Intervention for executive functions after traumatic brain injury: A systematic review, meta-analysis and clinical recommendations

Abstract

A systematic review of studies that focused on the executive functions of problem solving, planning, organising and multitasking by adults with traumatic brain injury (TBI) was performed through 2004. Qualitative and quantitative methods were used to evaluate the 15 studies that met inclusion criteria. Demographic variables, design and intervention features, and impairment and activity/participation outcomes (ICF) (World Health Organization, 2001) were documented. Five randomised control treatment (RCT) studies used step-by-step, metacognitive strategy instruction (MSI) and outcomes were evaluated in a meta-analysis. Effect sizes (ESs) from immediate impairment outcomes after MSI and “control” intervention were similar to each other, and both were significantly larger than chance. ESs from immediate activity/participation outcomes after MSI were significantly larger than the ESs from control intervention, and both were significantly larger than chance. These results, along with positive outcomes from the other group, single-subject design and single case studies, provided sufficient evidence to make the clinical recommendation that MSI should be used with young to middle-aged adults with TBI, when improvement in everyday, functional problems is the goal (Level A) (American Academy of Neurology, 2004). Although maintenance effects were generally positive, there was insufficient data quantitatively to evaluate this. Furthermore, there was insufficient evidence to make clinical recommendations for children or older adults. Intervention that trained verbal reasoning and multi-tasking was promising, although the evidence is insufficient to make clinical recommendations at this time. Additional research needs were highlighted.

Keywords:

• Brain injury
• Executive functions
• Intervention
• Problem solving
• Systematic review

Acknowledgments

This project was funded by the Academy of Neurologic Communication Disorders and Sciences (ANCDS), with additional support from American Speech-Language-Hearing Association (ASHA), ASHA's Division 2: Neurophysiology and Neurogenic Speech and Language Disorders, the Veterans Administration, and numerous universities. The authors want to especially thank graduate students, Miriam Krause, Sarah Linhart and Sharon Spencer, in the Department of Speech-Language-Hearing Sciences at the University of Minnesota for their contributions to this project.

Notes

¹This work is from the Writing Subcommittee on Evidence-based Practice for Cognitive-Communication Disorders after TBI, established by the Academy of Neurologic Communication Disorders and Sciences (ANCDS). See www.ancds.org/practice.shtml for other publications from this and other subcommittees.

²If there was bias in this meta-analysis, it was a conservative bias. Many studies did not provide raw data or the statistical information to be able to estimate ES when the comparison did not yield statistically significant results. If the control treatment did not result in significant pre- and post-treatment differences, ES could not be estimated because of missing data. For example, Rath et al. (2003) reported greater gains in problem solving by the innovative treatment group compared to the conventional group, whereas the conventional treatment group reported greater gains in memory and attention. We could not estimate these ESs from results that were not-significant, namely the effect of problem solving from participants in the conventional treatment condition. So, the between-group difference (in ES) would have been even larger, if studies had reported the data for all comparisons whether statistically significant or not.

³The American Academy of Neurology (2004) degrades RCTs from Class I to Class II if the dropout rate is higher than 25%.