Application of the Rasch Model to Develop a Simplified Version of a Multiattribute Utility Measurement on Attitude Toward Labor Epidural Analgesia

Abstract

Valid and reliable measures based on health behavior theory, such as multiattribute utility decision theory, are essential to elucidate complex relationships between psychological factors and labor pain. In this study we aimed to use Rasch analysis to simplify a previously developed 20-item multidimensional questionnaire on attitude toward labor epidural analgesia using multiattribute utility theory. The Rasch analysis was performed to condense item selection categories, to exclude misfit items and persons, and to generate a unidimensional attitude toward labor epidural analgesia (ATLEA) score. Item characteristics and thresholds of rating categories in the questionnaire were also estimated. Reliability and empirical validity of the simplified version were further compared with those of the full version. One hundred sixty-seven postpartum women completed the questionnaire. The original 10 rating scale categories were combined to make 4 without compromising reliability. Three respondents and 11 items were excluded because of misfit. Reliability indices of the simplified and full versions were 0.68 and 0.74, respectively. The correlation coefficient between ATLEA scores from the simplified and full versions was 0.89. Empirical validity values of ATLEA scores from the simplified and full versions for labor epidural analgesia decision, as assessed by area under the receiver operating characteristic curves, were 0.80 and 0.81, respectively. We demonstrated application of the Rasch analysis to simplifying a multiattribute utility questionnaire without compromising reliability. Further study is necessary to determine whether the simplified questionnaire is valid for use in clinical practice.