Traumatic spinal cord injury (tSCI) may involve new-onset anxiety and depression post-discharge. However, long-term population-based studies have lacked access to follow-up conditions in terms of new-onset anxiety and depression. The objective of this study was to estimate the long-term risk of new-onset anxiety and depression post-discharge. The Longitudinal Health Insurance Database 2000 (LHID2000) from Taiwan’s National Health Insurance Research Database was used in this study. Individuals with tSCI were identified using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnostic codes of 806 and 952 from 1999–2008. The comparison cohort (other health conditions group) was randomly selected from the LHID2000 and was 1:1 matched by age, sex, index year, and comorbidities to reduce the selection bias. All study participants were retrospectively followed for a maximum of 3 years until the end of follow-up, death, or new-onset anxiety (ICD-9-CM: 309.2–309.4) or depression (ICD-9-CM: 296.2, 296.5, 296.82, 300.4, 309.0–309.1, and 311). Persons who were issued a catastrophic illness card for tSCI were categorized as having a severe level of SCI (Injury Severity Score [ISS] ≥16). Poisson regression was used to estimate the incidence rate ratios of anxiety or depression between patients with tSCI and other health conditions. The relative risk of anxiety or depression was estimated using a Cox regression analysis, which was adjusted for potential confounding factors. Univariate analyses showed that the tSCI patients (n = 3556) had a 1.33 times greater incidence of new-onset anxiety or depression (95% confidence interval [CI]: 1.12–1.57) compared to the other health conditions group (n = 3556). After adjusting for potential risk factors, the tSCI patients had a significant 1.29-fold increased risk of anxiety or depression compared to the group with other health conditions (95% CI: 1.09–1.53). Individuals with tSCI, including patients who were under the age of 35, patients who were males, patients who had a low income, and patients without a Charlson Comorbidity Index score, all had a higher long-term risk of anxiety or depression than the other health conditions group (IRRs: 1.84, 1.63, 1.29, and 1.39, respectively). For all tSCI patients, those with an Injury Severity Score (ISS) ≥16 had an almost 2-fold higher risk of anxiety or depression (adjusted Hazard Ratio: 1.85; 95% CI: 1.17–2.92) compared to those with ISS <16. Our findings indicated that tSCI patients have a high risk of anxiety or depression post-discharge, especially among the younger tSCI patients (age <50 years), compared with the other health conditions group. This information could help physicians understand the long-term risk of new-onset anxiety or depression in tSCI patients post-discharge.

Traumatic brain injury (TBI) remains a critical public health challenge. Although studies have found several prognostic factors for TBI, a useful early predictive tool for mortality has yet to be developed in the triage of the emergency room. This study aimed to use machine learning algorithms of artificial intelligence (AI) to develop predictive models for TBI patients in the emergency room triage. We retrospectively enrolled 18,249 adult TBI patients in the electronic medical records of three hospitals of Chi Mei Medical Group from January 2010 to December 2019, and undertook the 12 potentially predictive feature variables for predicting mortality during hospitalization. Six machine learning algorithms including logistical regression (LR) random forest (RF), support vector machines (SVM), LightGBM, XGBoost, and multilayer perceptron (MLP) were used to build the predictive model. The results showed that all six predictive models had high AUC from 0.851 to 0.925. Among these models, the LR-based model was the best model for mortality risk prediction with the highest AUC of 0.925; thus, we integrated the best model into the existed hospital information system for assisting clinical decision-making. These results revealed that the LR-based model was the best model to predict the mortality risk in patients with TBI in the emergency room. Since the developed prediction system can easily obtain the 12 feature variables during the initial triage, it can provide quick and early mortality prediction to clinicians for guiding deciding further treatment as well as helping explain the patient’s condition to family members.