Evaluating correlation and interrater reliability for four performance scales in the palliative care setting

Area of Science:

• Palliative Care Medicine
• Clinical Assessment
• Health Services Research

Background:

• Performance scales objectively measure patient function and predict therapy response and survival.
• Four scales are used in palliative care, with two specifically designed for this population.
• The optimal performance scale for palliative care remains undetermined.

Purpose of the Study:

• To determine correlations among four common palliative care performance scales.
• To compare the interrater reliability of these four scales.

Main Methods:

• Three healthcare professionals assessed each patient using all four scales concurrently.
• Spearman correlation coefficients and kappa values were calculated.
• Interrater reliability was assessed using both exact agreement and partial credit measures.

Main Results:

• All four performance scales showed highly significant linear correlations.
• No single scale demonstrated significantly higher interrater agreement than others, regardless of the reliability measure used.
• Clinical experience or profession did not emerge as a significant factor in agreement.

Conclusions:

• The choice of performance scale may be less critical than the rater's training.
• Enhanced training in scale administration is recommended for consistent patient function assessment.
• Future studies should consider interrater agreement as a critical parameter.