Development and Validation of a Novel Motor-Cognitive Assessment Strategy of Compensatory Sit-to-Stand Maneuvers in People With Dementia

Background and Purpose: People with dementia show disease-specific sit-to-stand (STS) movement disorders, which relate to deficits of integrating cognitive aspects of motor processes into motor action organization. During STS training in rehabilitation therapy, compensatory STS movement maneuvers are taught aiming to improve patients' STS ability. Previous clinical STS measures do not address these maneuvers or assess cognitive aspects of their motor action organization. The purpose of this study was to develop and validate a motor-cognitive STS assessment instrument for people with dementia (Assessment of Compensatory Sit-to-Stand Maneuvers in People With Dementia, ACSID). Methods: The ACSID covers the recall, initiation, and effective performance of compensatory STS movement maneuvers. The inter- and intrarater reliability, concurrent validity, sensitivity to change, and feasibility were investigated by secondary analysis of data of 97 participants from a randomized controlled trial to improve motor-cognitive performances in people with mild to moderate dementia (mean [standard deviation] age: 82.5 [5.9] years, Mini-Mental Status Examination: 21.9 [2.9] points). Concurrent validity of the individual ACSID items was assessed against reference criteria derived from video-motion analysis. Results: Good to excellent inter- (kappa [] = 0.64-0.99; intraclass correlation coefficient [ICC] = 0.74-0.89) and intrarater (= 0.77-0.91; ICC = 0.77-0.91), concurrent validity (point-biserial correlation coefficients = |0.56|-|0.84|), and sensitivity to change (standardized response means = 0.61-1.00) were found. Feasibility was excellent with a high completion rate (96.9%), no critical events during assessment, and no floor or ceiling effects. Conclusions: The ACSID represents the first observation-based assessment instrument to document motor and cognitive aspects in the execution of a motor key feature in people with dementia, and has been shown to be reliable, valid, feasible, and sensitive to intervention-induced changes.