Global functional outcome in traumatic brain injury: Use in clinical trials, approaches to data collection, and role in multi-dimensional outcome assessment

Abstract

Traumatic brain injury (TBI) affects multiple aspects of health and daily functioning. However, TBI researchers whose primary interest is in the acute care setting have often used single measures of global functional outcome, such as the Glasgow Outcome Scale (GOS), to provide an overall assessment of recovery at follow-up, and have not routinely incorporated measures that capture the multi-dimensional impact of TBI. CENTER-TBI is part of an international initiative towards standardizing and refining outcome assessment in TBI. The aim of this initiative is to promote the use of common measures to provide a multi-dimensional description of TBI outcomes in a range of study contexts. This thesis uses data collected for CENTER-TBI to examine two main issues of relevance to outcome assessment in adult TBI: (1) methods of collecting information about global functional outcome; and (2) implementation of multi-dimensional approaches to outcome assessment in TBI. The systematic review in Chapter 2 examines the patterns of use and reporting quality of outcome measures in clinical trials in adult TBI. The findings from the review demonstrate heterogeneity in the use of outcome measures, limited use of multi-dimensional outcomes, and highlight the issue of incomplete reporting of outcomes in these studies, providing the impetus for the studies in Chapters 4, 5 and 6. Chapters 4 and 5 compare outcomes assigned using clinician ratings and patient reports on the extended GOS (GOSE). The two GOSE approaches were found to be broadly equivalent indicating that, in this context, patient reports generally provide information that is comparable to that obtained via clinician-rated interviews. Chapter 5 demonstrates that the GOSE has significant, but modest, associations with prognostic factors and other outcome measures. The role of the GOSE in implementing multi-dimensional outcome assessment is considered in Chapter 6. Chapter 6 demonstrates that the applicability of individual outcome assessments is strongly driven by level of disability. Thus, a tailored approach to outcome assessment is needed. The studies in this thesis indicate that mixed modes of GOSE data collection can be used to maximise follow-ups in studies with pragmatic constraints. Furthermore, outcome measures need to be carefully selected to capture the multi-dimensional impact of TBI across the spectrum of recovery. The findings have implications for further CENTER-TBI analyses, for selecting outcome measures in future prospective studies, and for pooling data for secondary analyses.