Common experiences of pain in children and adolescents--an exploratory factor analysis of a questionnaire

The aim of the study was to reduce everyday and dental treatment pain items included in the extended Children's Pain Inventory (CPI), used in a prior study on Swedish children and adolescents. Another aim was to, by means of exploratory factor analysis (EFA), expose hitherto undiscovered dimensions of the CPI pain variables and thus to improve the psychometric properties of CPI. As some pain items are relevant merely to some individuals, a new and more useful questionnaire construction would enhance the internal validity of the instrument in observational surveys. EFA was applied on the extended CPI instrument. 368 children, 8-19 years old, had answered a questionnaire comprising 10 dental and 28 everyday pain variables. These pain items were analysed using a series of sequentially implemented EFA. Interpretations and decisions on the final number of the extracted factors was based on accepted principles; Kaiser's Eigenvalue >1 criterion, inspection of the scree plot and the interpretability of the items loading. The factors were orthogonally rotated using the Varimax method to maximize the amount of variance. Of all tested EFA models in the analysis, a two, three, four, and five factor model surfaced. The interpretability of the factors and their items loading were stepwise examined; the items were modulated and the factors re-evaluated. A four factor pain model emerged as the most interpretable, explaining 79% of the total variance depicting Eigenvalues > 1.014. The factors were named indicating the profile of the content: Factor I cutting trauma to skin/mucosal pain, Factor II head/neck pain, Factor III tenderness/blunt trauma pain, Factor IV oral/dental treatment pain.