Responsiveness of performance-based outcome measures for mobility, balance, muscle strength and manual dexterity in adults with myotonic dystrophy type 1

• Published in: Journal of rehabilitation medicine 2018-03, Vol.50 (3), p.269-277
• Main Authors: Kierkegaard, Marie, Petitclerc, Émilie, Hébert, Luc J, Mathieu, Jean, Gagnon, Cynthia
• Format: Article
• Language: English
• Subjects: Balance; Climbing; Dexterity; Grip strength; Hypotheses; Manual dexterity; Meaning; Measurement; Medicin och hälsovetenskap; Mobility; Motor ability; Myotonic dystrophy; myotonicdystrophy; outcomemeasures; Perceived changes; psychometrics; Rehabilitation; Responsiveness; Strength; Walking
• ISSN: 1650-1977; 1651-2081; 1651-2081
• DOI: 10.2340/16501977-2304

To assess changes and responsiveness in outcome measures of mobility, balance, muscle strength and manual dexterity in adults with myotonic dystrophy type 1. A 9-year longitudinal study conducted with 113 patients. The responsiveness of the Timed Up and Go test, Berg Balance Scale, quantitative muscle testing, grip and pinch-grip strength, and Purdue Pegboard Test was assessed using criterion and construct approaches. Patient-reported perceived changes (worse/stable) in balance, walking, lower-limb weakness, stair-climbing and hand weakness were used as criteria. Predefined hypotheses about expected area under the receiver operating characteristic curves (criterion approach) and correlations between relative changes (construct approach) were explored. The direction and magnitude of median changes in outcome measures corresponded with patient-reported changes. Median changes in the Timed Up and Go test, grip strength, pinch-grip strength and Purdue Pegboard Test did not, in general, exceed known measurement errors. Most criterion (72%) and construct (70%) approach hypotheses were supported. Promising responsiveness was found for outcome measures of mobility, balance and muscle strength. Grip strength and manual dexterity measures showed poorer responsiveness. The performance-based outcome measures captured changes over the 9-year period and responsiveness was promising. Knowledge of measurement errors is needed to interpret the meaning of these longitudinal changes.