Assessing agreement between point of care and pathology laboratory results for INR: Experiences from the Point of Care Testing in General Practice Trial

Aims: To assess the level of agreement between international normalised ratio (INR) results obtained from pathology laboratories and point of care testing (PoCT) devices used in a general practice setting. Methods: INR pathology results were collected from multiple pathology laboratories and CoaguChek S PoCT devices over a 6 month period. Agreement was assessed using both clinically relevant agreement and the Bland Altman method. Results: Analysis was based on 1664 dual measurements collected on 417 patients from 26 general practices across Australia. The percentage of dual measurements satisfying the expanded and narrow agreement criteria were 91% and 89%, respectively. The mean difference in results and the 95 limits of agreement depended on the average INR result: mean difference =-0.30+0.08 × average; 95% limits of agreement =-0.30+0.08 × average±0.77. Conclusions: The current study provides further evidence that PoCT is an acceptable alternative to pathology laboratory testing in a general practice setting. The Bland Altman method is a useful and flexible tool for assessing agreement. Limits of agreement should be reported in future method comparison studies to assist clinicians in patient management.