Detecting malingered ADHD using the personality assessment inventory : an exploratory analysis in college students

Abstract

Background: There has been a growing interest in assessment of effort during psychoeducational evaluations, where students may feign symptoms of ADHD to obtain academic accommodations or stimulant medications. Current research suggests most ADHD questionnaires and neuropsychological tests do not adequately distinguish clinical ADHD from simulated ADHD. Objective: The purpose of the current study is to develop an embedded malingering index in the Personality Assessment Inventory (PAI) specifically for detecting feigned ADHD in college students. Method: A sample of 310 undergraduate students were separated into three groups, ADHD Simulators, Prospective ADHD, and College controls. In addition, this study used archival data from individuals diagnosed with Clinical ADHD, No Diagnosis, Psychopathology, Comorbid ADHD-Psychopathology, and Suspect Effort. ADHD Simulators’ scores on the items of the Personality Assessment Inventory were compared to the Clinical ADHD group. The item pool was narrowed by selecting the 40 items with the largest effect sizes. A discriminant function analysis was then used to select the items that discriminate best between the two groups. The items were weighted and summed into a scale. Next, logistic regression analyses and ROC curves were used to determine an appropriate cutoff score. Results: Fourteen items were summed into a scale. When various cutoff scores were examined, a score of > 16 yielded specificity of .95 and .96 for the Clinical ADHD groups and .98 for the No Diagnosis group and sensitivity of .64 for the ADHD Simulator groups. However, it did not yield adequate specificity for Psychopathology or Comorbid ADHD-Psychopathology groups. A cutoff score of greater than > 22 yielded specificity > .90 for all groups but sensitivity of .44 for the ADHD Simulators. Conclusion: The use of a cutoff score of > 16 is recommended when individuals do not complain of comorbid psychopathology but a cutoff score of > 22 when comorbid psychopathology is in question. The newly developed scale of the PAI shows promise in identifying college students malingering ADHD symptoms.