Predicting outcomes after traumatic brain injury: A novel hospital prediction model for a patient reported outcome

Estimation of long-term quality of life in patients sustaining Traumatic brain injuries is a difficult but important task during the early hospitalization. There are very limited tools to assess these outcomes, therefore we aimed to develop a predictive model for quality-of-life that could be used in hospitalized adults with TBIs.The TRACK-TBI dataset was used to identify adult patients with TBI from 2014 to 2018. Multiple variables were assessed to predict favorable versus unfavorable scores on the Quality of Life after Brain Injury-Overall Scale (QOLIBRI-OS).We included 1549 subjects. 57% had a favorable outcome, and were more likely to have private insurance, higher GCS scores, and fewer comorbidities. A model (TBI-PRO) for 3, 6, and 12-month QOLIBRI score was created. The AUROCs for predicting 3, 6 and 12-month favorable QOLIBRI scores were 0.81, 0.79, and 0.76, respectively.The TBI-PRO model adequately estimates long-term outcomes in patients with TBI.