1 Random-intercept logistic regression
analyses were performed using 2011 data from a national datab
2 Statistical
analyses were performed using a 2-tailed Student t-test.
3 was examined using Cox regression analysis, and sensitivity
analyses were performed using a competing risk approach.
4 Data
analyses were performed using a computational modeling approa
5 Adjusted
analyses were performed using a model of baseline risk determ
6 Discovery and replication
analyses were performed using a p-value-based meta-analysis.
7 Time trend
analyses were performed using a piecewise linear regression a
8 Mixed treatment comparison
analyses were performed using a random-effects Poisson regres
9 Statistical
analyses were performed using a software program, and P value
10 Analyses were performed using absorption and fluorescence spe
11 In vitro antioxidant
analyses were performed using biological samples.
12 Electrophysiological studies and molecular
analyses were performed using CAR haploinsufficient (CAR(+)/(
13 Analyses were performed using Cox proportional hazards regres
14 Univariate and multivariate survival
analyses were performed using Cox proportional hazards regres
15 Survival
analyses were performed using Cox proportional hazards.
16 Crude and propensity score-corrected
analyses were performed using Cox regression, with additional
17 Regressions
analyses were performed using Cox regression.
18 Uni- and multivariate survival
analyses were performed using Cox-proportional hazards models
19 Interrupted time series
analyses were performed using generalised additive mixed mode
20 Genomic
analyses were performed using haplotype sharing analysis and
21 Eplet and epitope
analyses were performed using HLAMatchmaker and Cn3D software
22 Statistics
analyses were performed using IBM SPSS 20.0.
23 alizations was noted in the prasugrel-treated patients when
analyses were performed using instrumental variable methods.
24 ence noted in any of the falsification end-point rates when
analyses were performed using inverse probability of treatmen
25 Analyses were performed using linear and logistic regression,
26 Data
analyses were performed using linear random-intercept models
27 Sample
analyses were performed using liquid chromatography coupled w
28 Univariable and multivariable
analyses were performed using log-binomial regression.
29 Primary
analyses were performed using mixed linear models.
30 Survival
analyses were performed using models adjusted for traditional
31 Analyses were performed using proportional hazards models.
32 Downstream
analyses were performed using QIIME, PICRUSt, and LEfSe.
33 All
analyses were performed using R version 2.15.0.
34 Meta-
analyses were performed using random effect models and standa
35 Using Seed-based d Mapping software, meta-
analyses were performed using random-effect nonparametric sta
36 All meta-
analyses were performed using Review Manager 5.2.
37 Test precision
analyses were performed using samples from 10 subjects.
38 Mendelian randomization
analyses were performed using single nucleotide polymorphisms
39 of bias tool was used to assess selected studies, and meta-
analyses were performed using statistical software.
40 The
analyses were performed using summary statistics obtained for
41 All
analyses were performed using the 2012 World Health Organizat
42 Untargeted data
analyses were performed using the comprehensive measurements.
43 Analyses were performed using the Cox proportional hazards mo
44 Meta-
analyses were performed using the Hartung-Knapp-Sidik-Jonkman
45 Meta-
analyses were performed using the Mantel-Haenszel random-effe
46 Statistical
analyses were performed using the Pearson correlation and Stu
47 Functional
analyses were performed using transport activity assays with
48 Heritability
analyses were performed using uni- and bivariate Cholesky dec
49 Association
analyses were performed using univariable and multivariable s
50 Molecular
analyses were performed using whole-genome sequencing or whol