Ann Hammer and Birgitta Lindmark
^A1 Department of Rehabilitation Medicine Örebro University Hospital Sweden
^A2 Department of Neuroscience Section of Physiotherapy University Hospital Uppsala Sweden

DOI: 10.1080/16501970306132

Objective: Coefficients of repeatability and reproducibility can be guides in differentiating between real changes and measurement error. The aim was to evaluate test-retest intra-rater reliability of a clinical procedure measuring grip force with Grippit(r) in stroke patients, to assess relationship between grip force of the hands and between sustained and peak grip force. Patients and methods: Eighteen patients were tested using the Grippit(r) at two occasions one hour apart. Each occasion comprised three consecutive trials per hand. Results: The paretic hand needs to score a 50 N change within and between occasions to exceed the measurement error in 95% of the observations, irrespective of calculation method. Expressed by CV within the measurement error was 10%. There was no learning or fatigue effect during measuring. There was a wide variation between subjects but the mean ratio between sides was 0. 66. The mean ratio between sustained and peak grip force was 0. 80-0. 84. Conclusion: The measurement errors were acceptable and the instrument can be recommended for the use in stroke patients at a department of rehabilitation medicine.