ibij_a_1683227_sm2669.xlsx (15.04 kB)

Prediction of long-term quality of life after severe traumatic brain injury based on variables at hospital admission

dataset

posted on 2019-10-25, 07:29 authored by Julien Maillard, Michael De Pretto, Cecile Delhumeau, Bernhard Walder

Objectives: Variables collected early after severe traumatic brain injury (sTBI) could predict health-related quality of life (HRQoL). Our aim was to determine the prevalence of patients with a low HRQoL 4 years after sTBI and to develop a prediction model including early variables.

Methods: Adult patients with both sTBI [abbreviated injury score of the head region (HAIS) >3] and disease-specific HRQoL assessments using the ‘Quality of Life after Brain Injury’ (QOLIBRI) were included. The outcome was the total score (TS) of QOLIBRI; cutoff for low HRQoL: <60 points. A multivariate logistic regression model and prediction model were performed.

Results: One hundred-sixteen patients [median age 50.8 years (IQR 25.9–62.8; 21.6% >65 years)] were included; 68 (58.6%) with HAIS = 4, 48 (41.4%) with HAIS = 5. Median Glasgow Coma Scale (GCS) was 13 (IQR 3–15). Median TS was 77 (IQR 60–88). Low HRQoL was observed in 28 patients (24.1%). Two variables were associated with low HRQoL: GCS <13, working situation other than employed or retired. The prediction model had an AUROC of 0.765; calibration was moderate (Hosmer Lemeshow Chi2 6.82, p = .556).

Conclusion: One in four patients had a low HRQoL after 4 years. A lower GCS and working situations were associated with low HRQoL.