BACKGROUND: Varus thrust is common in those with knee osteoarthritis. Varus thrust is traditionally identified with visual analysis or motion capture, methods that are either dichotomous or limited to the laboratory setting. Inertial measurement unit data has been found to correlate with motion capture measures of varus thrust in those with severe knee osteoarthritis, allowing for a quantitative and accessible way of measuring varus thrust. However, such measures have not been examined across a wider range of cartilage health. The goal of this study was to compare motion capture and inertial measurement unit estimates of varus thrust in adults who were asymptomatic or who had knee osteoarthritis. METHODS: Adults with (n = 17) and without (n = 10) knee osteoarthritis walked over-ground while motion capture and inertial measurement unit data were collected. We tested the correlations between motion capture variables (peak external knee adduction moment and knee adduction angular velocity during the first half of stance) and inertial measurement unit variables (peak frontal axis shank, thigh, and knee angular velocity during the first half of stance). FINDINGS: No significant relationships were found between the inertial measurement unit and motion capture variables. Between-study differences in cohorts or sensor-to-segment alignment methods may explain the conflicting findings. INTERPRETATION: Our findings suggest that assessing varus thrust across the spectrum of knee health (including those with and without knee osteoarthritis) may not be feasible using these inertial measurement unit measures. We should explore additional inertial measurement unit measures to enable accurate detection or monitoring of individuals with knee osteoarthritis.