Experts’ Failure to Consider the Negative Predictive Power of Symptom Validity Tests

Abstract

Feigning (i.e., grossly exaggerating or fabricating) symptoms distorts diagnostic evaluations. Therefore, dedicated tools known as symptom validity tests (SVTs) have been developed to help clinicians differentiate feigned from genuine symptom presentations. While a deviant SVT score is an indicator of a feigned symptom presentation, a non-deviant score provides support for the hypothesis that the symptom presentation is valid. Ideally, non-deviant SVT scores should temper suspicion of feigning even in cases where the patient fits the DSM’s stereotypical yet faulty profile of the “antisocial” feigner. Across three studies, we tested whether non-deviant SVT scores, indeed, have this corrective effect. We gave psychology students (Study 1, N = 55) and clinical experts (Study 2, N = 42; Study 3, N = 93) a case alluding to the DSM profile of feigning. In successive steps, they received information about the case, among which non-deviant SVT outcomes. After each step, participants rated how strongly they suspected feigning and how confident they were about their judgment. Both students and experts showed suspicion rates around the midpoint of the scale (i.e., 50) and did not respond to non-deviant SVT outcomes with lowered suspicion rates. In Study 4, we educated participants (i.e., psychology students, N = 92) about the shortcomings of the DSM’s antisocial typology of feigning and the importance of the negative predictive power of SVTs, after which they processed the case information. Judgments remained roughly similar to those in Studies 1–3. Taken together, our findings suggest that students and experts alike have difficulties understanding that non-deviant scores on SVTs reduce the probability of feigning as a correct differential diagnosis.