Repeatability of the modified Oxford foot model during gait in healthy adults

Background: The Oxford foot model (OFM) is a multi-segment model for calculating hindfoot and forefoot motion. Limited information is available regarding the repeatability and error of this model in adults. Therefore the purpose of this study was to assess the intra-tester reliability of OFM hindfoot and forefoot gait kinematics in adults at initial contact (IC) and toe-off (TO).

Methods: Seventeen healthy adults (age=25.1±4.8 years, height=1.75±0.10m, weight=74.0±12.4kg) were tested on a single visit, during which 1 examiner recorded 2 sessions. For each session, 10 walking trials were recorded using a 12-camera motion analysis system (Vicon, Oxford, UK). Markers were removed and re-applied between sessions. Dynamic hindfoot and forefoot angles were calculated both with and without referencing to neutral stance (assuming neutral stance angles are zero in all planes). Using the 10 trial average, intraclass correlation coefficients (ICC(2,k)) and standard errors of the measurement were calculated for each reference condition, anatomical plane, and joint (hindfoot, forefoot).

Results: Referencing to neutral stance resulted in good reliability (ICC≥0.83) and small error (≤2.45°) for hindfoot and forefoot angle in all planes. Without referencing to neutral stance, sagittal and transverse plane reliability were also good (ICC≥0.90) and error small (≤3.14°); however, frontal plane reliability was poor (ICC≤0.77), with large error (≥4.86°).

Discussion: Our results show that overall the OFM is reliable during adult gait. Reliability for adults is higher than previously reported in children. Referencing joint angles to neutral stance decreased error by up to 2° from previous reports.