Validation of the Multidimensional Affect and Pain Survey (MAPS).

UNLABELLED: The concurrent construct validity of the Multidimensional Affect and Pain Survey (MAPS) was determined in a sample of 100 oncology outpatients. As recommended by Jensen, we examined the frequency pattern of correlations between MAPS and standard questionnaires that reached a criterion level of significance. As predicted, the 17 subclusters in the MAPS Somatosensory Pain supercluster showed a higher mean frequency of criterion correlations with the 4 sensory-related groups of items from the MPQ and, equally important, a lower frequency of criterion correlations with unrelated constructs, POMS, and the MPQ Affective Class. The 8 subclusters in the MAPS Emotional Pain supercluster revealed a high frequency of correlations with related POMS and FACT-G scales and (negatively) with the KPS score; equally important, there were far fewer criterion correlations of these scales with unrelated MPQ sensory classes. The 5 subclusters of the MAPS Well-Being supercluster showed a high frequency of criterion correlations with the POMS Vigor/Activity Scale, related FACT-G scales, and the KPS score and lower frequencies of correlation with the unrelated MPQ sensory groups. In conclusion, the 101-item MAPS yields more information about a patient's pain, emotional, physical, and cognitive status than does a much longer test battery consisting of 191 items contained in the MPQ, POMS, and FACT-G questionnaires. PERSPECTIVE: This study demonstrates the validity of a new pain questionnaire developed objectively by multivariate cluster analysis rather than subjectively by expert opinion. MAPS assesses patients' somatosensory and emotional experiences, and feelings of well-being with greater scope and accuracy than a battery of questionnaires, and it does so in much less time.

Factor analysis validates the cluster structure of the dendrogram underlying the Multidimensional Affect and Pain Survey (MAPS) and challenges the a priori classification of the descriptors in the McGill Pain Questionnaire (MPQ).

The purpose of this study was to validate the content and structure of the Multidimensional Affect and Pain Survey (MAPS) by means of factor analysis. The 101-MAPS is based on a dendrogram obtained by cluster analysis and contains 30 clusters subsumed within three superclusters. If the MAPS is a valid questionnaire for the quantification of emotion and pain in patients, then factor analysis of patients' intensity ratings should produce factors which correspond to the cluster structure of the dendrogram. To confirm the structure of the dendrogram and hence, MAPS, factor analysis was applied to the responses by 100 outpatients diagnosed with early stage cancer. Principal components analysis of responses to the MAPS yielded six factors. In accordance with the hypothesis, 13 of the 17 clusters within the MAPS somatosensory pain supercluster loaded on three sensory factors: factor 1, severe sensory pain; factor 3, moderate sensory pain; and factor 6, numb/cold. Five of the eight clusters within the emotional pain supercluster loaded on factor 2, negative emotions. Four of the five clusters in the well-being supercluster loaded on factor 4, good health. Factor 5, manageable illness was loaded on by clusters from the well-being supercluster and the somatosensory pain supercluster. The homogeneity of the six factors found demonstrate the validity of the MAPS and the cluster structure of the dendrogram. MAPS proved sensitive to sex differences; women endorsed the negative emotions factor more strongly than did men. The MAPS factors were much more homogeneous than those reported in the literature for the McGill Pain Questionnaire.