A study of psychophysical scaling in chronic pain patients.
Chronic pain patients were asked to psychophysically scale two sets of word descriptors (intensity and unpleasantness) using a crossmodality matching procedure with line length and numerical estimates. In 22 patients (group I) descriptor scaling was performed once, in another 20 patients (group II) the procedure was repeated 3 times. 72.7% of patients in group I obtained a correlation of 0.9 or higher when scaling intensity descriptors, but only 18.2% obtained this correlation when scaling unpleasantness words. In group II, an average of 85% of patients reliably scaled intensity words, but only 50% could do so for the unpleasantness descriptors. Patients who reliably judged unpleasantness descriptors generally exhibited a higher level of psychological distress (SCL-90 R). Numerical estimates assigned by patients to individual word descriptors showed a smaller range than obtained from experimental pain studies. There was unequal spacing of values of adjacent descriptors with clustering of words in the low and high ends of the continuum and large gaps in midrange. Results indicate that chronic pain patients in general show a higher incidence of impairment in rendering proportional judgments than a healthy population. In addition, they are more likely to judge pain reliably on an intensity than on an unpleasantness dimension. The difference in performance between scales persists in spite of training and could not be explained by medical history, drug intake, or demographic characteristics. Patients who are unable to render reliable judgments are easily identified. For optimal clinical use words clustering close together should be combined and additional words should be added describing midrange intensities. Conversely, psychophysical scaling techniques may be used to calibrate category or analogue scales of pain.
Urban, BJ; Keefe, FJ; France, RD
Volume / Issue
Start / End Page
Pubmed Central ID
International Standard Serial Number (ISSN)
Digital Object Identifier (DOI)