Development and reliability of a system to classify the eating and drinking ability of people with cerebral palsy

AimThe aim of this study was to develop a valid classification system to describe eating and drinking ability in people with cerebral palsy (CP), and to test its reliability. MethodThe Eating and Drinking Ability Classification System (EDACS) was developed in four stages in consultation with individuals with CP, parents, and health professionals: Stage 1, drafting informed by literature and clinical experience; Stage 2, modification by nominal groups; Stage 3, refinement in an international Delphi survey; and Stage 4, testing of agreement and reliability between classifications made by speech and language therapists (SaLTs), and between SaLTs and parents. ResultsSeven nominal groups involved 56 participants; 95 people participated in two rounds of the Delphi survey. Using the version of EDACS produced from this process, SaLTs in pairs classified 100 children. The rate of absolute agreement was 78% (kappa=0.72; intraclass correlation coefficient [ICC]=0.93; 95% confidence interval [CI] 0.90-0.95). Any disagreement was only by one level, with one exception. SaLTs and parents classified 48 children. The rate of absolute agreement was 58% (kappa=0.45, ICC=0.86; 95% CI 0.76-0.92). Parents either agreed with SaLTs or rated their children as more able by one level. InterpretationThe EDACS provides a valid and reliable system for classifying eating and drinking performance of people with CP, for use in both clinical and research contexts.