Estimating Bias Between Quantitative Analytical Methods
Author Information
Author(s): William C. Griffiths, Paul Camara, Israel Diamond, John C. Pezzullo
Primary Institution: Roger Williams General Hospital and Brown University
Hypothesis
Can a simple graphical technique be used to estimate the bias between two quantitative clinical methods?
Conclusion
The study presents a method for estimating bias between analytical methods, showing significant bias in potassium, chloride, and carbon dioxide tests, while sodium tests showed no significant bias.
Supporting Evidence
- The study found significant bias in potassium, chloride, and carbon dioxide tests.
- Sodium tests showed no significant bias.
- A simple graphical technique was developed for estimating bias.
- Approximately 25 patient samples were analyzed for each method.
- Results were plotted to visually assess bias and non-linearity.
- Z-values were calculated to quantify the significance of observed bias.
- Control samples were included in the analysis to ensure accuracy.
- The method is designed to be simple and accessible for clinical scientists.
Takeaway
This study helps labs figure out if their testing methods are giving accurate results by comparing them to a trusted method.
Methodology
The study involved analyzing patient samples with both test and reference methods, plotting results, and calculating z-values to assess bias.
Potential Biases
The study may not account for all random sampling fluctuations and potential errors in measurement.
Limitations
The methods used in the examples are not evaluated, and the study does not account for all potential sources of error.
Participant Demographics
Patient samples were used, but specific demographics are not provided.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Want to read the original?
Access the complete publication on the publisher's website