epoc method comparison

The CLIA 2025 allowable differences apply specifically to proficiency testing (PT), not method comparison studies. For method comparisons, each site should establish its own allowable limits based on:

Comparing Vitros (dry slide chemistry), ABL (blood gas analyzer), Sysmex (hematology analyzer) to epoc (electrochemical POC device) introduces inherent variability due to:

This is why method comparison studies should be interpreted with caution and ideally include:

Thank you, Pamela.  We'll take your reply into consideration.

Hello,
I've been struggling with the same issue and learned recently that Pamela is correct. I'm meeting with medical directors and QA to determine new method comparison study limits. The CLIA allowable is just too impossible to maintain. 

Hello,
    We have also been experiencing issues, particularly with Hct. I agree with Jake that the current allowable limits are almost impossible to maintain. Pamela, I was wondering if you could share how your facility does your method comparisons based on what you referred to with bias analysis, regression statistics and clinical impact assessment. I am newer to POC and trying to get a handle on how method comparisons should work. Thank you!

Thanks everyone!  My medical director is currently working with the providers to find out what the clinical impact to patients would be if we widen our allowable limits.  He is wanting a bias analysis study, but we can't figure out how to do this.  Pamela, will you share how your facility does this?

Hello-

At our system, we conducted a comprehensive 2-year retrospective analysis comparing results across all platforms within our network, which includes approximately 26 hospitals. For each analyte, we compared results to the gold standard method/instrument used in the central lab, calculating the percent difference between the two.

In addition to internal data, we referenced several external resources to guide our evaluation, including: (All available through Data Innovations website)

We did not need to conduct additional prospective studies, as the only significant discrepancies were observed with hemoglobin (Hgb) and hematocrit (Hct). In most cases, the bias or difference was clinically insignificant, even when it exceeded CLIA’s allowable limits. This helped support our discussions with providers and leadership regarding the potential for adjusting internal thresholds based on clinical impact rather than regulatory limits alone.