A Century of Asthma Mortality
Richard Beasley, Neil E. Pearce in The Role of Beta Receptor Agonist Therapy in Asthma Mortality, 2020
Changes in diagnostic fashions are likely to happen slowly and therefore are more likely to account for slow increases than for rapid increases in reported mortality rates. However, arguments have also been advanced to explain more rapid changes in diagnostic fashions. The introduction of a new treatment, such as a bronchodilator, could act almost as a diagnostic test. Esdaile et al. pointed out that asthma could be defined as bronchodilator-reversible airway obstruction.3 Given the massive increase in sales of aerosolized bronchodilators in the early 1960s, they suggest this could have accounted for a substantial increase in the diagnosis of asthma and thus reported deaths. However, they also noted that there are a number of arguments that can be raised against this hypothesis. The most important is that it cannot account for the subsequent rapid decline in mortality despite increasing total beta agonist sales. In addition, aerosolized bronchodilator sales were high in many countries that did not experience the 1960s epidemic.9
Assessing the Microbiome—Current and Future Technologies and Applications
David Perlmutter in The Microbiome and the Brain, 2019
On the other hand, beta diversity is an analysis technique used to compare diversity between samples (Jovel et al. 2016 and Olsen 2016). It is typically used to determine “how different” samples are from each other by effectively measuring the distance between samples because similar samples are “closer” together. This technique can be done with the supervision of phylogenetic data (e.g., UniFrac) or without it (e.g., Bray–Curtis dissimilarity). Once the beta diversity is computed, it is often displayed graphically by reducing the dimensionality of the dataset using either non-metric multidimensional scaling or principal coordinate analysis. These methods are extremely useful for both data visualization and clustering based on the covariates under investigation, but both rely on the assumption that variation in beta diversity can be explained by a few independent factors.
A Brief Review of Sequential Testing Methods
Albert Vexler, Alan D. Hutson, Xiwei Chen in Statistical Testing Strategies in the Health Sciences, 2017
Therefore, one can easily obtain that where denotes the (1-α)th quantile of a normal distribution with a mean of μ0 and a variance of σ2/n. As a consequence, and hence the fixed sample size in the nonsequential LRT test can be obtained by solving. In this example, assuming α = 0.05 and β = 0.2, the nonsequential LRT test requires a sample size ≥6.1826 (6.1826 corresponds to how much sample information is required to achieve the desired error probabilities). Here we use the uniroot function in R to obtain the sample size. The R code to obtain the results is shown below: > ## set up parameters > alpha <- 0.05 # the significance level > beta <- 0.2 # power=1-beta=0.8 > muX0 <- -1/2 > muX1 <- 1/2 > > ### Likelihood ratio test > # power as a function of the number of observations n > get.power <- function(n){ + right.part <- qnorm(alpha,mean=muX0,sd=1/sqrt(n),lower=FALSE) + pnorm(right.part,mean=muX1,sd=1/sqrt(n),lower=FALSE) + } > power <- 1-beta > n.LR <- uniroot(function(n) get.power(n)-power, c(0, 10000))$root > n.LR [1] 6.182566
Distinct profiles of bile acid metabolism caused by gut microbiota in kidney transplantation recipients revealed by 16S rRNA gene sequencing
Published in Archives of Physiology and Biochemistry, 2023
Xiaoqiang Wu, Xiangyong Tian, Guanghui Cao, Zhiwei Wang, Xuan Wu, Yue Gu, Tianzhong Yan
Beta diversity refers to the diversity of species composition or the replacement rate of species along the environmental gradient between different communities that change along the environmental gradient, so it is also called between-habitat diversity (Walters and Martiny 2020). Beta diversity reflects the degree of similarity in species diversity of different sample groups, and the small value of beta diversity indicated that the species of the two groups were similar. When considering the existence of species, there is a difference in microbial species between the healthy control group and the renal transplant recipient, but it is not significant, with unweighted UniFrac distances (R2 =0.08162; P = 0.4929). When considering species abundance, we find that there were significant changes in the community structure of bacteria of experiment recipients compared to normal control, with weighted UniFrac accommodate (R2=0.2228; p = 0.0453) (Figure 4).
New nonparametric measures for instantaneous and granger-causality tail co-dependence
Published in Journal of Applied Statistics, 2022
Cees Diks, Marcin Wolski
In this section we introduce the main mechanics of the new risk transmission measures. While our methodology can be possibly applied to assess risk more broadly, we benchmark our framework against the literature on financial risk, in which investors and policy makers typically aim to understand probability of financial loss associated with a given scenario. This feeds into their decision making process, hoping to improve the expected outcome or better prepare for a possible disaster. Commonly used financial risk management techniques include measures of dispersion, Sharpe ratio, and so-called beta, which describes sensitivity of individual stocks to shocks in the market benchmark. More sophisticated methods include, for instance, Expected Shortfall or Value-at-Risk metrics, as we describe below. In this spirit, conditional measures, like CoVaR, allow to study how specific risk factors associated with one variable affect other variables, being particularly useful for spillover or contagion analysis. This is also an area which our methodology aims to contribute to.
Pseudoword spelling ability predicts response to word spelling treatment in acquired dysgraphia
Published in Neuropsychological Rehabilitation, 2022
Jennifer Shea, Robert Wiley, Natalie Moss, Brenda Rapp
Table 4 reports all of the values obtained from Analyses 1A and 1B that quantify the pre-treatment integrity of core spelling functions and treatment-related changes in spelling accuracy and which were then used for Analysis 2. The full results of the LMEM analyses as well as pairwise correlations between all variables are reported in the Supplemental Online Materials. For beta values corresponding to functional integrity such as word frequency, a more positive beta value indicates greater sensitivity. For example, for frequency, a beta = 0 would indicate no relationship between spelling performance and word frequency, whereas a beta = 0.2 would indicate better spelling performance on higher frequency words. For word length, a more negative value indicates greater sensitivity to length. For example, a beta = 0 would indicate no relationship between spelling performance and length, whereas a beta = −0.2 would indicate worse spelling performance on longer words. Beta values quantifying accuracy changes (e.g., Trained Pre vs. Post) reflect the change in log odds, a standardized effect size measure. They can be interpreted as odds ratios by exponentiating the value; for example, a Training Words Pre vs. Post beta = 1 corresponds to an odds ratio of exp(1) = 2.72, meaning that the participant was almost three times as likely to be correct after completing training. If such a participant had 1:1 odds of being correct at pre-training (50% accuracy), at post-training he/she would have 2.72:1 odds post-treatment (73% accuracy; for more information on this approach to analyzing training studies see Zorzi et al., 1998).
Related Knowledge Centers
- Alpha Particle
- Beta Decay
- Gamma Ray
- Ionizing Radiation
- Radioactive Decay
- Atomic Nucleus
- Radiation Protection
- Nuclear Fission Product
- Caesium-137
- Phosphorus-32