Run Passing Bablok regression to compare methods

 

Method comparison with the Passing and Bablok regression

When developing a new method to measure the concentration or the quantity of an element (molecule, micro organism, …) you might want to check whether it gives results that are similar to a reference or comparative method or not.

Passing and Bablok (1983) developed a regression method that allows comparing two measurement methods, which overcomes the assumptions of the classical linear regression that are inappropriate for this application. XLSTAT-LifeSciences provides the Passing and Bablok regression to evaluate the performance of a method compared to another.

Dataset for method comparison

The data correspond to a medical experiment during which the concentration of an antibody is measured for 8 mice submitted to 8 different doses of a new molecule being tested. For each mouse, a blood sample has been taken and divided into four homogeneous sub-samples. Two methods are being tested each on 2 of the 4 sub-samples. The first method is currently considered as the reference, but it is much more expensive than the second and new method.

Our goal is to check if it is possible to use the new method instead of the reference one.

Setting up

• Open XLSTAT

• Select the Method validation > Passing and Bablok regression feature.

• The Passing and Bablok regression dialog box appears.

• In the General tab, select the data that correspond to the first method, then to the second method.

• In the Options tab, we choose the first method which was developed by Passing and Bablok (1983). It should be used when both methods are on the same scale and move in the same direction (positive correlation between X and Y).

• Click OK to start the computations and display the results.

Interpreting the results

The first table displays the descriptive statistics for the two methods. The new method has a larger mean but a larger variance as well.

Then, the model coefficients are displayed.

The intercept value is -1.970 with a confidence interval including 0. This value measures that the systematic difference between the two methods is not significantly different from 0.

The slope coefficient is equal to 1.214 with confidence interval including 1. That means that the proportional difference between the two methods is equal to 1. If 1 is included in the confidence interval then the hypothesis that the slope is equal to 1 is not rejected.

We can say that there no systematic and no proportional differences between the two methods.

The regression plot confirms these remarks:

Before drawing any conclusion, we should test that our model fits a linear model. For that purpose a test of linearity is applied.

Since the test is not rejected, we can say that both methods have no significant difference and that the new less expensive method could be used to replace the former one.