AUS-TBI: The Australian Health Informatics Approach to Predict Outcomes and Monitor Intervention Efficacy after Moderate-to-Severe Traumatic Brain Injury

Melinda Fitzgerald, Jennie Ponsford, Natasha A. Lannin, Terence J. O’Brien, Peter Cameron, D. James Cooper, Nick Rushworth, Belinda J. GabbeNeurotrauma 2022 · DOI

TL;DR

This paper introduces AUS-TBI, a novel Australian initiative leveraging health informatics to create a comprehensive data resource for predicting outcomes and monitoring interventions in moderate-to-severe traumatic brain injury (TBI). By integrating diverse data types and employing a codesigned, culturally safe approach, AUS-TBI aims to enable personalized TBI care and advance research globally.

What Problem Does This Paper Address?

Traumatic Brain Injury (TBI) is a significant global health challenge, leading to diverse and complex outcomes that are difficult to predict and manage. Current approaches to TBI care and research are hampered by fragmented data systems, a lack of standardized data collection, and an incomplete understanding of the myriad factors influencing recovery. This heterogeneity makes it challenging to personalize treatment and develop effective interventions. Existing predictive models, while useful, often lack the sensitivity and specificity needed for individual patient care and may not adequately represent diverse populations, such as Aboriginal and Torres Strait Islander peoples. The paper highlights the need for a unified, comprehensive data resource to address these gaps and improve TBI outcomes.

Methods

The AUS-TBI initiative proposes a multi-faceted health informatics approach. Key methodological aspects include:

  • Data Integration: A core aim is to bring together a wide array of data, including psychosocial modulators, social determinants, clinical parameters, imaging data, biomarker profiles, and rehabilitation outcomes.
  • Data Management Systems: Designing systems for secure data collection, linkage, storage, curation, management, and analysis.
  • Consensus-Based Data Element Identification: A systematic process involving expert researchers, clinicians, and individuals with lived experience of TBI to identify the most relevant and feasible common data elements. This includes systematic literature reviews and stakeholder engagement.
  • Codesign and Cultural Safety: Emphasizing codesign with consumers and stakeholders, particularly Aboriginal and Torres Strait Islander persons, to ensure the approach is culturally safe and the outcomes are valuable to those with lived experience.
  • FAIR Data Principles: Adhering to the Findable, Accessible, Interoperable, and Reusable (FAIR) data principles to maximize data utility and accessibility.
  • Privacy-Preserving Linkage: Utilizing innovative linkage methods to connect TBI data with routinely collected administrative data, enabling a comprehensive view of a patient's journey across health and social services.
  • Machine Learning: Employing machine learning to develop accurate prediction models for TBI outcomes.

Datasets

The paper does not detail specific datasets used for training or evaluation in this foundational proposal. Instead, it outlines the plan to create a new, large-scale, nationally representative data resource for Australia. This resource will be built by collecting data from individuals who have experienced moderate-to-severe TBI across all Australian states and territories. The initiative aims to integrate this data with existing multi-sectoral administrative data.

Key Insights and Ideas

  • Holistic Data Integration: The recognition that TBI outcomes are influenced by a complex interplay of factors beyond just clinical injury parameters, necessitating the inclusion of psychosocial, social, and environmental data.
  • Patient-Centered Approach: The strong emphasis on codesigning the data elements and approach with individuals with lived experience of TBI, ensuring the research is relevant and valuable to those most affected.
  • Addressing Health Inequalities: A specific focus on including and engaging Aboriginal and Torres Strait Islander persons, acknowledging the disproportionate impact of injuries on these populations and the need for culturally appropriate data collection.
  • Bridging Service Gaps: The potential for the data resource to bridge the gap between child and adult services by integrating data across different care domains.
  • Foundation for Personalized Medicine: The vision of creating a data resource that moves beyond generalized predictions to enable highly personalized prognostication and treatment strategies for TBI patients.

Key Results

As this paper describes the establishment of the AUS-TBI initiative and its proposed approach, it does not present quantitative results from data analysis or model performance. The "results" are the planned outcomes of the initiative itself:

  • Development of a comprehensive health informatics approach for TBI data collection and management.
  • Identification of key common data elements that are feasible, reliable, and relevant for predicting TBI outcomes.
  • Creation of a large-scale, nationally representative data resource for Australia.
  • Facilitation of personalized TBI care through improved prediction and monitoring.
  • Advancement of TBI research by providing a robust data foundation for future studies and interventions.

Limitations and Future Work

Limitations:

  • Scope of Data Elements: The paper acknowledges that the focus on predicting outcomes may limit the inclusion of exploratory biomarker analyses or other avenues of investigation.
  • Feasibility of Data Collection: The common data elements will be limited to those that are feasible to collect from all participating institutions, which could further constrain the scope.
  • Representativeness: While aiming for national representation, the paper notes that existing data sets have not adequately included populations with disproportionate representation, such as Aboriginal and Torres Strait Islander persons, a gap AUS-TBI aims to address.

Future Work:

  • Implementation of the Data Management System: The ongoing work will involve designing and implementing the health informatics framework.
  • Development of Predictive Models: Utilizing the collected data to develop and validate machine learning-based prediction models.
  • Integration with Multi-Site Trials: Incorporating multi-site intervention trials into the data collection and informatics system to support streamlined and efficient clinical trials.
  • Global Contribution: The project aims to create a blueprint for a data resource that can facilitate world-leading interdisciplinary TBI research and potentially improve TBI treatment globally.
  • Addressing Health Inequalities: Continued engagement with the Aboriginal and Torres Strait Islander Expert Advisory Committee to ensure accuracy and cultural appropriateness for these vulnerable populations.

Summary updated

Abstract

Predicting and optimizing outcomes after traumatic brain injury (TBI) remains a major challenge because of the breadth of injury characteristics and complexity of brain responses. AUS-TBI is a new Australian Government-funded initiative that aims to improve personalized care and treatment for children and adults who have sustained a TBI. The AUS-TBI team aims to address a number of key knowledge gaps, by designing an approach to bring together data describing psychosocial modulators, social determinants, clinical parameters, imaging data, biomarker profiles, and rehabilitation outcomes in order to assess the influence that they have on long-term outcome. Data management systems will be designed to track a broad range of suitable potential indicators and outcomes, which will be organized to facilitate secure data collection, linkage, storage, curation, management, and analysis. It is believed that these objectives are achievable because of our consortium of highly committed national and international leaders, expert committees, and partner organizations in TBI and health informatics. It is anticipated that the resulting large-scale data resource will facilitate personalization, prediction, and improvement of outcomes post-TBI.
BibTeX
@inproceedings{fitzgerald2022aus,
  author = {Melinda Fitzgerald and Jennie Ponsford and Natasha A. Lannin and Terence J. O’Brien and Peter Cameron and D. James Cooper and Nick Rushworth and Belinda J. Gabbe and Belinda Gabbe},
  title = {{AUS-TBI: The Australian Health Informatics Approach to Predict Outcomes and Monitor Intervention Efficacy after Moderate-to-Severe Traumatic Brain Injury}},
  booktitle = {Neurotrauma},
  year = {2022},
  doi = {10.1089/neur.2022.0002},
}
Figure