The Mc Gill Pain Questionnaire – MPQ : Linguistic and Statistical Consideration

The McGill Pain Questionnaire (Melzack, 1975) is the most widely used pain

assessment instrument in the world. The MPQ comprises 20 subcategories of verbal (single-word) pain descriptors (e.g., cutting, crushing, burning, stinging) intended to measure the sensory, affective, and evaluative components of pain. Pain descriptors are ranked according to intensity within each subcategory. Six of the 84 items are incorporated into a Present Pain Index (PPI) which measures overall pain intensity. Nevertheless, despite its international popularity (translated into at least 15 different languages, including Spanish, French and Portugese), the MPQ has some severe psychometric limitations that undermine its validity. For instance, scoring systems remain psychometrically complicated, raising doubts about the reliability of subcategory scores. In addition, several verbal descriptors pertaining to the evaluative component of pain really describe affective aspects. Also, some descriptors (e.g., lancinating and rasping) are not easily understood by many pain patients. Problems also arise from unequal numbers of descriptors within subclasses, the discrepant number of subclasses within each component, variability in ranks and interval ratings of pain descriptors, ambiguity of some descriptors (e.g., flashing pertains to brightness

as well as to frequency), and the non-equivalence of translations of descriptors.

Compared with the sensory domain, the proportion of variance accounted for by the affective domain is no more than 50%, while the evaluative domain accounts for a mere 10% of variance. Instability of the MPQ factor structure is problematic, and reliance on only the tripartite (sensory, affective, and evaluative) domains may lead to loss of diagnostic information due to restriction of variance associated with possible underfactoring. Indeed, several studies have suggested that extensive restructuring of the MPQ instrument is needed. Accordingly, some positive suggestions based on empirical findings with the MPQ are offered in an attempt to refine what is undoubtedly a valuable and time-honoured pain assessment instrument.