Evaluating the Psychometric Properties of the Eating Assessment Tool (EAT-10) Using Rasch Analysis

R. Cordier*, A. Joosten, P. Clavé, A. Schindler, M. Bülow, N. Demir, S. Serel Arslan, R. Speyer

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

64 Citations (Scopus)

Abstract

Early and reliable screening for oropharyngeal dysphagia (OD) symptoms in at-risk populations is important and a crucial first stage in effective OD management. The Eating Assessment Tool (EAT-10) is a commonly utilized screening and outcome measure. To date, studies using classic test theory methodologies report good psychometric properties, but the EAT-10 has not been evaluated using item response theory (e.g., Rasch analysis). The aim of this multisite study was to evaluate the internal consistency and structural validity and conduct a preliminary investigation of the cross-cultural validity of the EAT-10; floor and ceiling effects were also checked. Participants involved 636 patients deemed at risk of OD, from outpatient clinics in Spain, Turkey, Sweden, and Italy. The EAT-10 and videofluoroscopic and/or fiberoptic endoscopic evaluation of swallowing were used to confirm OD diagnosis. Patients with esophageal dysphagia were excluded to ensure a homogenous sample. Rasch analysis was used to investigate person and item fit statistics, response scale, dimensionality of the scale, differential item functioning (DIF), and floor and ceiling effect. The results indicate that the EAT-10 has significant weaknesses in structural validity and internal consistency. There are both item redundancy and lack of easy and difficult items. The thresholds of the rating scale categories were disordered and gender, confirmed OD, and language, and comorbid diagnosis showed DIF on a number of items. DIF analysis of language showed preliminary evidence of problems with cross-cultural validation, and the measure showed a clear floor effect. The authors recommend redevelopment of the EAT-10 using Rasch analysis.

Original languageEnglish
Pages (from-to)250-260
Number of pages11
JournalDysphagia
Volume32
Issue number2
Early online date21 Nov 2016
DOIs
Publication statusPublished - 1 Apr 2017
Externally publishedYes

Keywords

  • Classic test theory
  • Deglutition
  • Deglutition disorders
  • Item response theory
  • Measurement
  • Oropharyngeal dysphagia
  • Reliability
  • Validity

Fingerprint

Dive into the research topics of 'Evaluating the Psychometric Properties of the Eating Assessment Tool (EAT-10) Using Rasch Analysis'. Together they form a unique fingerprint.

Cite this