An alternative simple method of assessment of clinical agreement of two methods of measuring blood glucose level without Bland–Altman limits

Abstract

New methods of measuring blood glucose levels are generally accepted when they show sufficient agreement with the existing established method. Such agreement is almost always assessed by Bland-Altman (B-A) method. This method of assessing agreement has several limitations such as this does not assess the degree of agreement and it requires uniform clinical tolerance limits. We describe a simple and robust method of assessing the agreement between two methods of measuring blood glucose levels that directly uses the clinical tolerance limits and assesses the degree of agreement without calculating the B-A limits. This alternative is more flexible also as the clinical tolerance limits can be varyingly set to provide more accuracy at critical values of the blood glucose level and relaxed at not-so-critical

Direct Use of Clinical Tolerance Limits for Assessing the Degree of Agreement between Two Methods of Measuring Blood Pressure

“…These limits can be directly used to assess agreement without going through the process of calculating Bland-Altman limits. 7 This direct method overcomes all of these problems. An additional advantage of this direct method is that the clinical tolerance limits can be varied for different values of the measurement when so required in the clinical context.…”

“…As an adjunct, the differences are plotted against the mean of the two values to provide a Bland-Altman plot 6 that helps to study the pattern of the values of the differences. The Bland-Altman method has several problems, however 7 …”

“…The Bland-Altman method has several problems, however. 7 The primary difficulty is that the Bland-Altman limits only describe the scatter of the differences and do not measure the degree of agreement. Many workers, such as Polonia et al, 2 do not specify the clinical tolerance limits and subjectively conclude that they are wide or narrow.…”

“…The biggest problem, however, is that agreement with the Bland-Altman method cannot be assessed without prespecifying the clinical tolerance limits. These limits can be directly used to assess agreement without going through the process of calculating Bland-Altman limits 7 . This direct method overcomes all of these problems.…”