Create Passing-Bablok regression online

To create a Passing-Bablok regression, you only need to select two metric variables — the measurements from the two methods you want to compare. The regression line and its confidence interval are then calculated directly here online.

Passing-Bablok regression is a method used to compare two measurement methods, for example a new test against an established gold standard. Both measurements are plotted against each other in a scatter plot, and a regression line is fitted that describes how the two methods relate to one another.

What is Passing-Bablok regression?

Passing-Bablok regression is a non-parametric method for method comparison. Unlike ordinary linear regression, it does not assume that the data are normally distributed and it does not require the measurements to be free of error. The slope and intercept are estimated from the median of the slopes of all lines between pairs of points, which makes the method robust against outliers.

Because it makes so few assumptions, Passing-Bablok regression is widely used in clinical chemistry, laboratory medicine, and other fields where two measurement methods have to be validated against each other.

Interpreting the results

The estimated regression line is described by an intercept and a slope, each reported with a confidence interval:

• Intercept — indicates a systematic (constant) difference between the two methods. If the confidence interval of the intercept includes 0, there is no evidence of a constant difference.
• Slope — indicates a proportional difference between the two methods. If the confidence interval of the slope includes 1, there is no evidence of a proportional difference.

If the intercept is not significantly different from 0 and the slope is not significantly different from 1, the two methods can be considered to agree over the measured range.

Passing-Bablok, Bland-Altman and scatter plots

Passing-Bablok regression is closely related to the Bland-Altman plot: both are used to compare two measurement methods. The Bland-Altman plot focuses on the agreement and the average difference between the methods, while Passing-Bablok regression estimates the systematic and proportional differences through a robust regression line. If you simply want to visualize the relationship between the two variables, you can also create a scatter plot.