Investigating the M-FAST: psychometric properties and utility to detect diagnostic specific malingering

Abstract

This study examined the ability of the M-FAST to differentiate a group of undergraduate students simulating one of four DSM-IV diagnoses (n = 190; schizophrenia, major depressive disorder, bipolar disorder, and posttraumatic stress disorder) and a clinical comparison sample drawn from previous M-FAST studies comprising individuals with the same diagnosis (n = 142). Across all diagnostic conditions, the simulators obtained higher M-FAST total scores than the clinical comparisons, and the rare combinations scale was equal or superior to the total score at differentiating the groups. The M-FAST was most efficient at distinguishing feigned from bona fide schizophrenia. Although the internal consistency of the total score was high (alpha = 0.88), inter-item correlations were lower than values reported in previous research. Lastly, given the importance of base rate considerations in the evaluation of diagnostic instruments, it was notable that the M-FAST was able to identify malingerers even at relatively low base rates.