Reliability and validity of a novel Kinect-based software program for measuring posture, balance and side-bending.

Area of Science:

• Biomechanics
• Rehabilitation Technology
• Clinical Measurement

Background:

• Clinical examinations for posture and balance often lack objectivity and reliability.
• 3D motion analysis offers objective data but is prohibitively expensive for routine clinical use.
• Qinematic™ is an accessible interactive system using Kinect for movement analysis.

Purpose of the Study:

• To evaluate the test-retest reliability and construct validity of the Qinematic™ system.
• To determine minimal clinically important differences for Qinematic™ variables.
• To assess the discriminative validity of Qinematic™ in individuals with low-back pain (LBP).

Main Methods:

• Test-retest reliability assessed in 37 healthy individuals.
• Construct validity compared Qinematic™ to 3D motion capture in 30 healthy individuals.
• Discriminative validity compared 20 LBP patients to 17 healthy controls.
• Evaluated 18 variables across posture, balance, and side-bending.

Main Results:

• Most posture and balance variables showed poor reliability (ICC < 0.4) and validity (Spearman < 0.4).
• Qinematic™ did not differentiate between LBP patients and healthy controls for most posture/balance variables.
• Left-side bending measurement exhibited excellent reliability (ICC = 0.898) and validity (r = 0.943), differentiating LBP patients.

Conclusions:

• Qinematic™ exhibits poor psychometric properties for posture and balance, limiting its use in monitoring individual changes or research.
• The system's variables for balance and posture are not recommended for clinical use.
• Further research into Qinematic™'s dynamic task capabilities is warranted, particularly for side-bending assessment.