Development of the Poor Validity Profile Analysis for the Medical Symptom Validity Test.

Résumé

OBJECTIVES: The Medical Symptom Validity Test (MSVT) offers a profile analysis for performance validity, but individuals performing with suboptimal performance validity are frequently misidentified by current profile rules. The aim of this study was to develop and validate a new profile analysis for the MSVT.METHOD: This study utilized 2 phases to investigate current profile analyses and construct a new analysis. Phase I compared graduate students and adult volunteers asked to simulate dementia to individuals with dementia in long-term care facilities. Phase I included development of a logistic equation with a logit model correction and introduced 4 new profiles. Phase II investigated the 4 proposed profiles using a new sample of graduate students asked to simulate dementia and outpatient dementia referrals.RESULTS: The 4 proposed profiles showed high specificity (94%-96%) and acceptable sensitivity (79.5%-95.5%) in Phase I. In Phase II, 1 profile out performed others by demonstrating 95.7% specificity and 86% sensitivity. High sensitivity (84%) and specificity (96%) were obtained for aggregate samples using this profile.CONCLUSIONS: This research provides an alternative profile analysis for the MSVT that outperforms the available analysis in identifying poor performance validity from genuine impairment.