Common in-lab, marker-based gait analyses may not represent daily, real-world gait. Real-world gait analyses are feasible using inertial measurement units (IMUs), but estimating traditional gait kinematics (e.g., joint angles) from IMU data is challenging. Recent advancements in open-source methods (e.g., OpenSense) may enable reliable and repeatable estimation of joint angles. Before using OpenSense to study real-world gait, we must determine whether these methods: (1) estimate joint kinematics similarly to traditional marker-based motion capture (MoCap) and (2) differentiate groups with clinically different gait mechanics. In this project, we compared inverse kinematics calculated using IMU- and marker-based data across young adults, healthy older adults, and older adults with knee osteoarthritis.