attending an emergency department can be managed more cost effectively with computed tomography rather than admission for observation in hospital.
Trial registration: ISRCTN81464462.