Intramural Healthcare Consumption and Costs After Traumatic Brain Injury: A Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) Study

Traumatic brain injury (TBI) is a global public health problem and a leading cause of mortality, morbidity, and disability. The increasing incidence combined with the heterogeneity and complexity of TBI will inevitably place a substantial burden on health systems. These findings emphasize the importance of obtaining accurate and timely insights into healthcare consumption and costs on a multi-national scale. This study aimed to describe intramural healthcare consumption and costs across the full spectrum of TBI in Europe. The Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) core study is a prospective observational study conducted in 18 countries across Europe and in Israel. The baseline Glasgow Coma Scale (GCS) was used to differentiate patients by brain injury severity in mild (GCS 13–15), moderate (GCS 9–12), or severe (GCS ≤8) TBI. We analyzed seven main cost categories: pre-hospital care, hospital admission, surgical interventions, imaging, laboratory, blood products, and rehabilitation. Costs were estimated based on Dutch reference prices and converted to country-specific unit prices using gross domestic product (GDP)-purchasing power parity (PPP) adjustment. Mixed linear regression was used to identify between-country differences in length of stay (LOS), as a parameter of healthcare consumption. Mixed generalized linear models with gamma distribution and log link function quantified associations of patient characteristics with higher total costs. We included 4349 patients, of whom 2854 (66%) had mild, 371 (9%) had moderate, and 962 (22%) had severe TBI. Hospitalization accounted for the largest part of the intramural consumption and costs (60%). In the total study population, the mean LOS was 5.1 days at the intensive care unit (ICU) and 6.3 days at the ward. For mild, moderate, and severe TBI, mean LOS was, respectively, 1.8, 8.9, and 13.5 days at the ICU and 4.5, 10.1, and 10.3 days at the ward. Other large contributors to the total costs were rehabilitation (19%) and intracranial surgeries (8%). Total costs increased with higher age and greater trauma severity (mild; €3,800 [IQR €1,400–14,000], moderate; €37,800 [IQR €14,900–€74,200], severe; €60,400 [IQR €24,400–€112,700]). The adjusted analysis showed that female patients had lower costs than male patients (odds ratio (OR) 0.80 [CI 0.75–1.85]). Increasing TBI severity was associated with higher costs, OR 1.46 (confidence interval [CI] 1.31–1.63) and OR 1.67 [CI 1.52–1.84] for moderate and severe patients, respectively. A worse pre-morbid overall health state, increasing age and more severe systemic trauma, expressed in the Injury Severity Score (ISS), were also significantly associated with higher costs. Intramural costs of TBI are significant and are profoundly driven by hospitalization. Costs increased with trauma severity and age, and male patients incurred higher costs. Reducing LOS could be targeted with advanced care planning, in order to provide cost-effective care.


Introduction
Each year, *1,500,000 people with traumatic brain injury (TBI) are hospitalized in the European Union, and *57 000 die as a result of a TBI, translating on average into 287 hospital admissions and *12 deaths per 100 000 inhabitants. 1,2The population-based incidence that includes those injuries that are not treated at hospitals can even be as high as 790 per 100,000. 35][6][7] The increasing number of cases combined with the heterogeneity and complexity of TBI will inevitably put a substantial burden on health systems, as the consumption of specialized acute care and long-term rehabilitation or chronic care will concomitantly increase. 1,83][14][15] It is therefore essential to obtain accurate and timely insight into healthcare consumption after TBI, and the cost effectiveness of TBI treatments, to optimize future allocation of restricted healthcare budgets. 16In view of these trends, cost studies have gained more importance, as measurement of healthcare consumption and accompanied costs serves as a fundament for improvement of access to and delivery of healthcare and for identification of potential savings. 1,2,8,17ublished studies report in-hospital costs of patients with TBI to range from $3,079 to $7,800 (e2,721-6,893) for mild TBI patients 16,[18][19][20][21] and from $2,130 to $401,808 (e1,882-355,117) for severe TBI patients. 177][18][19][20][21] Unfortunately, the interpretation, comparability, and generalizability of these study results are difficult and limited.Most available research on costs after TBI frequently suffers from major methodological heterogeneity and inadequate quality, and is commonly restricted to one TBI severity level.Additionally, implementation of clinical guideline recommendations and personnel costs differ across hospitals and countries, resulting in different treatment practices and cost patterns. 9,10,16,22As measurement of healthcare consumption and costs after TBI differs among countries, researchers usually assess strictly local or national expenses, which limits the understanding and possibility of comparisons on a multi-national scale.In order to address these shortcomings, this study aimed to provide a detailed overview of intramural healthcare consumption and healthcare costs arising from hospital admission and inpatient rehabilitation, across the full spectrum of TBI in Europe.

Study design and patients
The Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) core study is a prospective longitudinal non-randomized observational study, registered at clinicaltrials.govNCT02210221, which included patients with TBI from 18 countries across Europe and in Israel between 2014 and 2017.Inclusion criteria were: (1) a clinical diagnosis of TBI, (2) a clinical indication for a computed tomographic (CT) scan, (3) presentation within 24 h of injury, and (4) informed consent obtained according to local and national policies.Patients were excluded if they had a severe pre-existing neurological disorder that would confound outcome assessments.For this particular study, patients from Israel and those <16 years of age were excluded.Ethical approval for the CENTER-TBI study was obtained from all responsible medical ethical committees, and informed consent procedures followed applicable regulations. 23inical data Clinical data were prospectively collected by local research staff using electronic case report forms (eCRF).Data were de-identified using a randomly generated Global Unique Patient Identifier (GUPI) and stored on a secured database by the International Neuroinformatics Coordinating Facility (INCF) (www.incf.org) in Stockholm, Sweden.Data were extracted in January 2021 (version 3.1) and included demographic characteristics, trauma and injury information, results of neurological assessment, imaging, and patient outcomes.Using the baseline Glasgow Coma Scale (GCS) score, patients were classified into three categories of TBI severity: GCS 13-15 (mild TBI), GCS 9-12 (moderate TBI), and GCS 3-8 (severe TBI).24 The baseline GCS score is a derived variable and represents the total GCS score for baseline risk adjustment.The systemic injury severity score (ISS) was categorized into three groups: ISS £16 (minor injury), ISS 17-25 (major injury), and ISS >25 (critical injury).25 Pre-injury health status was classified using the American Society of Anesthesiologists (ASA) physical status classification.26 Brain injury is further described according to the Abbreviated Injury Scale (AIS) and classified as minor, moderate, serious, severe, critical, or unsurvivable.27

Healthcare consumption
Healthcare consumption data were extracted following the same procedure as with clinical data.The healthcare consumption of patients included seven main healthcare service categories: (1) pre-hospital care, including ambulance transportation and, for secondary referral patients, costs of TBI-related admission and any emergent surgical interventions in the ''referring hospital,'' before admission to a CENTER-TBI study hospital; (2) hospital admission, including initial assessment and care at the emergency room (ER) and length of stay (LOS) in days at the ward or ICU; (3) all surgical interventions, both intra-and extra-cranial; (4) imaging of the brain; (5) laboratory; (6)  blood products; and (7) rehabilitation; including only LOS at an inpatient rehabilitation center.Healthcare consumption of outpatient rehabilitation care facilities was not included.The transitions of care forms, in which the care pathway of patients was registered, were used to extract the in-hospital LOS of patients.Inpatient rehabilitation LOS was extracted using the transitions of care forms and patient-reported outcome forms.Missing LOS at the ward, ICU, and rehabilitation were imputed using single imputation.All healthcare services registered within CENTER-TBI and included in this study are reported in Supplementary Table S1.

Healthcare costs
Because of the unavailability of country-specific unit prices for each healthcare service, Dutch reference prices were used as fundament for this study.In addition, definitions, calculations, and sources of country-specific unit prices may vary (e.g., unit prices can differ based on the inclusion/exclusion of personnel costs), which could potentially lead to an over-or underestimation of costs when such unit prices are used.For example, it was found that the reported monthly salary for a senior resident ranged from a low between e500 and e800 in Eastern Europe to a high of e7900 in Norway. 28By using a uniform price list, this study focuses on differences in healthcare consumption rather than price differences among countries.
Reference prices were extracted from the Dutch Guidelines for economic healthcare evaluations. 29Reference prices not mentioned in the Dutch Guidelines were complemented using unit prices reported by the Netherlands Healthcare Authority or by using the average national price, based on declared fees 30,31 (Supplementary Table S1).First, using the Dutch national general consumer price index, all reference prices were corrected to EURO 2017, the last year of patient inclusion (Supplementary Table S2). 32econd, in order to calculate the economic burden of a patient with TBI within Europe, the Dutch reference prices were converted to country-specific unit prices by correcting the Dutch reference prices for the purchasing power parity (PPP) for the general domestic product (GDP) (Supplementary Table S3).The GDP-PPP is the standard measure when comparing differences in life standards among countries. 33hird, the total intramural healthcare costs were calculated by multiplying the number of healthcare units (e.g.length of days at ward and ICU for hospitalization costs) with the corresponding reference price, according to country of admission.See Supplemental Methods, Supplementary Tables S2 and S3 for further details about the calculations.

Statistical analysis
Data were analyzed using descriptive statistics.Baseline characteristics of patients are based on crude data and presented as absolute numbers and percentages.Continuous variables are presented as medians (interquartile range [IQR]) and means (standard deviation [SD]).Median and mean prices were rounded to hundreds.To compare continuous and categorical variables across all subgroups, the Kruskal-Wallis test and the v 2 test were applied respectively.A p value <0.05 was considered statistically significant.Healthcare consumption (i.e., LOS at ICU, ward, and rehabilitation unit) and total healthcare costs were presented for the total study population, including all severities, and according to TBI severity.
Missing data were statistically imputed based on correlations among baseline characteristics, healthcare consumption, in-hospital mortality, and Glasgow Outcome Scale Extended (GOSE) score at 6 months using the mice package in R. 34 To determine betweencountry differences in ICU and ward LOS, a mixed linear regression model was applied, with results presented in forest plots.The country effect was included in the model as a random intercept, and case-mix adjustment was performed using variables in the International Mission for Prognosis and Analysis of Clinical Trials in TBI (IMPACT) prognostic model: age, pupils, GCS score, hypoxia, hypotension, traumatic subarachnoid hemorrhage, epidural hemorrhage, Marshall CT classification, hemoglobin, and glucose measurements. 35Countries including fewer than five patients per severity group were excluded from this analysis.
We used a mixed general linear model (GLM) with gamma distribution and log link function to determine which baseline characteristics were associated with the total intramural healthcare costs.GLM models are recommended for use in linear regression of costs data, as they provide parametric methods of analysis in which non-normal distributions can be specified. 36A random effect for country was added to both the univariable and multi-variable models to account for between-country differences in costs.Statistical analysis were performed in STATA and R version 4.0.4. 37,38

Healthcare consumption
Hospital admission (i.e.including ICU and ward admission) accounted for over half (60%) of the mean total intramural costs (mild TBI: e8,200 [55%], moderate TBI: e33,400 [61%], severe TBI: e48,500 [61%]), of which 47% were related to ICU admission and 13% were related to ward admission (Fig. 1 and Supplementary Table S4).For the total study population, the mean LOS at the ICU and ward were 5.1 and 6.3 days respectively (Table 2).For mild, moderate, and severe TBI, mean LOS was 1.8, 8.9, and 13.5 days in the ICU and 4.5, 10.1, and 10.3 days on the ward, respectively.The mean LOS for inpatient rehabilitation was 13.5 days for the total population and 5.8, 22.1, and 32.6 days, respectively, for mild, moderate, and severe TBI.Rehabilitation costs (19%; e6,400) and intracranial surgeries (8%; e2,700) were also large cost contributors (Fig. 1 and Supplementary Table S4).Costs for all categories were higher for each TBI severity level.Proportion of total costs related to ICU admission and intracranial surgery increased with TBI severity, while proportion of costs related to ward admission, pre-hospital expenses, and extracranial surgery decreased.Patients who sustained TBI as a result of self-harm had the longest ICU and ward LOS (11 and 17 days, respectively).Patients who died during admission had higher median total costs (e18.900 vs. e8,500) (Table 2).
Between-country differences in healthcare consumption Case-mix of patients varied substantially among countries.The total number of patients per country ranged from 15 to 962.France (52%), Sweden (35%), and Lithuania (33%) had a high percentage of severe TBI patients.Patients with critical injury (Injury Severity Score [ISS] = critical) were mostly found in France (67%), Italy (42%) and the United Kingdom (37%) (Supplementary Table S7).Throughout Europe, costs related to hospitalization were the largest contributor to the total intramural costs, especially in Romania (83%), Austria (76%), and France (72%) (Supplementary Fig. S1).The costs generated from intracranial surgery were the highest in Denmark (12%), Lithuania (12%), Sweden (13%), and Hungary (13%).The multi-variable linear regression model showed that across all TBI severities and adjusted for patient characteristics, some differences among countries in the LOS in the ICU and on the ward were present (Fig. 3A-3F).Most profound differences were visible in the LOS in the ICU, especially in the moderate and severe patient groups (Fig. 3D and 3F).Outliers within this analysis are most profoundly caused by the selective sampling of countries.The median b value indicates that mild, moderate, and severe TBI patients with the same baseline characteristics from a random country will have an average ICU LOS longer by 0.33 days, 0.54 days, and 0.29 days, respectively, when compared with another random country (Fig. 3A-F).

Generalized linear model
Female patients showed lower total intramural costs with an OR of 0.80 [CI 0.75-0.85]times lower than male patients.Increasing TBI severity was associated with higher costs for moderate and severe patients: OR 1. 46   4).Hypotension at admission was also associated with higher costs with an OR of 1.18 [CI 1.03-1.35].Increasing severity of CT abnormalities, as measured by the Marshall CT score, was also associated with higher costs.
FIG. 1. Proportion of mean total intramural costs per cost-category according to severity of traumatic brain injury (TBI).The proportion of the total intramural costs from each cost category are plotted in a histogram for each TBI severity level separately.The exact percentage for each cost category (including pre-hospital costs, intensive care unit and ward admission costs, intra-and extracranial surgery costs, laboratory costs, imaging costs, blood products costs, and rehabilitation costs) are presented in the table below the figure.For example, of the total costs within the mild TBI category, 7% of the expenses were from pre-hospital costs.

Patient population
Studies describing the global burden of TBI, estimated that mild TBI accounted for 81% of injuries, moderate TBI for 11% and severe TBI for 8% and estimated that the first-year lifetime costs per person for mild TBI was between US$3395 and US$4636 and respectively US$21379 and US$36648 for moderate and severe patients. 20,39In comparison to these studies, the CENTER-TBI population included only those patients with a CT indication and recruited mostly patients from academic medical centers, leading to a lower proportion of mild TBI patients and higher rates of severely injured patients.1][42][43][44] The exclusion of TBI patients without a CT indication combined with higher proportions of severely injured patients show that the CENTER-TBI study is not fully representative of the European TBI population.As mentioned, the European TBI population is composed mostly of mild TBI patients, for whom CT is not always indicated, and neurosurgical interventions are required in <1%. 45Notwithstanding, stratification on injury severity in our study was based on the baseline clinical assessment wherein clinical deterioration was not accounted for.Additionally, the mild TBI population is a highly heterogeneous group, and although classified as mild, *50% do not reach full recovery 6 months after injury.The possibility of clinical deterioration combined with the heterogeneity of this population and possible presence of extracranial injury could explain their comparable need for inpatient rehabilitation and the observed inhospital mortality rate. 46

Sex differences
We showed that male patients incurred higher total intramural costs, in almost all age and severity groups, than female patients.1][52] Compared with patients with isolated TBI, defined as brain injury without concomitant severe The p value assesses the null hypothesis of no differences among the mild, moderate, and severe subgroups.IQR, interquartile range; AIS, Abbreviated Injury Score; ISS, Injury Severity Score; CT, computed tomography; GOSE, Glasgow Outcome Scale.Extended extracranial injury, patients with severe extracranial injury have longer hospitalizations because of the necessity of continuing treatment for body sites other than the head. 53he presence of severe extracranial injury could lead to longer hospital LOS resulting in higher intramural costs and causing differences in costs between males and females.However, higher costs for male patients remained after adjustment for relevant confounders, including extracranial injury.Several studies have shown that in comparison to male TBI patients, female TBI patients have lower access to trauma centers and are less often admitted to the ICU.5][56] Within CENTER-TBI, differences in care pathways were most frequently observed in patients who sustained mild TBI, wherein women with comparable injury severity and demographic characteristics were more likely to be discharged home after presenting to the ER and were less likely to be admitted to the ICU. 56The differences in healthcare consumption and costs between males and females could therefore be explained by differences in management of TBI and suboptimal healthcare access among female TBI patients.

The elderly and TBI
We reported that an increase in age is associated with an increase in costs, which is line with previous studies showing that increasing age, severe brain injury, and extracranial injury are related to higher hospital costs. 41,57he cost pattern of the elderly did however, differ from

COSTS OF TBI
the younger patient group, as they had shorter ICU LOS and lower costs for surgical interventions.The difference in healthcare consumption by the elderly could be explained by (1) mechanism of injury and (2) their premorbid health state.
In the elderly population, low energy falls are a common cause of TBI, which are most commonly adjoined by injuries to the lower extremities.9][60] Additionally, although most older patients initially had mild TBI, proportions of in-hospital mortality remained high. 61Because of vulnerability and preexisting comorbidities, older adults are less likely to survive their TBI than are their younger counterparts, which could presumably lead to higher consumption of care during the end phase life. 61,62tween-country differences in healthcare consumption In this study, we found some differences in LOS of TBI patients in the ICU and on the ward across countries.Although part of this difference could be explained by a different case mix of patients in each country, differences in ward and ICU LOS remained within each TBI severity level.When interpreting these differences, we should acknowledge that the design of CENTER-TBI, with enrollment of patients in three admission strata (ER, ward, and ICU) led to different recruitment procedures of TBI severities among countries (i.e.some countries enrolled only patients in the ICU stratum, meaning patients admitted directly to the ICU upon presentation).Although we performed extensive case-mix adjustment, we cannot exclude the possibility of remaining differences in the patient population.Besides differences in patient population, the observed between-country differences in healthcare consumption can still be for other reasons, such as the overall health status of the residential population, the proportion of patients with insurance, pharmaceutical costs, and personnel costs. 63Additionally, the economic development of a country determines the health spending per person. 64In general, differences in expenditure also affect the outcome of TBI patients, as lower-resource, developing countries experience significant higher mortality rates than the higher-resource countries. 65Using GDP-corrected prices, we have adjusted for this factor within this study.In addition to these economic factors, the organization of care and guidelines adaptation is an important key factor in healthcare expenditure.The difference in organization of care can result in a difference of guidelines being used; for example, it is known that some countries are more likely to perform CT scans in patients with mild TBI. 54,66Within TBI care, clinical guidelines are scarce and adherence is suboptimal, resulting in considerable between-country variation in treatment of TBI and subsequently different expenditure patterns across countries. 54,67A previous study has shown that there is considerable variation regarding ICU admission policies, especially in the mild TBI population, wherein it is unclear whether a liberal admission policy is truly benefiting the patient while costs are rising. 68

Strength and limitations
The most important strength of this study is the availability of detailed data of high quality collected from several European countries.The data provide a detailed perspective for all severities of TBI, including data about different age groups with detailed clinical presentation, neuroimaging, and performed interventions.However, several limitations should be acknowledged.The CENTER-TBI study consisted mostly of trauma levels I and II hospitals, resulting in a population of relatively severely injured patients.This may not correctly represent the total TBI population in Europe, as trauma level I centers are known to have overall higher expenses resulting in higher costs. 69This, combined with the selective sampling per country, makes it overall difficult to interpret between-country differences.Total costs were calculated using inflation-and GDPcorrected cost prices, as health financial systems are determinative of the care products' cost prices.Because of the use of inflation-and GDP-corrected prices in this study, we were able to compare the cost of TBI across countries, and focus on healthcare consumption rather than price differences.However, it should be noted that adjustment for GDP-PPP does not fully compensate for actual cost differences among countries.Second, our study did not include detailed information about the interventions in the first hospital for referred patients, despite the burden of TBI in acute care being substantial. 11With 17% of our study population consisting of secondary referrals, missing data on the total healthcare consumption in acute care setting at the referring hospital, could cause an underestimation of the total costs.
In our study, information on long-term healthcare consumption, such as outpatient rehabilitation care and outpatient clinic visits, was not available.Outpatient rehabilitation care and outpatient clinic visits are inevitably large contributors to the overall costs of TBI.After TBI, a range of problems can persist, including cognitive impairment, post-concussion symptoms, emotional difficulties, and functional limitations, requiring long-term outpatient care. 46A study conducted in the United States has shown that patients receiving inpatient rehabilitation still experience major health consequences 5 years after injury, wherein 12% were living in an institutional setting and ‰ FIG. 3.This panel shows forest plots reporting the random country effect (random intercept estimate and 95% confidence intervals) on the length of stay at the ICU and ward for mild (A-B), moderate (C-D) and severe (E-F) TBI patients.Countries including fewer than five patients per severity group were excluded from this analysis.The models included adjustment according to the International Mission for Prognosis and Analysis of Clinical Trials in TBI (IMPACT) prognostic model.The median b reflects the between-country variation; a median b equal to 0 represents no variation, the larger the median b, the larger the variation.

COSTS OF TBI 2137
almost 50% were readmitted to the hospital at least once. 70 study from New Zealand showed that in the first year after trauma, patients use their general practitioner in 36% of the cases, allied health in 18% of cases, and specialized services in 14% of cases, increasing respectively with TBI severity. 20In our study, we observed that inpatient rehabilitation accounted for 19% of the total costs across all TBI severities.This is most probably an underestimated contribution to the total costs, as a previous study has shown that the need for rehabilitation services is largely unmet within the TBI population. 71We should additionally acknowledge that the long-term consequences of TBI are the drivers of the indirect costs caused by loss of productivity, disability, and reduced quality of life. 46These indirect costs are contemplated to be the largest contributors to the overall costs related to TBI, indicating that the economic impact of TBI is even higher than shown in this study.

Recommendations
Intramural costs of TBI are significant, with hospital admission being the largest contributor.Costs increased with trauma severity, male patients incurred higher costs, and cost patterns of the elderly differed from those of the overall TBI population.This knowledge about healthcare expenses could be a leading step toward more cost-efficient TBI care.Hospitalization (ICU LOS in particular), incurs the highest costs and differs among countries.Improvements in resource allocation and eventual reduction of costs could be effected by the development of admission guidelines wherein only those who would truly benefit will be admitted to the ICU, combined with special attention to gender differences in assessment of patients.A leading step toward tailored and costeffective TBI treatment, is, for example, the use of acute serum biomarkers to determine CT indication in mild TBI patients, thereby preventing unnecessary imaging. 72Additionally, discharge planning according to patient needs and preventive interventions targeting in-hospital complications are highly valuable in reducing unnecessary healthcare consumption.The long-term consequences of TBI are of substantial concern for the patient, the healthcare provider, and, eventually, society.Advanced care planning, wherein patients start early on with rehabilitation, could lead to reduction of hospitalization and better patient outcome, which will subsequently lead to a reduction of the indirect costs related to TBI.Differences in healthcare consumption between males and females should also be explored more extensively, as differences in the management of TBI could also lead to different outcomes.Conclusively, TBI patients must be considered as a distinct patient population, with targeted interventions that suit the different subgroups within TBI, in order to reduce costs.

FIG. 2 .
FIG.2.The median total intramural costs for male and female patients are plotted according to injury severity and age category.The injury severity was determined using the baseline systemic Injury Severity Score (ISS) and was categorized into three groups: ISS £16 (minor injury); ISS 17-25 (major injury); ISS >25 (critical injury).The four panels represent the four different age categories: (A) 16-25 years, (B) 26-40 years, (C) 41-64 years, and (D) ‡ 65 years.

Table 1 .
Baseline Characteristics of Patients According to Trauma Severity A total of 157 patients were missing information on the baseline Glasgow Coma Scale score.The p value assesses the null hypothesis of no differences among the mild, moderate, and severe subgroups.IQR, interquartile range; AIS, Abbreviated Injury Score; ISS, Injury Severity Score; CT, computed tomography; GOSE, Glasgow Outcome Scale Extended.

Table 2 .
Median Intramural Costs for Each Cost Category According to Baseline Characteristics

Table 3 .
Median Intramural Costs of Traumatic Brain Injury (TBI) According to Trauma Severity ICU, intensive care unit; IQR, interquartile range; SD, standard deviation; TBI, traumatic brain injury; AIS, Abbreviated Injury Score; ISS, Injury Severity Score; CT, computed tomography; GOSE, Glasgow Outcome Scale Extended

Table 4 .
Associations With Total Healthcare Costs Based on Generalized Linear Models CI, confidence interval; TBI, traumatic brain injury; AIS, Abbreviated Injury Score; ISS, Injury Severity Score; CT, computed tomography.