12 Dic

The calculation of distortion and match limits from a given sample are estimates of the actual value of the underlying population. The results are therefore subject to random variation. Assuming that differences in measured values follow a normal distribution, it is therefore necessary to calculate standard errors and confidence intervals for distortion and compliance limits  . For the data in example 1, the 95% confidence interval is sufficient for the estimated bias value of -8.4 from 8.1 to 17.7. Confidence intervals for the lower and upper match limits are -42.6 to -25.8 and 9.0 to 25.8. Due to the rather small sample size, confidence intervals are quite wide and the results are therefore very uncertain. If a method can actually measure the interested gre without error, this method is commonly referred to as the gold standard. However, if a new method, such as measuring the volume of the tumour with less effort or softer for the patient, is to be implemented, it should be determined whether the results obtained are consistent with those of the established method. In the case of a quantitative gre with a constant level of scale (e.g., tumor volume), it is common to calculate the correlation between the two measurement methods. But if I wanted to be sure, I could have mathematically calculated our relational potential. In 1986, Martin Bland and Douglas Altman, two British epidemiologists, published a simple and ingenious method of measuring adequacy. Background: In this article, we describe qualitative and quantitative methods for assessing the degree of agreement (concordance) between two measuring or rating techniques.

An assessment of concordance is particularly important when a new measuring technique is introduced. Tab  shows manual values (SBD1) and automatic measurements (SBD2) of systolic blood pressure in 30 people (example 1, partial data set ). It is interesting to see how well the two measurement methods match (“agreement”). For the first case, it is advisable to use descriptive and graphic methods such as the representation of the dot cloud, as well as the straight line of the mood, and the Bland-Altman diagram.