Comparison of clinical data based on limits of agreement

Grilo, Luís M.; Grilo, Helena L.

doi:10.2478/bile-2013-0003

Cited by 7 publications

(13 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…▪ Proof of ( 16) and (17). Since the mean for Gamma(a, b) is ab, we have Plim n→∞ for both = 1 − ( ∕2) and = ∕2, which proves (16). Equation ( 17) can be proved similarly.…”

Section: Appendixmentioning

confidence: 70%

“…16 Results of different approaches on data from Grilo and Grilo. 16 Points are data, in which solid dots are points within the assessment set, and crosses are points outside of the assessment set. In A, the dotted lines are the corresponding 95% confidence intervals.…”

Section: F I G U R Ementioning

confidence: 99%

“…Results for different choices of

γ

and F are presented in Table 4, in which we present scenarios with a smaller sample size ( n = 50) as well as scenarios with a larger sample size ( n = 200). Such sample sizes are common, for example, the sample sizes for data used in Zhou et al, 15 Grilo and Grilo, 16 and Lin et al 4 are 24, 68, and 299, respectively.…”

Section: Nonparametric Approach Via Quantile Regressionmentioning

confidence: 99%

See 2 more Smart Citations

Parametric and nonparametric improvements in Bland and Altman's assessment of agreement method

Chen

Kao

2021

Statistics in Medicine

View full text Add to dashboard Cite

The Bland-Altman method, which assesses agreement via an assessment set constructed by the difference of the measurement variables, has received great attention. Other assessment approaches have been proposed following the same difference-based framework. However, the exact assessment set constructed by the difference is achievable only for measurements with certain joint distributions. To provide a more general assessment framework, we propose two approaches. First, when the measurement distribution is known, we propose a parametric approach that constructs the assessment set through a measure of closeness corresponding to the distribution. Second, when the measurement distribution is unknown, we propose a nonparametric approach that constructs the assessment set through quantile regression. Both approaches quantify the degree of agreement with the presence of both systematic and random measurement errors, and enable one to go beyond the difference-based approach. Results of simulation and data analyses are presented to compare the two approaches.

show abstract

“…▪ Proof of ( 16) and (17). Since the mean for Gamma(a, b) is ab, we have Plim n→∞ for both = 1 − ( ∕2) and = ∕2, which proves (16). Equation ( 17) can be proved similarly.…”

Section: Appendixmentioning

confidence: 70%

Section: F I G U R Ementioning

confidence: 99%

“…Results for different choices of

γ

Section: Nonparametric Approach Via Quantile Regressionmentioning

confidence: 99%

See 1 more Smart Citation

Parametric and nonparametric improvements in Bland and Altman's assessment of agreement method

Chen

Kao

2021

Statistics in Medicine

View full text Add to dashboard Cite

show abstract

“…To estimate the level of agreement bounds, the Bland-Altman analysis assumes homoscedasticity. The assumption of homoscedasticity was evaluated using the Goldfeld-Quandt test (Hoffman, 2015) with the "lmtest" package in R. If the assumption of homoscedasticity was violated, then the level of agreement bounds was estimated to enable the bounds to increase/decrease with increasing mean of both assessments (Grilo & Grilo, 2012).…”

Section: Statistical Analysesmentioning

confidence: 99%

Test–Retest Reliability of Concussion Baseline Assessments in United States Service Academy Cadets: A Report from the National Collegiate Athletic Association (NCAA)–Department of Defense (DoD) CARE Consortium

Houston

Pelt

D’Lauro

et al. 2020

J Int Neuropsychol Soc

View full text Add to dashboard Cite

Objective: In response to advancing clinical practice guidelines regarding concussion management, service members, like athletes, complete a baseline assessment prior to participating in high-risk activities. While several studies have established test stability in athletes, no investigation to date has examined the stability of baseline assessment scores in military cadets. The objective of this study was to assess the test–retest reliability of a baseline concussion test battery in cadets at U.S. Service Academies. Methods: All cadets participating in the Concussion Assessment, Research, and Education (CARE) Consortium investigation completed a standard baseline battery that included memory, balance, symptom, and neurocognitive assessments. Annual baseline testing was completed during the first 3 years of the study. A two-way mixed-model analysis of variance (intraclass correlation coefficent (ICC)3,1) and Kappa statistics were used to assess the stability of the metrics at 1-year and 2-year time intervals. Results: ICC values for the 1-year test interval ranged from 0.28 to 0.67 and from 0.15 to 0.57 for the 2-year interval. Kappa values ranged from 0.16 to 0.21 for the 1-year interval and from 0.29 to 0.31 for the 2-year test interval. Across all measures, the observed effects were small, ranging from 0.01 to 0.44. Conclusions: This investigation noted less than optimal reliability for the most common concussion baseline assessments. While none of the assessments met or exceeded the accepted clinical threshold, the effect sizes were relatively small suggesting an overlap in performance from year-to-year. As such, baseline assessments beyond the initial evaluation in cadets are not essential but could aid concussion diagnosis.

show abstract

“…If the normality assumption is violated, this interpretation may fail, and the consequence may depend on the degree of departure from the normality assumption. If the normality assumption is violated, a logarithmic transformation (in a case of skewness) or a nonparametric approach may be a suitable alternative (Bland & Altman, 1999;Grilo & Grilo, 2012). Consider the case where 1 − γ = 0.95.…”

mentioning

confidence: 99%

A Paradox in Bland-Altman Analysis and a Binary Approach

Kim¹,

Wand²

2020

IJSP

View full text Add to dashboard Cite

A reliable method of measurement is important in various scientific areas. When a new method of measurement is developed, it should be tested against a standard method that is currently in use. Bland and Altman proposed limits of agreement (LOA) to compare two methods of measurement under the normality assumption. Recently, a sample size formula has been proposed for hypothesis testing to compare two methods of measurement. In the hypothesis testing, the null hypothesis states that the two methods do not satisfy a pre-specified acceptable degree of agreement. Carefully considering the interpretation of the LOA, we argue that there are cases of an acceptable degree of agreement inside the null parameter space. We refer to this subset as the paradoxical parameter space in this article. To address this paradox, we apply a Bernoulli approach to modify the null parameter space and to relax the normality assumption on the data. Using simulations, we demonstrate that the change in statistical power is not negligible when the true parameter values are inside or near the paradoxical parameter space. In addition, we demonstrate an application of the sequential probability ratio test to allow researchers to draw a conclusion with a smaller sample size and to reduce the study time. MethodsThis section is structured as follows. The formulation of the hypothesis testing by Lu et al. (2016) is reviewed in Section 2.1. The hypothesis testing is formulated under the normality assumption, and it is referred to as the normal approach throughout the paper. In Section 2.2, a paradoxical case of the normal approach is introduced, and an alternative approach is considered in Section 2.3. In the new approach, a Bernoulli distribution (i.e., binary outcome) is used to address the paradox and to relax the normality assumption. We refer to this alternative approach as the Bernoulli approach throughout the paper. In Section 2.4, the required sample size is compared between the normal approach and the Bernoulli approach. One caveat of the Bernoulli approach is the minimum sample size requirement, and it is briefly discussed in Section 2.5. In Section 2.6, the application of sequential analysis (Wald, 1945;Wald, 1947) is discussed to overcome the caveat. Normal Approach (Lu et al., 2016)Let D i be the difference between two measurements for i = 1, 2, . . . , n, where n is a fixed sample size. Assume D 1 , . . . , D n are independent random variables, and assume D i ∼ N(µ, σ 2 ). Let A = µ − z 1−γ/2 σ be the true lower limit of an acceptable degree of agreement and B = µ + z 1−γ/2 σ be the true upper limit, where z 1−γ/2 is the 100(1 − γ/2)-th percentile of the standard normal distribution (e.g., z 0.975 = 1.96). The probabilistic interpretation of (Point estimators for A and B areÂ =D − z 1−γ/2 S D andB =D + z 1−γ/2 S D , respectively, and their standard errors can be estimated byGiven a confidence level 1 − α, CIs for A and B areGiven a pre-defined acceptable limit, δ > 0, according to Lu et al. (2016), the null hypothesis is formulated as H...

show abstract

Comparison of clinical data based on limits of agreement

Cited by 7 publications

References 8 publications

Parametric and nonparametric improvements in Bland and Altman's assessment of agreement method

Parametric and nonparametric improvements in Bland and Altman's assessment of agreement method

Test–Retest Reliability of Concussion Baseline Assessments in United States Service Academy Cadets: A Report from the National Collegiate Athletic Association (NCAA)–Department of Defense (DoD) CARE Consortium

A Paradox in Bland-Altman Analysis and a Binary Approach

Contact Info

Product

Resources

About