Inter-rater reliability of postural observation after stroke

Area of Science:

• Rehabilitation Medicine
• Clinical Assessment Tools
• Stroke Care

Background:

• Accurate assessment of stroke patients' posture is crucial for effective rehabilitation.
• Existing bedside observation methods may lack standardized tools, potentially leading to inconsistent data.
• A reliable pictorial tool could improve the objectivity and consistency of posture assessment.

Purpose of the Study:

• To evaluate the inter-observer reliability of a new pictorial tool designed for bedside stroke patient posture observation.
• To compare the reliability of two versions of the pictorial tool across different patient positions and body segments.

Main Methods:

• Three projects utilized two versions of a pictorial tool to assess 19 aspects of stroke patients' posture.
• Simultaneous observations were made by a main observer and co-observers with varying experience in 57 stroke patients.
• Inter-observer agreement was quantified using kappa values and percentage agreement across seated, supine, and side-lying positions.

Main Results:

• Acceptable agreement (>70%) was achieved for 67% and 73% of observations concerning the affected upper and lower limbs, respectively.
• Agreement for head, neck, and trunk posture observations was considerably lower, with only 34% achieving acceptable agreement.
• Inter-observer agreement did not significantly improve with increased observer experience.

Conclusions:

• The pictorial tool shows promise for quick, bedside stroke patient posture data collection.
• Refinements to the tool, enhanced observer training, and further validation are necessary before widespread adoption.
• The tool's current reliability is higher for limb posture compared to trunk and head/neck positioning.