<b>CLME</b>: An <i>R</i> Package for Linear Mixed Effects Models under Inequality Constraints

Jelsema, Casey; Peddada, Shyamal D.

doi:10.18637/jss.v075.i01

Cited by 63 publications

(21 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We checked the variance inflation factor (VIF) of all variables in multivariable models and no pair of independent variables had a VIF greater than 1.3, indicating no problems with multicollinearity. An adjusted test for trend using a constrained interference for linear mixed effects (CLME) test was also performed15 16 (online Supplementary table 3). Statistical analyses were performed using SAS V.9.4.…”

Section: Methodsmentioning

confidence: 99%

Association of income and health-related quality of life in atrial fibrillation

et al. 2019

View full text Add to dashboard Cite

ObjectiveHealth-related quality of life (HRQoL) is a patient-centred benchmark promoted by clinical guidelines in atrial fibrillation (AF). Income is associated with health outcomes, but how income effects HRQoL in AF has limited investigation.MethodsWe enrolled a convenience cohort with AF receiving care at a regional healthcare system and assessed demographics, medical history, AF treatment, income, education and health literacy. We defined income as a categorical variable (<$20 000; $20 000–$49 999; $50 000–$99 999; >$100 000). We used two complementary HRQoL measures: (1) the atrial fibrillation effect on quality of life (AFEQT), measuring composite and domain scores (daily activity, symptoms, treatment concerns, treatment satisfaction; range 0–100); (2) the 12-item Short Form Survey (SF-12), measuring general HRQoL with physical and mental health domains (range 0–100). We related income to HRQoL and adjusted for relevant covariates.ResultsIn 295 individuals with AF (age 71±10, 40% women), we observed significant differences in HRQoL by income. Higher mean composite AFEQT scores were observed for higher income groups: participants with income <$20 000 had the lowest HRQoL (n=35, 68.2±21.4), and those with income >$100 000 had the highest HRQoL (n=64, 81.9±17.0; p=0.04). We also observed a significant difference by income in the AFEQT daily activity domain (p=0.02). Lower income was also associated with lower HRQoL in the mental health composite score of the SF-12 (59.7±21.5, income <$20 000 vs 79.3±16.3, income >$100 000; p<0.01).ConclusionWe determined that income was associated with HRQoL in a cohort with prevalent AF. Given the marked differences, we consider income as essential for understanding patient-centred outcomes in AF.

show abstract

Section: Methodsmentioning

confidence: 99%

Association of income and health-related quality of life in atrial fibrillation

et al. 2019

View full text Add to dashboard Cite

show abstract

“…For each pattern we construct a suitable order restricted test and the final test statistic is taken to be the maximum of all test statistics. The null distribution of the test statistic is derived using the residual bootstrap based procedure developed in Farnan et al ( 2014 ) which is implemented in the package called constrained linear mixed effects (CLME), an R code developed by Casey Jelsema and is described in Jelsema and Peddada ( 2016 ). The R code allows for modeling covariates as well as longitudinal/repeated measurements data.…”

Section: Analysis Of Two or More Groupsmentioning

confidence: 99%

“…For example, one may be interested in comparing the mean relative abundances of individual taxon in subjects ordered by different levels of fat intake or levels of dietary supplements or subjects belong to different age groups etc. In all such situations the classical two-sided tests are not as informative or powerful as the constrained inference (or order restrictions) based tests (Farnan et al, 2014 ; Jelsema and Peddada, 2016 ). Since the proposed methodology converts the simplex data to Euclidean space data, constrained inference theory developed in Farnan et al ( 2014 ) is directly applicable to the present setting.…”

Section: Introductionmentioning

confidence: 99%

Analysis of Microbiome Data in the Presence of Excess Zeros

et al. 2017

Self Cite

View full text Add to dashboard Cite

Motivation: An important feature of microbiome count data is the presence of a large number of zeros. A common strategy to handle these excess zeros is to add a small number called pseudo-count (e.g., 1). Other strategies include using various probability models to model the excess zero counts. Although adding a pseudo-count is simple and widely used, as demonstrated in this paper, it is not ideal. On the other hand, methods that model excess zeros using a probability model often make an implicit assumption that all zeros can be explained by a common probability models. As described in this article, this is not always recommended as there are potentially three types/sources of zeros in a microbiome data. The purpose of this paper is to develop a simple methodology to identify and accomodate three different types of zeros and to test hypotheses regarding the relative abundance of taxa in two or more experimental groups. Another major contribution of this paper is to perform constrained (directional or ordered) inference when there are more than two ordered experimental groups (e.g., subjects ordered by diet or age groups or environmental exposure groups). As far as we know this is the first paper that addresses such problems in the analysis of microbiome data.Results: Using extensive simulation studies, we demonstrate that the proposed methodology not only controls the false discovery rate at a desired level of significance while competing well in terms of power with DESeq2, a popular procedure derived from RNASeq literature. As expected, the method using pseudo-counts tends to be very conservative and the classical t-test that ignores the underlying simplex structure in the data has an inflated FDR.

show abstract

“…For all 3 sets of response models, a Markov chain Monte Carlo approach using Gibbs sampling was performed with 4 chains in parallel with a total of 100,000 iterations (R runjags package; Denwood, 2016). Visual inspection of the chains, autocorrelation plots, and effective sample sizes were used as measures of efficacy.…”

Section: Discussionmentioning

confidence: 99%

Diagnostic accuracy of Somaticell, California Mastitis Test, and microbiological examination of composite milk to detect Streptococcus agalactiae intramammary infections

Rossi

Amarante

Correia

et al. 2018

Journal of Dairy Science

View full text Add to dashboard Cite

The objectives of this study were to estimate the accuracy of Somaticell (Idexx Laboratories Inc., Westbrook, ME), California Mastitis Test (CMT), and microbiological examination of composite milk (MEC) to diagnose Streptococcus agalactiae intramammary infections (IMI), and to assess the agreement between Somaticell and CMT to detect these infections. A secondary objective was to estimate quarter- and cow-level prevalence of S. agalactiae IMI in the herds included in the study. Seven farms were included in the study. The CMT was performed and aseptic milk samples were collected from all quarters of all lactating cows. Composite milk samples were produced in the laboratory by mixing milk from all quarters of each sampled cow. The Somaticell test was performed on a subset of S. agalactiae-positive (n = 167) and S. agalactiae-negative (n = 152) quarter milk samples. Microbiological examination of quarter milk samples (MEQ) was considered the reference test for diagnosing S. agalactiae IMI. The accuracy of all tests at various thresholds was estimated using Bayesian latent class models. Apparent prevalence of S. agalactiae IMI was 15.8% (n = 184/1,164) at the quarter level (based on MEQ) and 28.5% (n = 83/291) at the cow level (based on MEC). True prevalence, as determined by Bayesian models, was 13.0% [95% credible interval (CR): 6.4-24.4%] at the quarter level, and 25.6% (95% CR: 15.3-39.5%) at the cow level. At the cow level (n = 285), sensitivity and specificity of MEC were 95.6 and 99.5%, respectively. The accuracy of Somaticell (n = 319 quarters) to identify S. agalactiae-infected quarters was 75.4, 86.4, 88.9, 89.4, and 91.0% at thresholds of 98,000, 147,000, 205,000, 244,000, and 282,000 cells/mL, respectively. The accuracy of CMT was 87.6, 90.7, 90.8, and 87.4% at thresholds of trace, 1, 2, and 3, respectively. The areas under the receiver operating characteristic curve for Somaticell and CMT were 94.5% (95% confidence interval: 91.8-97.2%) and 92.0% (88.6-95.4%), respectively. At the tested thresholds, the sensitivity of Somaticell ranged from 94.9 to 99.5% to detect S. agalactiae IMI, and specificity ranged from 48.1 to 87.1%. The sensitivity of Somaticell at the lowest threshold (69,000 cells/mL; sensitivity = 99.9%; 95% CR: 98.2-100%) was higher than that of CMT at any tested threshold. Results of this study could be used at the farm level to reduce the use of antimicrobials and reach specific goals in S. agalactiae eradication programs.

show abstract

CLME: An R Package for Linear Mixed Effects Models under Inequality Constraints

Cited by 63 publications

References 17 publications

Association of income and health-related quality of life in atrial fibrillation

Association of income and health-related quality of life in atrial fibrillation

Analysis of Microbiome Data in the Presence of Excess Zeros

Diagnostic accuracy of Somaticell, California Mastitis Test, and microbiological examination of composite milk to detect Streptococcus agalactiae intramammary infections

Contact Info

Product

Resources

About