Utility of the Neurobehavioral Symptom Inventory Validity-10 index to detect symptom exaggeration: An analogue simulation study

Abstract

ABSTRACTThe Neurobehavioral Symptom Inventory (NSI) has been recommended by the interagency Traumatic Brain Injury (TBI) Outcome Workgroup as an outcome measure for TBI research. A new symptom exaggeration index—the NSI Validity-10—can be calculated from its items, but its utility has not been evaluated in a malingering simulation study. Data from a prior analogue study were reanalyzed to examine the NSI Validity-10 test properties. The data were from a sample of 85 Australian undergraduate students. A battery of measures was completed under 1 of 3 experimental conditions: control (i.e., honest responding, n = 24), feign postconcussional disorder (PCD; n = 29), and feign posttraumatic stress disorder (PTSD; n = 32). Participants who feigned PTSD or PCD had significantly higher scores on the NSI Validity-10 compared with controls. There were minimal differences between the 2 feigning groups. Using the combined data from the feigning groups and assuming a 35% symptom exaggeration base rate, the optimal NSI Validity-10 cutoff score was ≥10. This cutoff score identified “probable exaggeration” (sensitivity = .75, specificity = 1.0, positive predictive power = 1.0, negative predictive power = .88). Diagnostic efficiency statistics for 25% and 45% base rates were also generated. The cutoff score identified in this study is lower than previously reported. Its properties are promising, but its usage requires careful consideration.