CENTER-TBI: Collaborative European NeuroTrauma Effectiveness Research in TBI

Collaborative European NeuroTrauma Effectiveness Research in TBI (CENTER-TBI)

The research aims of the CENTER-TBI study are to:

1. better characterize Traumatic Brain Injury (TBI) as a disease and describe it in a European context, and
2. identify the most effective clinical interventions for managing TBI.

Specific aims

1. To collect high quality clinical and epidemiological data with repositories for neuro-imaging, DNA, and serum from patients with TBI.
2. To refine and improve outcome assessment and develop health utility indices for TBI.
3. To develop multidimensional approaches to characterisation and prediction of TBI.
4. To define patient profiles which predict efficacy of specific interventions ("Precision Medicine").
5. To develop performance indicators for quality assurance and quality improvement in TBI care.
6. To validate the common data elements (CDEs) for broader use in international settings, and to develop a user-friendly web based data entry instrument and case report form builder.
7. To develop an open database compatible with Federal Interagency Traumatic Brain Injury Research (FITBIR).
8. To intensify networking activities and international collaborations in TBI.
9. To disseminate study results and management recommendations for TBI to health care professionals, policy makers and consumers, aiming to improve health care for TBI at individual and population levels.
10. To develop a "knowledge commons" for TBI, integrating CENTER-TBI outputs into systematic reviews.

• Introduction:

CENTER-TBI (Collaborative European NeuroTrauma Effectiveness Research in TBI) (www.center-tbi.eu) is a project embedded within the International Initiative on TBI Research (InTBIR) (http://intbir.nih.gov/), as a collaboration between the European Commission (EC), the US National Institute of Neurological Disorders and Stroke (NIH-NINDS) and the Canadian Institute of Health Research (CIHR).

• Research aims:

The basic concept of this project is to exploit the existing heterogeneity in biology, care and outcome of TBI patients to discover underlying pathophysiology, to refine characterisation, and to identify effective clinical interventions. The key driver of our research plan is to collect data from a large number of European centres and sufficiently large cohort to enable 'Comparative effectiveness research' (CER) analyses of differences in clinical care and management pathways in TBI.

Improved disease characterization will aid Precision Medicine, a concept recently enunciated by the US National Academy of Science. Such improved characterization and stratification will allow for more targeted therapies. Further, CER provides a promising framework to identify best practices and improve outcome after TBI. CER is the generation and synthesis of evidence that compares the benefits and harms of alternative methods to prevent, diagnose, treat, and monitor a clinical condition or to improve the delivery of care. The purpose of CER is to assist consumers, clinicians, purchasers, and policy makers to make informed decisions that will improve health care at both the individual and population levels.

• Cohort:

A large core cohort of patients across the severity spectrum in TBI will be recruited at approximately 77 sites in Europe and Israel (the CENTER-TBI Core Study): 5400 patients differentiated into 3 equal strata of approximately 1800:

• ER stratum: patients seen and discharged from the ER;
• Admission stratum: patients admitted to the hospital but not to the ICU;
• ICU stratum: patients admitted directly to the ICU. Balance in numbers between the strata will be aimed for, but sites will be allowed to arrange recruitment strategies to best suit their local requirements.

The core cohort will be underpinned by comparison with a larger registry (the CENTER-TBI registry) based on pragmatic data collection of all patients with TBI seen in participating centres (to establish the internal generalizability of our study), and by comparison with national trauma registries (to establish the external generalizability of our findings).

In the core cohort, detailed data will be collected on clinical parameters, neuroimaging studies, biomarker analyses, DNA analyses, and longitudinal outcome assessments. In selected centres, extended studies will additionally focus on advanced magnetic resonance (MR) imaging, detailed coagulation profiling and high resolution ICU monitoring.

• Sample size assessment:

The sample size estimate (n=5400) was based on:

• Practical logistic considerations
• Power calculations for the different strata, targeting comparative effectiveness analyses, assuming a between-centre and between-country heterogeneity as identified in previous research (expressed by variance parameter from a random effects model, Tau^2 of 0.431)
• Postulated odds ratios for intervention effects of approximately 5% improvement in outcome.

Overall, these calculations provided a statistical power to detect odds ratios of ~1.2 associated with differences in process or intervention variables across the core dataset with a power of 80%; and require somewhat larger odds ratios in each of the three individual strata. In the registry we expect to be able to detect differences (predominantly in organizational or system variables) with an odds ratio of 1.2 with a power of 82%.

• Quality control and assurance:

Continuous monitoring of enrolment and completeness of data will be performed by ICON as contract research organization (CRO). Source data verification (SDV) will be performed in 10% of subjects by the CRO. The quality of data collection will be further enhanced by implementing automated data entry checks for impossible/implausible values and implementing data checks for consistency between variables. A task force of study personnel will evaluate the completeness, consistency and validity of submitted data and function as support desk for participating sites.

• Data management:

Prior to upload to the study database, all acquired data will be stored locally. All patients will be allocated a random Global Unique Personal Identification number (GUPI) which will be linked locally to hospital identifiers. All uploaded data will be de-identified and images will be defaced prior to upload. While blood samples and clinical data will be linked, both sets of data will be kept confidential and anonymised beyond the initial stage of correlation for analysis. All imaging and electronic data will be kept on individually password protected servers. All clinical data will be entered into electronic Case Report Forms (eCRFs) and managed by the 'QuesGen data management platform' (http://www.quesgen.com/) which will be developed in collaboration with Karolinska Institutet International Neuroinformatics Coordinating Faculty (KI-INCF). As data is entered into each form, the system will run data validation checks that include conditionally required data, validation across fields, and validation requirements based on subject type. If any validation check fails, the user is alerted immediately that the data does not meet Quality Assurance (QA) criteria and the issue can be addressed and corrected at that point. All de-identified electronic study data in the CENTER-TBI database will be stored securely in the European data space under supervision of KI-INCF for the duration of subject enrolment and follow-up and for a period afterwards for data analysis and preparation of publications. We estimate that the analysis and publication period will last for several years after the conclusion of subject enrolment.

Together with QuesGen Systems, KI-INCF will ensure that data standards are established for the data model e.g. conformity of field formats, field codes and names to ensure consistency across all datasets. Any approved changes will be fully documented with dataset updates to maintain data quality and accuracy. KI-INCF will be responsible for importing cleaned datasets to other analytic platforms as determined by the coordinators.

Where applicable, information relevant to the patient's care will be made available to the physician responsible. Data, including blood samples collected as part of this study will be shared in an anonymised form with collaborators from other European states (this is part of a European Commission Framework7 funded program), and with selected collaborators in other countries who form part of an emerging International Traumatic Brain Injury Research initiative.

• Computing platform and Neuroinformatics Resource:

The KI-INCF will coordinate the establishment of an informatics platform for acquisition, storage and analysis of CDE-based clinical data. The goal is to develop a next generation open standards-based platform to support advanced large-scale analytics and model building. Such a platform also provides a model for future clinical studies on brain diseases and disorders. This development will receive additional support from One Mind for Research.

• Statistical analysis plan:

Statistical analyses for the Comparative Effectiveness Research (CER) questions will primarily apply random effects modelling, in which center is included at the higher level, and patients are considered clustered within centers. In some analyses, higher levels of clustering will also be considered, e.g. country, or European region; or lower levels, e.g. physicians within hospitals. Confounding factors as measured at the individual patient and/or center level, will be considered extensively, and will be targeted to the specific research question.

Statistical analyses for better characterization of TBI will be exploratory, aiming to better understand the complexity of the disease and to discover new associations. In addition to standard statistical descriptive and inferential techniques, we will also employ novel machine learning techniques as appropriate.

Prognostic analyses will consider a range of variables, including genetic, demographic and clinical data, physiological signals, imaging results, and biomarkers as predictors of early endpoints and physiologic derangement (e.g. raised ICP), and late outcome, including mortality, functional outcome, quality of life and neuropsychological performance. Previously and newly developed prediction models will be validated by comparison of observed to predicted outcome risks, with predictive performance summarized by measures for model fit, discrimination, and calibration.

• Missing data:

Every effort will be made to limit the number of missing data in the CENTER-TBI study. Missing values are, however, inherent to any clinical study. Missing values may confound the descriptive, prognostic and CER analyses. Thus, appropriate techniques for dealing with missing values are required. First, we will evaluate the various reasons for missingness per centre. Next, we will explore the use of alternative statistical approaches, including inverse probability weighting and multiple imputation. We will consider missing values in baseline characteristics as well as in short and long term outcomes. Based on simulation studies and practical considerations we will develop standard operating procedures towards the analyses with missing values, respecting the differences in multiple research questions.

• Informed Consent:

Informed consent procedures will follow local and national requirements in all cases. We anticipate that many potential patients will not be able to consent themselves to participate in this project. The nature of TBI means that some patients may lack capacity to decide to participate in this study especially at the earliest time point. It is important to try and include these patients to ensure that representative samples of patients are included to avoid bias in the study findings. Every step will be taken to ensure that a test of capacity is undertaken before a decision on a person's capacity to consent or not to consent to participation in research is taken. If the subject is not capable of self-consent, all efforts will be made to locate a legally acceptable representative to act on behalf of the subject. When a legally acceptable representative (e.g. consultee/proxy) is identified, their opinion will be sought about the potential participant's wishes and feelings in relation to the project, and whether he or she would have wanted to take part in the study.

Subjects are free to withdraw, or be withdrawn by their consultee/proxy if appropriate, at any point in the study, and they need not state a reason.

• Impact:

The CENTER-TBI project will contribute towards the overall goals of InTBIR, by identifying more effective and efficient treatment provision, thus improving outcome and reducing costs. The science in the project will provide novel information on disease processes, treatment, outcome, and prognosis in TBI, identifying new therapeutic targets and therapies; while the CENTER-TBI repositories will ensure opportunities for legacy research. Thus, the project has the potential to improve current health care and its delivery at both population and individual levels, deliver early scientific advances that could improve the care of patients with TBI, and provide a rich investment for future biomedical research.

Inclusion Criteria:

• Clinical diagnosis of TBI
• Clinical indication for CT scan
• Presentation within 24 hours of injury
• Informed consent obtained according to local and national requirements

Exclusion Criteria:

• Severe pre-existing neurological disorder that would confound outcome assessments