Administration of the German Pediatric Evaluation of Disability Inventory (PEDI-G) Using the Mode of Observation in Children Undergoing Inpatient Rehabilitation: A Reliability and Validity Study.

To investigate inter-rater reliability, concurrent validity, and feasibility of the German Pediatric Evaluation of Disability Inventory (PEDI-G) using the mode of observation in a Swiss inpatient rehabilitation setting with the Functional Independence Measure for Children (WeeFIM®) as criterion. Cross-sectional clinimetric study including 36 children and adolescents with median age 10.8 (quartiles 8.7, 13.0) years with neurological/neuro-orthopedic disorders. Data were collected by healthcare professionals through observation. Analyses were performed using intraclass correlation coefficients (ICC), standard error of measurement (SEM), Bland-Altman plots, Cohen's Kappa κ, percentage agreement, and correlations. Excellent inter-rater reliability (ICCs ≥ 0.97), small SEMs and acceptable limits of agreement for the Functional Skills Scale (FSS) and Caregiver Assistance Scale (CAS) were found. No systematic differences between raters existed. Cohen's Kappa for inter-rater agreement of the Modifications Scale (MS) ranged from poor to strong (-0.06 ≤ κ ≤0.85). Excellent concurrent validity for FSS and CAS with the WeeFIM (ρ  ≥ 0.96), and excellent correlations of FSS and CAS with each other (ρ  ≥ 0.98) were identified. The German PEDI-G seems to be a reliable and valid, but time-consuming tool when applied in an inpatient setting using observation.