Inter-tester comparison between visual estimation and goniometric measurement of ankle dorsiflexion

Abstract

The aim of this study was to examine the inter-tester variability, measurement error and concurrent validity of measurements of active ankle dorsiflexion using the universal goniometer (UG) and visual estimation (VE). Twelve experienced orthopaedic physiotherapists participated in the study. They were asked to perform two measurements of ankle dorsiflexion, the first using visual estimation, the second using a masked universal goniometer. The results demonstrated inter-tester variation in visual estimation to be twice that of the universal goniometer (coefficient of variation: VE = 69.4%, UG = 34.87%). The measurement error was 11.1° for visual estimation and 5.5° for the universal goniometer. This suggests that measures of improvement lower than these figures constitute measurement error and not actual improvement in joint range of movement. The results suggest that a universal goniometer should be used to minimise measurement error between therapists when assessing active ankle dorsiflexion in a clinical setting. They also imply that there is a need to heighten the awareness of the importance of adhering to established and recognised protocols for measuring joint range of motion.Since clinical decisions with regard to patient treatment are often based on measured changes in range of movement, it is important that levels of accuracy in range of movement assessment are taken into consideration. The results of this study suggest that clinicians should only assume that a real clinical change in ankle dorsiflexion has occurred when there has been a change of more than 5° if using goniometry or 11° if using visual estimation. To validate this assumption, however, a further study to examine intra-tester variability should be undertaken.Inter-tester variation is significantly greater when visual examination is used to assess ankle dorsiflexion, which reduces the comparability of results between two or more clinicians. Furthermore, the greater measurement error associated with the use of visual estimation makes real clinical change in ankle dorsiflexion more difficult to quantify, which suggests that clinicians should preferably use a universal goniometer. This large variation among specialist therapists also suggests the need for frequent refresher courses and standardisation of measurement protocols.