Validity of the Groningen Effort Test in patients with suspected chronic solvent-induced encephalopathy.

Abstract

The use of performance validity tests (PVTs) in a neuropsychological assessment to determine indications of invalid performance has been a common practice for over a decade. Most PVTs are memory-based; therefore, the Groningen Effort Test (GET), a non-memory-based PVT, has been developed. This study aimed to validate the GET in patients with suspected chronic solvent-induced encephalopathy (CSE) using the criterion standard of 2PVTs. A second goal was to determine diagnostic accuracy for GET. Sixty patients with suspected CSE referred for NPA were included. The GET was compared to the criterion standard of 2PVTs based on the Test of Memory Malingering and the Amsterdam Short Term Memory Test. The frequency of invalid performance using the GET was significantly higher compared to the criterion of 2PVTs (51.7% vs. 20.0% respectively; p < 0.001). For the GET index, the sensitivity was 75% and the specificity was 54%, with a Youden's Index of 27. The GET showed significantly more invalid performance compared to the 2PVTs criterion suggesting a high number of false positives. The general accepted minimum norm of specificity for PVTs of >90% was not met. Therefore, the GET is of limited use in clinical practice with suspected CSE patients.