Correlation of SF-12 and SF-36 in a trauma population.

ABSTRACT

BACKGROUND: The SF-36 is a commonly used general measure of health-related quality of life (QoL). The SF-12 is a related tool with less response burden, but its performance in a general trauma population is unknown. HYPOTHESIS: The SF-12 would provide similar QoL information to the SF-36 in blunt trauma patients. METHODS: Adults with nonneurological blunt injury were prospectively enrolled. Demographic, injury, and socioeconomic data were collected. Patients were assessed with functional and psychologic questionnaires 1 and 6 months after injury. Physical (PCS) and mental (MCS) component scores of the SF-36 and SF-12 were compared using Pearson's correlation coefficient. Linear regression identified factors associated with the SF-12 and SF-36 PCS and MCS. Responsiveness to change was assessed using the standardized response mean. RESULTS: Correlation of the PCS was 0.924 and MCS was 0.925 (both P < 0.001). QoL remained below population norms at 6 months. PCS was moderately responsive to change and was equivalent using either the SF-12 or the SF-36. MCS was not responsive to change using either tool. At both time points, at least 25% of patients with normal SF-12 PCS or MCS had SF-36 subscale scores significantly below the normal population. CONCLUSIONS: The SF-12 can be used to assess QoL in trauma patients. The lack of responsiveness to change of the MCS suggests other methods may be necessary to fully evaluate mental QoL. Summary scores may not be sufficient to fully assess QoL in this population. Combining the SF-12 with measures to assess psychosocial variables should be further investigated.