Statistical methods for assessing agreement between double readings of clinical measurements

Abstract

Statistical analysis of data is crucial in cephalometric investigations. There are certainly excellent examples of good statistical practice in the field, but some articles published worldwide have carried out inappropriate analyses.ObjectiveThe purpose of this study was to show that when the double records of each patient are traced on the same occasion, a control chart for differences between readings needs to be drawn, and limits of agreement and coefficients of repeatability must be calculated. Material and methodsData from a well-known paper in Orthodontics were used for showing common statistical practices in cephalometric investigations and for proposing a new technique of analysis. ResultsA scatter plot of the two radiograph readings and the two model readings with the respective regression lines are shown. Also, a control chart for the mean of the differences between radiograph readings was obtained and a coefficient of repeatability was calculated. ConclusionsA standard error assuming that mean differences are zero, which is referred to in Orthodontics and Facial Orthopedics as the Dahlberg error, can be calculated only for estimating precision if accuracy is already proven. When double readings are collected, limits of agreement and coefficients of repeatability must be calculated. A graph with differences of readings should be presented and outliers discussed.

Prediction of maximum oxygen uptake through incremental exercise testing using ventilometry: a cross-sectional study

“…The following statistical analysis were applied for the development and cross-validation of the estimating equations: (a) analysis of data distribution using the Kolmogorov---Smirnov test and Q-Q normality plots. 18 During the application of the statistical analysis all theoretical presuppositions were reached: normality, linearity, homoscedasticity and independence of variables; (b) multiple stepwise linear regression using the method and independent variables: ''forward'' 21 ; (c) comparison between measured and estimatedV O 2max using the equations developed by the paired t-test; (d) Pearson's correlation; (f) analysis of individual residual scores using the Bland & Altman method 22 and standard error of the estimate (SEE) of 5 ml/kg −1 min −1 as well as reliability analysis using Cronbach's alpha. We used this degree of SEE based on the criterion to accept equation validity for estimatingV O 2max according to recommendations by Jette et al 23 and Gibson et al 24 The significance level adopted was p < 0.05.…”

Test–retest: Agreement or reliability?

“…During the last decades, many research reported on the two related but clearly different notions of test-retest, namely, reliability and agreement (e.g. Bland and Altman, 1986;Grenier et al, 2000;Halligan, 2002;Kottner et al, 2011;Vieira and Corrente, 2011). In this study, following Kottner et al (2011), we define reliability as the capacity of a test or any other measurement tool to differentiate between respondents when measured twice under the same conditions.…”

Retinal oximetry using ultrahigh-resolution optical coherence tomography

“…Information about the measurement repeatability of a developing method like OCT retinal oximetry is important for calculating appropriate study sample sizes in future studies 37. In the present study, there were obvious variations in artery ODR, vein ODR, and artery-vein ODR.…”