Background: Quantitative sensory testing (QST) conventionally identifies threshold elevation as reflecting sensory deficit. A major disadvantage of the technique is its inability to distinguish organic from feigned sensory deficit, as both are characterized by an elevated threshold.

Aim: To distinguish organic from feigned sensory deficit.

Method: Vibratory thresholds and their variances were measured, at foot L4, L5 and S1 sites, in 14 patients with low back pain (LBP) suspected of non-organic sensory loss by clinical criteria of Waddell, 14 patient controls with abnormal neurological examination and CT of the low back, and 20 healthy controls.

Results: Thresholds of non-organic patients and of patient controls were elevated to a similar extent compared to those of healthy controls. Variances, however, were higher for non-organic patients (6.7–10.5 for the various test sites) than for either patient controls (0.39–0.80, p: 0.001–0.05) or for healthy controls (0.20–0.54, p: 0.001–0.02). Of non-organic patients with high thresholds who would otherwise be identified as pathological, 30–67% were ‘spared’ the misdiagnosis and correctly identified as non-organic because of the inclusion of the variance criterion.

Conclusion: Variance evaluation is therefore suggested for inclusion into QST methodology, together with threshold itself, as a quality assurance parameter.