Validation of the Spanish version of DYsphagia in MUltiple Sclerosis questionnaire (DYMUS).

Abstract

Dysphagia is a common symptom in multiple sclerosis that can occur even early in the disease course and can lead to serious complications. Early recognition and treatment can promote comfort, safety and optimal nutritional status. Few dysphagia rating scales are available in Spanish. The aim of this study was to translate the Dysphagia in Multiple Sclerosis Questionnaire (DYMUS) into Spanish and to validate it. Forward and backward translation method was used to translate the original English version of DYMUS into Spanish. A pilot-study with 10 PwMS was carried on in order to improve the intelligibility of the instrument, comprehensibility and content validity of the questionnaire. The questionnaire was filled out by 100 PwMS who were asked a dichotomous question on their swallowing ("Do you have swallowing troubles?"). Descriptive data are presented as median and quartiles for continuous variables and frequency and percentage for categorical ones. Internal consistency reliability was estimated by Cronbach's alfa. Test-retest reliability was estimated by intraclass correlation coefficient. Concurrent validity with a speech and language therapy assessment (SLT-A) was measured with the weighted kappa statistic for the concordance for both dysphagia type and degree categories. Confirmatory factor analysis by means of structural equation models was used to verify the two-factor (solids and liquids) structure of the DYMUS questionnaire. As the goodness of fit evaluation was poor, an additional exploratory factor analysis was carried out. Internal consistency was high. The globus sensation question and the weight loss questions (item 3 and 10) are the least specific with dysphagia symptomatology so they are worst correlated with the sum of the others (item-rest correlation, 0.243 and 0.248, respectively). The test-retest reliability of the DYMUS among 40 patients using ICC was 0.75 (95% CI 0.57 - 0.86). Concurrent validity with SLT-A was poor (weighted kappa 0.37 for dysphagia type and 0.38 for dysphagia degree). The DYMUS questionnaire detected three times more dysphagia (53% versus 17%) than the dichotomous question. Confirmatory factors analysis failed to confirm the bidimensional structure (solid and liquid items) often reported in other validation studies. The subsequent exploratory factor analysis also identified two factors, but with poor interpretability. DYMUS-SP scale is not a sufficiently useful scale to detect dysphagia in PwMS due to the poor concurrent validity and the probable overdiagnosis of the condition; however, it can be helpful as a screening tool when combined with other measures.