Examining the cross-cultural validity of the english short-form McGill Pain Questionnaire using the matched moderated regression methodology.

OBJECTIVES: Pain is identified as the third largest health problem in the world, impacting sleep, physical functioning, and psychologic functioning. Pain has been viewed as a multidimensional experience influenced by biology, psychology, cultural conditioning, expectancies, and social contingencies. The aims of this study were to examine the psychometric properties and the differential item functioning (DIF) of the short-form McGill Pain Questionnaire (SF-MPQ) to test the cultural differences in verbally based pain expression.
METHODS: Data were drawn from the medical records of 134 Hispanic and 134 non-Hispanic White patients with pain. Cases were matched on total SF-MPQ score, sex, and age. Estimates of reliability and validity were assessed. Using a traditional moderated regression approach, each of the 15 items on the SF-MPQ was analyzed for DIF using matched moderated regression methodology.
RESULTS: Internal consistency was high: Hispanics 0.90 and non-Hispanics 0.89. Factor analysis revealed 2 components for each ethnic group. DIF analyses revealed statistically significant group differences for the 5 items (ie, throbbing, gnawing, aching, tiring-exhausting, and sickening) potentially due to cultural-linguistic factors. The remaining 10 items revealed no significant differences. Group differences were not accounted for by sex, age, marital status, education, pain location, or pain duration. DISCUSSION: Test bias was relatively small (0.74) given the entire scale of the SF-MPQ (0 to 45), indicating that this measure seems to be used equivalently across these 2 groups. This study shows 1 method in evaluating the cross-cultural validity of pain assessment instruments, and contributes to the understanding of cross-cultural variability in item reporting on the SF-MPQ.