Performance of International Classification of Diseases-based injury severity measures used to predict in-hospital mortality and intensive care admission among traumatic brain-injured patients

Economic Optimization Through Adherence to Best Practice Guidelines: A Decision Analysis of Traumatic Spinal Cord Injury Care Pathways in Australia

Traumatic spinal cord injuries (TSCIs) have significant health, economic, and social effects on individuals, families, and society. In this economic analysis modeling study, we used record-linked administrative patient data from New South Wales, Australia, to construct a decision tree model to compare the economic cost of acute care for patients with TSCI under current clinical pathways with an optimal care (consensus guidelines-informed) modeled pathway. The optimal care pathway included direct transfer to a specialist SCI Unit (SCIU) or indirect transfer to SCIU within 24 h of injury, surgical intervention within 12 h of injury, and subsequent inpatient rehabilitation. Propensity score matching with inverse probability of treatment weighting (IPTW) was used to reduce potential confounding from baseline differences in patient characteristics. A generalized linear model regression with gamma distribution and log link, weighted with IPTW scores, was used for cost and length of stay (LoS) estimations to reduce any residual bias. Sensitivity analyses quantified the sensitivity of the findings to key model parameters. From the healthcare payer perspective, our economic analysis found acute TSCI care at an SCIU was more expensive, with delayed patient transfer pathways, surgery, and timing of surgery driving higher per-patient costs ($14,322 at specialist centers). Probabilistic sensitivity analysis (PSA) using 10,000 Monte Carlo iterations showed the modeled optimal pathway as the expensive option in the majority (86%) of stimulations. However, the modeled direct transfer care pathway demonstrated economic improvements compared to current care pathways, despite a higher upfront cost ($25,428 per patient), the modeled pathway reduced the episode LoS by 5 days (23 days vs. 28 days) on average, generating system-level savings of $20,628 per patient. In PSA, increasing the proportion of patients directly transferred to SCIU by 25%, the optimized pathway was preferred in 28.3% of the simulations. Furthermore, adopting this pathway lowered the incremental per patient cost to $17,157 while preserving a 5-day LoS benefit compared to current pathways (22 days vs. 27 days), which could generate potential savings of $3,471 per patient. Our findings show that guideline-based acute care management is initially resource-intensive but efficient in terms of patient LoS, with a higher proportion of direct transfers resulting in cost savings of $3,471 per patient, which represent system-level benefits from adopting the modeled pathway, rather than episode-level savings. Following consensus guidelines for acute care can provide an economically sustainable approach to resource-intensive patient needs while improving outcomes, as demonstrated in previous studies. In summary, while more intensive, adhering to clinical guidelines of direct transfer to SCIU demonstrates value for patients and health systems. Standardization to optimize time to surgery can achieve improved outcomes through earlier access to rehabilitation and substantial care efficiencies. These findings highlight the economic case for adherence to best practice care guidelines at the healthcare system level to inform future healthcare planning for patients with TSCI.

Performance of ICD-10-based injury severity scores in pediatric trauma patients using the ICD-AIS map and survival rate ratios

OBJECTIVES

The performance of injury severity scores (ISSs), used widely to quantify injury severity and predict outcomes, has not been investigated in German pediatric cases. This study aims to identify the most feasible and accurate injury score predictor of mortality in German children with trauma using International Classification of Diseases 10 (ICD-10).

STUDY DESIGN AND SETTING

Between 2014 and 2020, a retrospective observational cohort study of hospital admissions cases aged <18 years with injury-related ICD-10 codes, using the German hospital database (GHD), was conducted. The maximum abbreviated injury scale and ISS were calculated using the International Classification of Diseases-Abbreviated Injury Scale (ICD-AIS) map provided by the Association for the Advancement of Automotive Medicine, adjusted to the German modification of the ICD-10 classification. The survival risk ratio was used to calculate the single-worst ICD-derived injury (single International Classification of Disease Injury Severity Score [ICISS]) and a multiplicative ICISS. Logistic regressions were conducted for each of the four above-mentioned scores (predictors) to predict in-hospital mortality (outcome) in the selected trauma population and within four clinically relevant subgroups using discrimination and calibration.

RESULTS

1,720,802 were trauma patients, and ICD-AIS mapping was possible in 1,328,377 cases. Cases with mapping failure (n = 392,425; 22.8%) were younger and had a higher mortality rate were excluded from the performance analysis. ICISS-derived scores had a better discrimination and calibration than ICD-AIS based scores in the overall cohort and all four subgroups (area under the curve [AUC] ranges between 0.985 and 0.998 vs 0.886 and- 0.972, respectively).

CONCLUSION

Empirically derived measures of injury severity were superior to ICD-AIS mapped scores in the GHD to predict mortality in pediatric trauma patients. Given the high percentage of mapping failure and high mortality among cases with single-coded injury, the single ICISS may be the most suitable measure of injury severity in this group of patients.

Traumatic brain injury in elderly population: A global systematic review and meta-analysis of in-hospital mortality and risk factors among 2.22 million individuals

BACKGROUND

Traumatic brain injury (TBI) among elderly individuals poses a significant global health concern due to the increasing ageing population.

METHODS

We searched PubMed, Cochrane Library, and Embase from database inception to Feb 1, 2024. Studies performed in inpatient settings reporting in-hospital mortality of elderly people (≥60 years) with TBI and/or identifying risk factors predictive of such outcomes, were included. Data were extracted from published reports, in-hospital mortality as our main outcome was synthesized in the form of rates, and risk factors predicting in-hospital mortality was synthesized in the form of odds ratios. Subgroup analyses, meta-regression and dose-response meta-analysis were used in our analyses.

FINDINGS

We included 105 studies covering 2217,964 patients from 30 countries/regions. The overall in-hospital mortality of elderly patients with TBI was 16 % (95 % CI 15 %-17 %) from 70 studies. In-hospital mortality was 5 % (95 % CI, 3 %-7 %), 18 % (95 % CI, 12 %-24 %), 65 % (95 % CI, 59 %-70 %) for mild, moderate and severe subgroups from 10, 7, and 23 studies, respectively. A decrease in in-hospital mortality over years was observed in overall (1981-2022) and in severe (1986-2022) elderly patients with TBI. Older age 1.69 (95 % CI, 1.58-1.82, P < 0.001), male gender 1.34 (95 % CI, 1.25-1.42, P < 0.001), clinical conditions including traffic-related cause of injury 1.22 (95 % CI, 1.02-1.45, P = 0.029), GCS moderate (GCS 9-12 compared to GCS 13-15) 4.33 (95 % CI, 3.13-5.99, P < 0.001), GCS severe (GCS 3-8 compared to GCS 13-15) 23.09 (95 % CI, 13.80-38.63, P < 0.001), abnormal pupillary light reflex 3.22 (95 % CI, 2.09-4.96, P < 0.001), hypotension after injury 2.88 (95 % CI, 1.06-7.81, P = 0.038), polytrauma 2.31 (95 % CI, 2.03-2.62, P < 0.001), surgical intervention 2.21 (95 % CI, 1.22-4.01, P = 0.009), pre-injury health conditions including pre-injury comorbidity 1.52 (95 % CI, 1.24-1.86, P = 0.0020), and pre-injury anti-thrombotic therapy 1.51 (95 % CI, 1.23-1.84, P < 0.001) were related to higher in-hospital mortality in elderly patients with TBI. Subgroup analyses according to multiple types of anti-thrombotic drugs with at least two included studies showed that anticoagulant therapy 1.70 (95 % CI, 1.04-2.76, P = 0.032), Warfarin 2.26 (95 % CI, 2.05-2.51, P < 0.001), DOACs 1.99 (95 % CI, 1.43-2.76, P < 0.001) were related to elevated mortality. Dose-response meta-analysis of age found an odds ratio of 1.029 (95 % CI, 1.024-1.034, P < 0.001) for every 1-year increase in age on in-hospital mortality.

CONCLUSIONS

In the field of elderly patients with TBI, the overall in-hospital mortality and its temporal-spatial feature, the subgroup in-hospital mortalities according to injury severity, and dose-response meta-analysis of age were firstly comprehensively summarized. Substantial key risk factors, including the ones previously not elucidated, were identified. Our study is thus of help in underlining the importance of treating elderly TBI, providing useful information for healthcare providers, and initiating future management guidelines. This work underscores the necessity of integrating elderly TBI treatment and management into broader health strategies to address the challenges posed by the aging global population.

REVIEW REGISTRATION

PROSPERO CRD42022323231.

International Classification of Disease based Injury Severity Score (ICISS): a comparison of methodologies applied to linked data from New South Wales, Australia

BACKGROUND

The International Classification of Disease Injury Severity Score (ICISS) provides an efficient method to determine injury severity in hospitalised injury patients. Injury severity metrics are of particular interest for the tracking of road transport injury rates and trends. The aims of this study were to calculate ICISS using linked morbidity and mortality datasets and to compare predictive ability of various methods and metrics.

METHODS

This was a retrospective analysis of Admitted Patient Data Collection records from New South Wales, Australia, linked with mortality data. Using a split sample approach, design data (2008-2014; n=1 035 174 periods of care) was used to derive survival risk ratios and calculate various ICISS scales based on in-hospital death and 3-month death. These scales were applied to testing data (2015-2017; n=575 306). Logistic regression modelling was used to determine model discrimination and calibration.

RESULTS

There were 12 347 (1.19%) in-hospital deaths and 29 275 (2.83%) 3-month deaths in the design data. Model discrimination ranged from acceptable to excellent (area under the curve 0.75-0.88). Serious injury (ICISS≤0.941) rates in the testing data varied, with a range of 10%-31% depending on the methodology. The 'worst injury' ICISS was always superior to 'multiplicative injury' ICISS in model discrimination and calibration.

CONCLUSIONS

In-hospital death and 3-month death were used to generate ICISS; the former is recommended for settings with a focus on short-term threat to life, such as in trauma care settings. The 3-month death approach is recommended for outcomes beyond immediate clinical care, such as injury compensation schemes.

Model for Predicting In-Hospital Mortality of Physical Trauma Patients Using Artificial Intelligence Techniques: Nationwide Population-Based Study in Korea

BACKGROUND

Physical trauma-related mortality places a heavy burden on society. Estimating the mortality risk in physical trauma patients is crucial to enhance treatment efficiency and reduce this burden. The most popular and accurate model is the Injury Severity Score (ISS), which is based on the Abbreviated Injury Scale (AIS), an anatomical injury severity scoring system. However, the AIS requires specialists to code the injury scale by reviewing a patient's medical record; therefore, applying the model to every hospital is impossible.

OBJECTIVE

We aimed to develop an artificial intelligence (AI) model to predict in-hospital mortality in physical trauma patients using the International Classification of Disease 10th Revision (ICD-10), triage scale, procedure codes, and other clinical features.

METHODS

We used the Korean National Emergency Department Information System (NEDIS) data set (N=778,111) compiled from over 400 hospitals between 2016 and 2019. To predict in-hospital mortality, we used the following as input features: ICD-10, patient age, gender, intentionality, injury mechanism, and emergent symptom, Alert/Verbal/Painful/Unresponsive (AVPU) scale, Korean Triage and Acuity Scale (KTAS), and procedure codes. We proposed the ensemble of deep neural networks (EDNN) via 5-fold cross-validation and compared them with other state-of-the-art machine learning models, including traditional prediction models. We further investigated the effect of the features.

RESULTS

Our proposed EDNN with all features provided the highest area under the receiver operating characteristic (AUROC) curve of 0.9507, outperforming other state-of-the-art models, including the following traditional prediction models: Adaptive Boosting (AdaBoost; AUROC of 0.9433), Extreme Gradient Boosting (XGBoost; AUROC of 0.9331), ICD-based ISS (AUROC of 0.8699 for an inclusive model and AUROC of 0.8224 for an exclusive model), and KTAS (AUROC of 0.1841). In addition, using all features yielded a higher AUROC than any other partial features, namely, EDNN with the features of ICD-10 only (AUROC of 0.8964) and EDNN with the features excluding ICD-10 (AUROC of 0.9383).

CONCLUSIONS

Our proposed EDNN with all features outperforms other state-of-the-art models, including the traditional diagnostic code-based prediction model and triage scale.

Mortality and Risk Factors in Isolated Traumatic Brain Injury Patients: A Prospective Cohort Study

INTRODUCTION

Outcomes in patients with isolated traumatic brain injury (iTBI) have not been evaluated comprehensively in low-income and middle-income countries. We aimed to study the in-hospital iTBI mortality and its associated risk factors in a prospective multicenter Indian trauma registry.

METHODS

Patients with iTBI (head and neck Abbreviated Injury Score ≥2 and other region Abbreviated Injury Score ≤2) were included. Study variables comprised age, gender, mechanism of injury, systolic blood pressure (SBP) at arrival, Glasgow Coma Scale (GCS) score - classified as mild (13-15), moderate (9-12), and severe (3-8), transfer status, and time to presentation at any participating hospital. A multivariable logistic regression was performed to assess the impact of these factors on 24-h and 30-d mortality following iTBI.

RESULTS

Among 5042 included patients, 24-h and 30-d in-hospital mortalities were 5.9% and 22.4%. On a regression analysis, 30-d mortality was associated with age ≥45 y (odds ratio [OR] = 2.1 [1.6-2.7]), railway injury mechanisms (OR = 2.1 [1.3-3.5]), SBP <90 mmHg (OR = 2.6 [1.6-4.1]), and moderate (OR = 3.8 [3.0-5.0]) to severe (OR = 21.1 [16.8-26.7]) iTBI based on GCS scores. 24-h mortality showed similar trends. Patients transferred to the participating hospitals from other centers had higher odds of 30-d mortality (OR = 1.4 [1.2-1.8]) compared to those arriving directly. Those who received neurosurgical intervention had lower odds of 24-h mortality (0.3 [0.2-0.4]).

CONCLUSIONS

Age ≥45 y, GCS score ≤12, and SBP <90 mmHg at arrival increased the risk of in-hospital mortality from iTBI.

Decoding health status transitions of over 200 000 patients with traumatic brain injury from preceding injury to the injury event

For centuries, the study of traumatic brain injury (TBI) has been centred on historical observation and analyses of personal, social, and environmental processes, which have been examined separately. Today, computation implementation and vast patient data repositories can enable a concurrent analysis of personal, social, and environmental processes, providing insight into changes in health status transitions over time. We applied computational and data visualization techniques to categorize decade-long health records of 235,003 patients with TBI in Canada, from preceding injury to the injury event itself. Our results highlighted that health status transition patterns in TBI emerged along with the projection of comorbidity where many disorders, social and environmental adversities preceding injury are reflected in external causes of injury and injury severity. The strongest associations between health status preceding TBI and health status at the injury event were between multiple body system pathology and advanced age-related brain pathology networks. The interwoven aspects of health status on a time continuum can influence post-injury trajectories and should be considered in TBI risk analysis to improve prevention, diagnosis, and care.

The derivation of an International Classification of Diseases, Tenth Revision-based trauma-related mortality model using machine learning

BACKGROUND

Existing mortality prediction models have attempted to quantify injury burden following trauma-related admissions with the most notable being the Injury Severity Score (ISS). Although easy to calculate, it requires additional administrative coding. International Classification of Diseases (ICD)-based models such as the Trauma Mortality Prediction Model (TMPM-ICD10) circumvent these limitations, but they use linear modeling, which may not adequately capture the intricate relationships of injuries on mortality. Using ICD-10 coding and machine learning (ML) algorithms, the present study used the National Trauma Data Bank to develop mortality prediction models whose performance was compared with logistic regression, ISS, and TMPM-ICD10.

METHODS

The 2015 to 2017 National Trauma Data Bank was used to identify adults following trauma-related admissions. Of 8,021 ICD-10 codes, injuries were categorized into 1,495 unique variables. The primary outcome was in-hospital mortality. eXtreme Gradient Boosting (XGBoost), a ML technique that uses iterations of decision trees, was used to develop mortality models. Model discrimination was compared with logistic regression, ISS, and TMPM-ICD10 using receiver operating characteristic curve and probabilistic accuracy with calibration curves.

RESULTS

Of 1,611,063 patients, 54,870 (3.41%) experienced in-hospital mortality. Compared with those who survived, those who died more frequently suffered from penetrating trauma and had a greater number of injuries. The XGBoost model exhibited superior receiver operating characteristic curve (0.863 [95% confidence interval (CI), 0.862-0.864]) compared with logistic regression (0.845 [95% CI, 0.844-0.846]), ISS (0.828 [95% CI, 0.827-0.829]), and TMPM-ICD10 (0.861 [95% CI, 0.860-0.862]) (all p < 0.001). Importantly, the ML model also had significantly improved calibration compared with other methodologies (XGBoost, coefficient of determination (R2) = 0.993; logistic regression, R2 = 0.981; ISS, R2 = 0.649; TMPM-ICD10, R2 = 0.830).

CONCLUSION

Machine learning models using XGBoost demonstrated superior performance and calibration compared with logistic regression, ISS, and TMPM-ICD10. Such approaches in quantifying injury severity may improve its utility in mortality prognostication, quality improvement, and trauma research.

LEVEL OF EVIDENCE

Prognostic and Epidemiologic; level III.

External validation of International Classification of Injury Severity Score to predict mortality in a Greek adult trauma population

INTRODUCTION

The International Classification of diseases- based Injury Severity Score (ICISS) obtained by empirically derived diagnosis-specific survival probabilities (DSPs) is the best-known risk-adjustment measure to predict mortality. Recently, a new set of pooled DSPs has been proposed by the International Collaborative Effort on Injury Statistics but it remains to be externally validated in other cohorts. The aim of this study was to externally validate the ICISS using international DSPs and compare its prognostic performance with local DSPs derived from Greek adult trauma population.

MATERIALS AND METHODS

This retrospective single-center cohort study enrolled adult trauma patients (≥ 16 years) hospitalized between January 2015 and December 2019 and temporally divided into derivation (n = 21,614) and validation cohorts (n = 14,889). Two different ICISS values were calculated for each patient using two different sets of DSPs: international (ICISSint) and local (ICISSgr). The primary outcome was in-hospital mortality. Models' prediction was performed using discrimination and calibration statistics.

RESULTS

ICISSint displayed good discrimination in derivation (AUC = 0.836 CI 95% 0.819-0.852) and validation cohort (AUC = 0.817 CI 95% 0.797-0.836). Calibration using visual analysis showed accurate prediction at patients with low mortality risk, especially below 30%. ICISSgr yielded better discrimination (AUC = 0.834 CI 95% 0.814-0.854 vs 0.817 CI 95% 0.797-0.836, p ˂ .05) and marginally improved overall accuracy (Brier score = 0.0216 vs 0.0223) compared with the ICISSint in the validation cohort. Incorporation of age and sex in both models enhanced further their performance as reflected by superior discrimination (p ˂ .05) and closer calibration curve to the identity line in the validation cohort.

CONCLUSION

This study supports the use of international DSPs for the ICISS to predict mortality in contemporary trauma patients and provides evidence regarding the potential benefit of applying local DSPs. Further research is warranted to confirm our findings and recommend the widespread use of ICISS as a valid measure that is easily obtained from administrative data based on ICD-10 codes.

Quantifying injury severity for traumatic brain injury with routinely collected health data

BACKGROUND

Routinely collected health data (RCHD) offers many opportunities for traumatic brain injury (TBI) research, in which injury severity is an important factor.

OBJECTIVE

The use of clinical injury severity indices in a context of RCHD is explored, as are alternative measures created for this specific purpose. To identify useful scales for full body injury severity and TBI severity this study focuses on their performance in predicting these currently used indices, while accounting for age and comorbidities.

DATA

This study utilized an extensive population-based RCHD dataset consisting of all patients with TBI admitted to any Belgian hospital in 2016.

METHODS

Full body injury severity is scored based on the (New) Injury Severity Score ((N)ISS) and the ICD-based Injury Severity Score (ICISS). For TBI specifically, the Abbreviated Injury Scale (AIS) Head, Loss of Consciousness and the ICD-based Injury Severity Score for TBI injuries (ICISS) were used in the analysis. These scales were used to predict three outcome variables strongly related to injury severity: in-hospital death, admission to intensive care and length of hospital stay. For the prediction logistic regressions of the different injury severity scales and TBI severity indices were used, and error rates and the area under the receiver operating curve were evaluated visually.

RESULTS

In general, the ICISS had the best predictive performance (error rate between 0.06 and 0.23; AUC between 0.82 [0.81;0.83] and 0.86 [0.85;0.86]). A clearly increasing error rate can be noticed with advancing age and accumulating comorbidity.

CONCLUSION

Both for full body injury severity and TBI severity, the ICISS tends to outperform other scales. It is therefore the preferred scale for use in research on TBI in the context of RCHD. In their current form, the severity scales are not suitable for use in older populations.

The Impact of a Comorbid Spinal Cord Injury on Cognitive Outcomes of Male and Female Patients with Traumatic Brain Injury

INTRODUCTION

Evidence of the effect of comorbid spinal cord injury (SCI) on cognitive outcomes in persons undergoing rehabilitation following newly diagnosed traumatic brain injury (TBI) is limited. We conducted a population-based study to investigate this effect.

OBJECTIVE

To compare cognitive outcomes in patients with TBI with and without a comorbid SCI.

SETTING/PARTICIPANTS

Adult patients diagnosed with TBI were identified and followed for 1 year through provincial health administrative data; those who entered inpatient rehabilitation were studied.

DESIGN

A retrospective matched cohort study using the National Rehabilitation Reporting System data of all acute care and freestanding rehabilitation hospitals in Ontario, Canada.

MAIN MEASURES

The exposure was a comorbid SCI in patients with diagnosed TBI. Exposed patients were matched to unexposed (TBI-only) on sex, age, injury severity, and income, in a ratio of one to two. Gain differences in the cognitive subscale of the Functional Independence Measure were compared between exposed and unexposed patients using multivariable mixed linear model, controlling for comorbidity propensity score, gains in motor function, and rehabilitation care indicators.

RESULTS

Over the first year post injury, 12 750 (0.84%) of all TBI patients entered inpatient rehabilitation, of whom 1359 (10.66%) had a comorbid SCI. A total of 1195 exposed patients (65.4% male, mean age 50.9 ± 20.6 for male and 61.8 ± 21.8 for female patients) were matched to 2390 unexposed patients. Controlling for confounding, exposed patients had lower cognitive gain (beta -0.43; 95% CI -0.72, -0.15), for both male (beta -0.39; 95% CI -0.75, -0.03) and female (beta -0.51; 95% CI -0.97, -0.05) patients. The adverse effects of comorbid SCI were driven largely by lower gains in problem solving and comprehension.

CONCLUSIONS

Adult patients with TBI and comorbid SCI showed a lower cognitive domain response to inpatient rehabilitation than patients with TBI alone. Identifying patients at risk for worse cognitive outcomes may facilitate the development of targeted strategies that improve cognitive outcomes.

Impact of ICD-9-CM to ICD-10-CM coding transition on trauma hospitalization trends among young adults in 12 states

BACKGROUND

We aimed to estimate the impact of the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) coding transition on traumatic injury-related hospitalization trends among young adults across a geographically and demographically diverse group of U.S. states.

METHODS

Interrupted time series analyses were conducted using statewide inpatient databases from 12 states and including traumatic injury-related hospitalizations in adults aged 19-44 years in 2011-2017. Segmented regression models were used to estimate the impact of the October 2015 coding transition on external cause of injury (ECOI) completeness (percentage of hospitalizations with a documented ECOI code) and on population-level rates of injury-related hospitalizations by nature, intent, mechanism, and severity of injury.

RESULTS

The transition to ICD-10-CM was associated with a drop in ECOI completion in the transition month (- 3.7%; P < .0001), but there was no significant change in the positive trend in ECOI completion from the pre- to post-transition periods. There were significant increases post-transition in the measured rates of hospitalization for traumatic brain injury (TBI), unintentional injury, mild injury (injury severity score (ISS) < 9), and injuries caused by drowning, firearms, machinery, other pedestrian, suffocation, and unspecified mechanism. Conversely, there were significant decreases in October 2015 in the rates of hospitalization for assault, injuries of undetermined intent, injuries of moderate severity (ISS 9-15), and injuries caused by fire/burn, other pedal cyclist, other transportation, natural/environmental, and other specified mechanism. A significant increase in the percentage of hospitalizations classified as resulting from severe injury (ISS > 15) was observed when the general equivalence mapping maximum severity method for converting ICD-10-CM codes to ICD-9-CM codes was used. State-specific results for the outcomes of ECOI completion and TBI-related hospitalization rates are provided in an online supplement.

CONCLUSIONS

The U.S. transition from ICD-9-CM to ICD-10-CM coding led to a significant decrease in ECOI completion and several significant changes in measured rates of injury-related hospitalizations by injury intent, mechanism, nature, and severity. The results of this study can inform the design and analysis of future traumatic injury-related health services research studies that use both ICD-9-CM and ICD-10-CM coded data.

LEVEL OF EVIDENCE

II (Interrupted Time Series).

Pre-injury health status and excess mortality in persons with traumatic brain injury: A decade-long historical cohort study

An increasing number of patients are able to survive traumatic brain injuries (TBIs) with advanced resuscitation. However, the role of their pre-injury health status in mortality in the following years is not known. Here, we followed 77,088 consecutive patients (59% male) who survived the TBI event in Ontario, Canada for more than a decade, and examined the relationships between their pre-injury health status and mortality rates in excess to the expected mortality calculated using sex- and age-specific life tables. There were 5792 deaths over the studied period, 3163 (6.95%) deaths in male and 2629 (8.33%) in female patients. The average excess mortality rate over the follow-up period of 14 years was 1.81 (95% confidence interval = 1.76-1.86). Analyses of follow-up time windows showed different patterns for the average excess rate of mortality following TBI, with the greatest rates observed in year one after injury. Among identified pre-injury comorbidity factors, 33 were associated with excess mortality rates. These rates were comparable between sexes. Additional analyses in the validation dataset confirmed that these findings were unlikely a result of TBI misclassification or unmeasured confounding. Thus, detection and subsequent management of pre-injury health status should be an integral component of any strategy to reduce excess mortality in TBI patients. The complexity of pre-injury comorbidity calls for integration of multidisciplinary health services to meet TBI patients' needs and prevent adverse outcomes.

Sex-specific incident dementia in patients with central nervous system trauma

INTRODUCTION

Despite evidence that central nervous system (CNS) trauma, including traumatic brain injury and spinal cord injury, can cause sustained neurocognitive impairment, it remains unclear whether trauma-related variables are associated with incident dementia independently of other known risk factors.

METHODS

All adults without dementia entering the health-care system with diagnoses of CNS trauma were examined for occurrence of dementia. All trauma-related variables were examined as predictors in sex-specific Cox regression models, controlling for other known risk factors.

RESULTS

Over a median follow-up of 52 months, 32,834 of 712,708 patients (4.6%) developed dementia. Traumatic brain injury severity and spinal cord injury interacted with age to influence dementia onset; women were at a greater risk of developing dementia earlier than men, all other factors being equal.

DISCUSSION

Risk stratification of patients with CNS trauma by sex is vital in identifying those most likely to develop dementia and in understanding the course and modifying factors.

Both the multiplicative and single-worst-injury International Classification of Diseases Injury Severity Score underperform in urban Indian hospitals

Introduction

Trauma accounts for 9% of all deaths worldwide, killing almost five million people annually. As India accounts for more than one million of these deaths, research on local trauma care is of great importance. A key aspect of such research is outcome comparisons between contexts. One tool to adjust these comparisons for trauma severity is the International Classification of Diseases Injury Severity Score. The aim was to assess two versions of this score in India.

Methods

The data used were from the project Towards Improved Trauma Care Outcomes in India. Published survival risk ratios were used to calculate multiplicative-International Classification of Diseases Injury Severity Score and single-worst-injury-International Classification of Diseases Injury Severity Score for the 200 most recent non-surviving patients and the surviving patients during the same period. Score performance was measured in discrimination and calibration.

Results

The 30-day prediction single-worst-injury-International Classification of Diseases Injury Severity Score discriminated best with an area under the receiver operating characteristics curve of 0.668 (95% CI 0.645–0.690) and a calibration slope of 0.830 (95% CI 0.708–0.940).

Conclusions

The single-worst-injury-International Classification of Diseases Injury Severity Score applied on 30-day mortality was the only score to calibrate on a satisfactory level. None of the scores had an acceptable discrimination. In interpreting these findings, we see that none of the tested scores can currently be implemented in the studied hospitals.

Task-Sharing for Emergency Neurosurgery: A Retrospective Cohort Study in the Philippines

Objective

The safety and effectiveness of task-sharing (TS) in neurosurgery, delegating clinical roles to non-neurosurgeons, is not well understood. This study evaluated an ongoing TS model in the Philippines, where neurosurgical workforce deficits are compounded with a large neurotrauma burden.

Methods

Medical records from emergency neurosurgical admissions to 2 hospitals were reviewed (January 2015-June 2018): Bicol Medical Center (BMC), a government hospital in which emergency neurosurgery is chiefly performed by general surgery residents (TS providers), and Mother Seton Hospital, an adjacent private hospital where neurosurgery consultants are the primary surgeons. Univariable and multivariable linear and logistic regression compared provider-associated outcomes.

Results

Of 214 emergency neurosurgery operations, TS providers performed 95 and neurosurgeons, 119. TS patients were more often male (88.4% vs. 73.1%; P = 0.007), younger (mean age, 27.6 vs. 50.5 years; P < 0.001), and had experienced road traffic accidents (69.1% vs. 31.4%; P < 0.001). There were no significant differences between Glasgow Coma Scale (GCS) scores on admission. Provider type was not associated with mortality (neurosurgeons, 20.2%; TS, 17.9%; P = 0.68), reoperation, or pneumonia. No significant differences were observed for GCS improvement between admission and discharge or in-hospital GCS improvement, including or excluding inpatient deaths. TS patients had shorter lengths of stay (17.3 days vs. 24.4 days; coefficient, -6.67; 95% confidence interval, -13.01 to -0.34; P < 0.05) and were more likely to undergo tracheostomy (odds ratio, 3.1; 95% confidence interval, 1.30-7.40; P = 0.01).

Conclusions

This study, one of the first to examine outcomes of neurosurgical TS, shows that a strategic TS model for emergency neurosurgery produces comparable outcomes to the local neurosurgeons.

Identifying Predictors of Higher Acute Care Costs for Patients With Traumatic Spinal Cord Injury and Modeling Acute Care Pathway Redesign: A Record Linkage Study

STUDY DESIGN

Record linkage study using healthcare utilization and costs data.

OBJECTIVE

To identify predictors of higher acute-care treatment costs and length of stay for patients with traumatic spinal cord injury (TSCI).

SUMMARY OF BACKGROUND DATA

There are few current or population-based estimates of acute hospitalization costs, length of stay, and other outcomes for people with TSCI, on which to base future planning for specialist SCI health care services.

METHODS

Record linkage study using healthcare utilization and costs data; all patients aged more than or equal to 16 years with incident TSCI in the Australian state of New South Wales (June 2013-June 2016). Generalized Linear Model regression to identify predictors of higher acute care treatment costs for patients with TSCI. Scenario analysis quantified the proportionate cost impacts of patient pathway modification.

RESULTS

Five hundred thirty-four incident cases of TSCI (74% male). Total cost of all acute index episodes approximately AUD$40.5 (95% confidence interval [CI] ±4.5) million; median cost per patient was AUD$45,473 (Interquartile Range: $15,535-$94,612). Patient pathways varied; acute care was less costly for patients admitted directly to a specialist spinal cord injury unit (SCIU) compared with indirect transfer within 24 hours. Over half (53%) of all patients experienced at least one complication during acute admission; their care was less costly if they had been admitted directly to SCIU. Scenario analysis demonstrated that a reduction of indirect transfers to SCIU by 10% yielded overall cost savings of AUD$3.1 million; an average per patient saving of AUD$5,861.

CONCLUSION

Direct transfer to SCIU for patients with acute TSCI resulted in lower treatment costs, shorter length of stay, and less costly complications. Modeling showed that optimizing patient-care pathways can result in significant acute-care cost savings. Reducing potentially preventable complications would further reduce costs and improve longer-term patient outcomes.

LEVEL OF EVIDENCE

3.

Neuro-ICU patient disposition: optimal venue for acute needs

PURPOSE OF REVIEW

This article revises the recent evidence on ICU admission criteria for acute neurological patients [traumatic brain injury (TBI) patients, postoperative neurosurgical procedures and stroke].

RECENT FINDINGS

The appropriate utilization of ICU beds is essential, but it is complex and a challenge to attain. To date there are no widely accepted international guidelines for managing these acute brain-injured patients (stroke, TBI, postneurosurgery) in the ICU. The criteria for ICU admission after neurological acute injury, high-dependency unit or a specialized neurosurgical ward vary from institution to institution depending on local structures and characteristics of the available resources. Better evidence to standardize the treatment and the degree of monitoring is needed during neurological acute injury. It is highly recommended to implement clinical vigilance in these patients regardless of their destination (ICU, stroke unit or ward).

SUMMARY

Currently evidence do not allow to define standardized protocol to guide ICU admission for acute neurological patients (TBI patients, postoperative neurosurgical procedures and stroke).

Dynamic multi-outcome prediction after injury: Applying adaptive machine learning for precision medicine in trauma

OBJECTIVE

Machine learning techniques have demonstrated superior discrimination compared to conventional statistical approaches in predicting trauma death. The objective of this study is to evaluate whether machine learning algorithms can be used to assess risk and dynamically identify patient-specific modifiable factors critical to patient trajectory for multiple key outcomes after severe injury.

METHODS

SuperLearner, an ensemble machine-learning algorithm, was applied to prospective observational cohort data from 1494 critically-injured patients. Over 1000 agnostic predictors were used to generate prediction models from multiple candidate learners for outcomes of interest at serial time points post-injury. Model accuracy was estimated using cross-validation and area under the curve was compared to select among predictors. Clinical variables responsible for driving outcomes were estimated at each time point.

RESULTS

SuperLearner fits demonstrated excellent cross-validated prediction of death (overall AUC 0.94-0.97), multi-organ failure (overall AUC 0.84-0.90), and transfusion (overall AUC 0.87-0.9) across multiple post-injury time points, and good prediction of Acute Respiratory Distress Syndrome (overall AUC 0.84-0.89) and venous thromboembolism (overall AUC 0.73-0.83). Outcomes with inferior data quality included coagulopathic trajectory (AUC 0.48-0.88). Key clinical predictors evolved over the post-injury timecourse and included both anticipated and unexpected variables. Non-random missingness of data was identified as a predictor of multiple outcomes over time.

CONCLUSIONS

Machine learning algorithms can be used to generate dynamic prediction after injury while avoiding the risk of over- and under-fitting inherent in ad hoc statistical approaches. SuperLearner prediction after injury demonstrates promise as an adaptable means of helping clinicians integrate voluminous, evolving data on severely-injured patients into real-time, dynamic decision-making support.

Comparison of two prehospital predictive models for mortality and impaired consciousness after severe traumatic brain injury

BACKGROUND

The primary aim was to investigate the performance of a National Advisory Committee for Aeronautics based predictive model (NACA-BM) for mortality at 14 days and a reference model using motor GCS (GCS-RM). The secondary aim was to compare the models for impaired consciousness of survivors at 14 days (IC-14; GCS ≤ 13).

METHODS

Patients ≥16 years having sustained TBI with an abbreviated injury scale score of head region (HAIS) of >3 were included. Multivariate logistic regression models were used to test models for death and IC-14. The discrimination was assessed using area under the receiver-operating curves (AUROCs); noninferiority margin was -5% between the AUROCs. Calibration was assessed using the Hosmer Lemeshow goodness-of-fit test.

RESULTS

Six hundred and seventy seven patients were included. The median age was 54 (IQR 32-71). The mortality rate was 31.6%; 99 of 438 surviving patients (22.6%) had an IC-14. Discrimination of mortality was 0.835 (95%CI 0.803-0.867) for the NACA-BM and 0.839 (0.807-0.872) for the GCS-RM; the difference of the discriminative ability was -0.4% (-2.3% to +1.7%). Calibration was appropriate for the NACA-BM (χ² 8.42; P = 0. 393) and for the GCS-RM (χ² 3.90; P = 0. 866). Discrimination of IC-14 was 0.757 (0.706-0.808) for the NACA-BM and 0.784 (0.734-0.835) for the GCS-RM; the difference of the discriminative ability was -2.5% (-7.8% to +2.6%). Calibration was appropriate for the NACA-BM (χ² 10.61; P = 0.225) and for the GCS-RM (χ² 6.26; P = 0.618).

CONCLUSIONS

Prehospital prediction of mortality after TBI was good with both models, and the NACA-BM was not inferior to the GCS-RM. Prediction of IC-14 was moderate in both models.

Temporal Delays Along the Neurosurgical Care Continuum for Traumatic Brain Injury Patients at a Tertiary Care Hospital in Kampala, Uganda

BACKGROUND

Significant care continuum delays between acute traumatic brain injury (TBI) and definitive surgery are associated with poor outcomes. Use of the "3 delays" model to evaluate TBI outcomes in low- and middle-income countries has not been performed.

OBJECTIVE

To describe the care continuum, using the 3 delays framework, and its association with TBI patient outcomes in Kampala, Uganda.

METHODS

Prospective data were collected for 563 TBI patients presenting to a tertiary hospital in Kampala from 1 June to 30 November 2016. Four time intervals were constructed along 5 time points: injury, hospital arrival, neurosurgical evaluation, computed tomography (CT) results, and definitive surgery. Time interval differences among mild, moderate, and severe TBI and their association with mortality were analyzed.

RESULTS

Significant care continuum differences were observed for interval 3 (neurosurgical evaluation to CT result) and 4 (CT result to surgery) between severe TBI patients (7 h for interval 3 and 24 h for interval 4) and mild TBI patients (19 h for interval 3 and 96 h for interval 4). These postarrival delays were associated with mortality for mild (P = .05) and moderate TBI (P = .03) patients. Significant hospital arrival delays for moderate TBI patients were associated with mortality (P = .04).

CONCLUSION

Delays for mild and moderate TBI patients were associated with mortality, suggesting that quality improvement interventions could target current triage practices. Future research should aim to understand the contributors to delays along the care continuum, opportunities for more effective resource allocation, and the need to improve prehospital logistical referral systems.

Identification and internal validation of models for predicting survival and ICU admission following a traumatic injury

Background

Measures to improve the accuracy of determining survival and intensive care unit (ICU) admission using the International Classification of Injury Severity Score (ICISS) are not often conducted on a population-wide basis. The aim is to determine if the predictive ability of survival and ICU admission using ICISS can be improved depending on the method used to derive ICISS and incremental inclusion of covariates.

Method

A retrospective analysis of linked injury hospitalisation and mortality data during 1 January 2010 to 30 June 2014 in New South Wales, Australia was conducted. Both multiplicative-injury and single-worst-injury ICISS were calculated. Logistic regression examined 90-day mortality and ICU admission with a range of predictor variables. The models were assessed in terms of their ability to discriminate survivors and non-survivors, model fit, and variation explained.

Results

There were 735,961 index injury admissions, 13,744 (1.9%) deaths within 90-days and 23,054 (3.1%) ICU admissions. The best predictive model for 90-day mortality was single-worst-injury ICISS including age group, gender, all comorbidities, trauma centre type, injury mechanism, and nature of injury as covariates. The multiplicative-injury ICISS with age group, gender, all comorbidities, injury mechanism, and nature of injury was the best predictive model for ICU admission.

Conclusions

The inclusion of comorbid conditions, injury mechanism and nature of injury, improved discrimination for both 90-day mortality and ICU admission. Moves to routinely use ICD-based injury severity measures, such as ICISS, should be considered for hospitalisation data replacing more resource-intensive injury severity classification measures.

Electronic supplementary material

The online version of this article (10.1186/s13049-018-0563-5) contains supplementary material, which is available to authorized users.

Predicting mortality with the international classification of disease injury severity score using survival risk ratios derived from an Indian trauma population: A cohort study

Background

Trauma is predicted to become the third leading cause of death in India by 2020, which indicate the need for urgent action. Trauma scores such as the international classification of diseases injury severity score (ICISS) have been used with great success in trauma research and in quality programmes to improve trauma care. To this date no valid trauma score has been developed for the Indian population.

Study design

This retrospective cohort study used a dataset of 16047 trauma-patients from four public university hospitals in urban India, which was divided into derivation and validation subsets. All injuries in the dataset were assigned an international classification of disease (ICD) code. Survival Risk Ratios (SRRs), for mortality within 24 hours and 30 days were then calculated for each ICD-code and used to calculate the corresponding ICISS. Score performance was measured using discrimination by calculating the area under the receiver operating characteristics curve (AUROCC) and calibration by calculating the calibration slope and intercept to plot a calibration curve.

Results

Predictions of 30-day mortality showed an AUROCC of 0.618, calibration slope of 0.269 and calibration intercept of 0.071. Estimates of 24-hour mortality consistently showed low AUROCCs and negative calibration slopes.

Conclusions

We attempted to derive and validate a version of the ICISS using SRRs calculated from an Indian population. However, the developed ICISS-scores overestimate mortality and implementing these scores in clinical or policy contexts is not recommended. This study, as well as previous reports, suggest that other scoring systems might be better suited for India and other Low- and middle-income countries until more data are available.

Prevalence of a definitive airway in patients with severe traumatic brain injury received at four urban public university hospitals in India: a cohort study

AIM

To estimate the proportion of patients arriving with a Glasgow Coma Scale (GCS) less than 9 who had a definitive airway placed prior to arrival.

METHODS

We conducted a retrospective analysis of the data from a multicentre, prospective observational research project entitled Towards Improved Trauma Care Outcomes in India. Adults aged ≥18 years with an isolated traumatic brain injury (TBI) who were transferred from another hospital to the emergency department of the participating hospital with a GCS less than 9 were included. Our outcome was a definitive airway, defined as either intubation or surgical airway, placed prior to arrival at a participating centre.

RESULTS

The total number of patients eligible for this study was 1499. The median age was 40 years and 84% were male. Road traffic injuries and falls comprised 88% of the causes of isolated TBI. The number of patients with GCS<9 who had a definitive airway placed before reaching the participating centres was 229. Thus, the proportion was 0.15 (95% CI 0.13 to 0.17). The proportions of patients with a definitive airway who arrived after 24 hours (19%) were approximately double the proportion of patients who arrived within 6 hours (10%) after injury to the definitive care centre.

CONCLUSION

The rates of definitive airway placement are poor in adults with an isolated TBI who have been transferred from another health facility to tertiary care centres in India.

A prospective neurosurgical registry evaluating the clinical care of traumatic brain injury patients presenting to Mulago National Referral Hospital in Uganda

Background

Traumatic Brain Injury (TBI) is disproportionally concentrated in low- and middle-income countries (LMICs), with the odds of dying from TBI in Uganda more than 4 times higher than in high income countries (HICs). The objectives of this study are to describe the processes of care and determine risk factors predictive of poor outcomes for TBI patients presenting to Mulago National Referral Hospital (MNRH), Kampala, Uganda.

Methods

We used a prospective neurosurgical registry based on Research Electronic Data Capture (REDCap) to systematically collect variables spanning 8 categories. Univariate and multivariate analysis were conducted to determine significant predictors of mortality.

Results

563 TBI patients were enrolled from 1 June– 30 November 2016. 102 patients (18%) received surgery, 29 patients (5.1%) intended for surgery failed to receive it, and 251 patients (45%) received non-operative management. Overall mortality was 9.6%, which ranged from 4.7% for mild and moderate TBI to 55% for severe TBI patients with GCS 3–5. Within each TBI severity category, mortality differed by management pathway. Variables predictive of mortality were TBI severity, more than one intracranial bleed, failure to receive surgery, high dependency unit admission, ventilator support outside of surgery, and hospital arrival delayed by more than 4 hours.

Conclusions

The overall mortality rate of 9.6% in Uganda for TBI is high, and likely underestimates the true TBI mortality. Furthermore, the wide-ranging mortality (3–82%), high ICU fatality, and negative impact of care delays suggest shortcomings with the current triaging practices. Lack of surgical intervention when needed was highly predictive of mortality in TBI patients. Further research into the determinants of surgical interventions, quality of step-up care, and prolonged care delays are needed to better understand the complex interplay of variables that affect patient outcome. These insights guide the development of future interventions and resource allocation to improve patient outcomes.