Construct Validity of the Spanish Versions of the Memorial Symptom Assessment Scale Short Form and Condensed Form: Rasch Analysis of Responses in Oncology Outpatients

Abstract

ContextThe Memorial Symptom Assessment Scale (MSAS) is a self-rating instrument for the assessment of symptom distress in cancer patients. The Spanish version of the MSAS has recently been validated. However, we lack evidence of the internal construct validity of the shorter versions (short form [MSAS-SF] and condensed form [CMSAS]). In addition, rigorous testing of these scales with modern psychometric methods is needed. ObjectivesThe aim of this study was to evaluate the internal construct validity and reliability of the Spanish versions of the MSAS-SF and CMSAS in oncology outpatients using Rasch analysis. MethodsData from a convenience sample of oncology outpatients receiving chemotherapy (n = 306; mean age 60 years; 63% women) at a university hospital were analyzed. The Rasch unidimensional measurement model was used to examine response category functioning, item hierarchy, targeting, unidimensionality, reliability, and differential item functioning by age, gender, and marital status. ResultsThe response category structure of the symptom distress items was improved by collapsing two categories. The scales were adequately targeted to the study patients, showed overall Rasch model fit (mean Infit MnSq ranged from 0.98 to 1.05), met criteria for unidimensionality, and the reliability of scores was good (person reliability > 0.80), except for the CMSAS prevalence scale. Only four items showed differential item functioning. ConclusionThe present study demonstrated that the Spanish versions of the MSAS-SF and CMSAS have adequate psychometric properties to evaluate symptom distress in oncology outpatients. Additional studies of the CMSAS are recommended.