Abstract
This archival study examined the effectiveness of using a normative floor effect method to detect suboptimal performance in a clinical sample of 120 cases consecutively referred for a neuropsychological evaluation. These cases were divided into moderate-severe TBI, mild TBI, and poor effort litigants (demonstrating suboptimal effort). Percentiles, sensitivity, specificity, positive and negative predictive power, and overall correct classification rates were calculated using the moderate-severe TBI sample as the clinical reference group. Several levels of stringency for the floor effect using different base rates were also examined. Setting the floor at the 50th percentile consistently appeared to provide for the best overall hit rate when comparing cases with mild brain injury versus poor effort. At this level, a strong trend (greater than one third of total scores) toward the generation of positive scores offers compelling evidence that, in general, the neuropsychological data have been invalidated by poor motivation. Clinical implications are discussed.
This study was supported, in part, by a grant from the US Department of Education-National Institute of Disability Research and Rehabilitation – The Traumatic Brain Injury Model Systems Project (H133A020515-03)
Notes
a < .01.
b < .05.
aAdjusted scores.
aFloor level > 50th percentile
bFloor level > 75th percentile
cFloor level > 90th percentile.
Sen = sensitivity; Spec = specificity; PPP = positive predictive power; NPP = negative predictive power; OCC = overall correct classification rate; PM = probability of malingering.
aFloor > 50th percentile for Trail Making Test and Temporal Orientation Test.
bFloor > 75th percentile for Trail Making Test and Temporal Orientation Test.
(+): positive score for floor set at 50th percentile, (++): positive score for 2 floor levels (50th and 25th/75th percentiles), (+++): positive score for all 3 floors levels (50th, 25th/75th and 10th/90th percentiles).
aAdjusted score; (−): negative score relative to all 3 floor levels.