The article by Davide Giavarina discusses the Bland-Altman (B&A) method for assessing the agreement between two quantitative measurement methods in clinical laboratories. The author emphasizes that correlation and regression studies, while commonly used, are not suitable for evaluating method comparability because they focus on the relationship between variables rather than the differences between them. The B&A method, introduced by Altman and Bland in 1983, quantifies agreement by studying the mean difference and constructing limits of agreement. This method is illustrated through a hypothetical example, where the differences between two measurement methods are plotted against their mean, allowing for the identification of bias and the estimation of an agreement interval. The article also highlights the importance of verifying the normal distribution of differences and the use of confidence intervals to assess the precision of the estimates. The B&A plot can be used to evaluate both unit differences and percentage differences, providing a comprehensive approach to method comparison. The article concludes by emphasizing that the B&A method defines intervals of agreement but does not determine whether these limits are acceptable, which must be defined based on clinical, biological, or other relevant criteria.The article by Davide Giavarina discusses the Bland-Altman (B&A) method for assessing the agreement between two quantitative measurement methods in clinical laboratories. The author emphasizes that correlation and regression studies, while commonly used, are not suitable for evaluating method comparability because they focus on the relationship between variables rather than the differences between them. The B&A method, introduced by Altman and Bland in 1983, quantifies agreement by studying the mean difference and constructing limits of agreement. This method is illustrated through a hypothetical example, where the differences between two measurement methods are plotted against their mean, allowing for the identification of bias and the estimation of an agreement interval. The article also highlights the importance of verifying the normal distribution of differences and the use of confidence intervals to assess the precision of the estimates. The B&A plot can be used to evaluate both unit differences and percentage differences, providing a comprehensive approach to method comparison. The article concludes by emphasizing that the B&A method defines intervals of agreement but does not determine whether these limits are acceptable, which must be defined based on clinical, biological, or other relevant criteria.