Test-retest studies in quantitative sensory testing: a critical review

Abstract

Quantitative sensory testing (QST) investigates the graded psychophysical response to controlled thermal, mechanical, electrical or chemical stimuli, allowing quantification of clinically relevant perception and pain thresholds. The methods are ubiquitously used in experimental and clinical pain research, and therefore, the need for uniform assessment procedures has been emphasised. However, varying consistency and transparency in the statistical methodology seem to occur in the QST literature. Sixteen publications, evaluating aspects of QST variability, from 2010 to 2012, were critically reviewed in detail. A considerable heterogeneity in the statistical evaluations of test-retest data was demonstrated. The authors, using a secondary analysis of published data for didactic purposes, propose and present minimal requirements for reporting of test-retest QST data.

Test-retest Agreement and Reliability of Quantitative Sensory Testing 1 Year After Breast Cancer Surgery

“…21 Our data show a large variability in agreement across all modalities assessed by Bland-Altman plots, which is comparable to other studies in postsurgical patients. 12 The minimal detectable change is analogous to the limits of agreement used in the Bland-Altman plots, and shows that generally a change of temperature threshold of 5 to 13 is needed to be 95% certain that a real change has occurred on the pathologic side.…”

Reliability of Quantitative Sensory Tests in a Low Back Pain Population

“…The reasons for choosing the CR is the fact that it measures absolute reliability in the unit of the original measure and its derivation from the Bland-Altman analysis, which is the current reference in comparison studies. 8,9 Although arbitrary reliability thresholds have been reported for ICC and CV, these measures have raised serious concerns. 2,12 For example, if the ICC is applied to values with a large variation, the ICC and reliability will appear higher than when applied to values with a narrower range with the same characteristics.…”

“…The standard error of measurement (SEM), the coefficient of variation (CV), and the coefficient of repeatability (CR) are examples of absolute reliability indexes. 2,8,9 Comprehensive QST studies assessing absolute and relative reliability in patients with chronic low back pain have not been published to date. Thus, the aim of the present study was to assess the test-retest reliability of multiple QST measures in patients with chronic low back pain.…”

Quantitative Sensory Testing in Pain Management

“…In one study, QST assessments in healthy volunteers were made in duplicate sessions separated by 21 days; some gender-related differences in pain perception were evident [56]. In a 2013 review of 16 test-retest studies from 2010 to 2012, which primarily involved healthy volunteers, Werner et al concluded that 80% of the variation in the retest variables could not be explained by the test variables and that this finding indicated a low strength of association [57]. Similarly, in 2015, Hafner et al studied thermal thresholds in 101 healthy volunteers (21-70 years old) using two different thermodes several testers and a test-retest data collection format [58].…”