A Systematic Review of Assessments for Identifying Executive Function Impairment in Adults With Acquired Brain Injury

Abstract

Purpose To synthesize quantitative findings concerning measures of executive function (EF) in adults with acquired brain injury (ABI). Method Electronic databases were searched for studies of EF assessments in adults with ABI that reported any of 3 values: likelihood ratios (LRs), standardized group mean comparisons ( d s), or correlations ( r s) among EF tests. Forest plots were constructed for each value. Results Searches yielded 1,417 unique citations. Full texts of 129 articles were reviewed; 34 reported at least 1 value of interest. Nineteen positive and negative LRs were calculated from 8 studies of 8 EF measures; some point estimates were in the clinically informative range, but all confidence intervals extended beyond it. From 24 studies of 14 measures, 114 d values were calculated; d values for only 11 measures had lower bounds > 0.80. From 8 studies involving 10 EF measures, 104 correlations were reported; in only 5 cases were r 2 values > .5. Conclusions Strong evidence concerning diagnostic accuracy and concurrent validity of EF measures for adults with ABI is lacking. Better specification of the construct of EF as well as research aimed at improving the quality of evidence concerning EF tests are needed.