Death after discharge: predictors of mortality in older brain-injured patients

Abstract

Background: Older patients with traumatic brain injury (TBI) may be at high risk of death after hospitalization. The purpose of this study was to characterize long-term mortality of older TBI patients who survived to discharge. We hypothesized that predictors of postdischarge mortality differed from those of inpatient mortality.

Methods: A retrospective cohort study was performed on TBI patients older than 55 years admitted to our Level I trauma center between July 1, 2006, and December 31, 2011. Postdischarge deaths were identified by matching patient data with local vital records up to December 31, 2011, when data collection was terminated (censoring). Patients were categorized by age, comorbidities, history of preinjury anticoagulant/prescription antiplatelet agent therapy, injury severity indices, initial TBI type, prehospital living status, discharge location, and discharge condition. The effect of risk factors on postdischarge mortality was evaluated by Cox proportional hazards modeling.

Results: Of 353 patients, 322 (91.2%) survived to discharge. Postdischarge mortality was 19.8% (n = 63) for the study period. Of the postdischarge deaths, 54.0% died within 6 months of discharge, and 68.3% died within 1 year. Median days to death after discharge or censoring were 149 and 410, respectively. Factors associated with death after discharge included age, preinjury anticoagulant use, higher number of Charlson comorbidities, discharge to a long-term care facility, and severe disability. Factors related to injury severity (i.e., Injury Severity Score [ISS], initial Glasgow Coma Scale [GCS] score) and preinjury prescription antiplatelet agent use, previously found to predict inpatient death, did not predict postdischarge mortality.

Conclusion: Older TBI patients who survive to discharge have a significant risk of death within 1 year. Predictors of postdischarge mortality and inpatient death differ. Death after discharge is largely a function of overall health status. Monitoring health status and continued aggressive management of comorbidities after discharge may be essential in determining long-term outcomes.

Level of evidence: Epidemiologic study, level III.