Inter-assessor reliability of practice based biomechanical assessment of the foot and ankle

BACKGROUND: There is no consensus on which protocols should be used to assess foot and lower limb biomechanics in clinical practice. The reliability of many assessments has been questioned by previous research. The aim of this investigation was to (i) identify (through consensus) what biomechanical examinations are used in clinical practice and (ii) evaluate the inter-assessor reliability of some of these examinations. METHODS: Part1: Using a modified Delphi technique 12 podiatrists derived consensus on the biomechanical examinations used in clinical practice. Part 2: Eleven podiatrists assessed 6 participants using a subset of the assessment protocol derived in Part 1. Examinations were compared between assessors. RESULTS: Clinicians choose to estimate rather than quantitatively measure foot position and motion. Poor inter-assessor reliability was recorded for all examinations. Intra-class correlation coefficient values (ICC) for relaxed calcaneal stance position were less than 0.23 and were less than 0.14 for neutral calcaneal stance position. For the examination of ankle joint dorsiflexion, ICC values suggest moderate reliability (less than 0.61). The results of a random effects ANOVA highlight that participant (up to 5.7°), assessor (up to 5.8°) and random (up to 5.7°) error all contribute to the total error (up to 9.5° for relaxed calcaneal stance position, up to 10.7° for the examination of ankle joint dorsiflexion). Kappa Fleiss values for categorisation of first ray position and mobility were less than 0.05 and for limb length assessment less than 0.02, indicating slight agreement. CONCLUSION: Static biomechanical assessment of the foot, leg and lower limb is an important protocol in clinical practice, but the key examinations used to make inferences about dynamic foot function and to determine orthotic prescription are unreliable.