The development of a German language (Berne) pain questionnaire and its application in a situation causing acute pain

Abstract

The adjectives used in the McGill Pain Questionnaire were translated into German. Nine of the 76 adjectives could not be translated satisfactorily. Accordingly, 10 new German adjectives were added by 22 physicians and psychologists, who were also asked to judge the grouping of the words and how adequately they express pain. Concordance of grouping was reached by more than 82% of the raters. The adjectives were assigned to 20 groups. In a next step 80 subjects assessed the words in each group on a visual analogue scale (VAS) with respect to intensity; for 3 of the 20 groups we found no difference in intensity between the adjectives within a group. They were not retained in the final version tested. Additionally each group was reduced to 3 adjectives by dropping the least discriminating words of each group; 17 groups of 3 words each were retained. In a third step a comparably composed of group 82 subjects rated the words in each group on VAS with respect to intensity; for each of the 17 triads there was a significant overall difference between the 3 adjectives on the VAS. In all but 4 of the 17 triads all 3 possible pairwise differences were significant as well. In spite of the significant differences of mean values there was considerable disagreement in individual intensity rankings of the 3 adjectives within the 17 group. The validity of the German language (Berne) pain questionnaire (BPQ) was tested together with a verbal rating scale (VRS) and a VAS in a double-blind, complete cross-over study. A low osmolar (LO), a high osmolar (conventional) (H) compound and 0.9% NaCl (placebo) (PL) were injected intra-arterially in patients with arteriovascular disease. We expected the LO compound to cause much less pain than the HO compound. The 3 conditions HO, LO and PL had significantly different effects on all 4 scales: VAS, VRS, RaW (scale of rank of words), and NoW (scale of number of words). The pairwise comparisons of PL with HO and LO resulted in significant differences also. The square of the correlation r for VAS/VRS was 0.64, for RaW/NoW 0.61, all other possible correlations were not significant. Our conclusions are: the translation of the MPQ into German was successful, and its validity could be shown. In our experiment, it differentiated acute, short lasting pain as well as the visual analogue and the verbal rating scales. The BPQ is a tool worth being examined in the laboratory and in studies of clinical pain syndromes.