Jump to content

Property:Specifications on analyses

From CAMIH

This is a property of type Text.

Showing 20 pages using this property.
K
Statistical analysis was carried out according to the intention to treat protocol. Continuous variables were tested for normal distribution by the Kolmogorov-Smirnov test, then reported as mean (SD) or median as appropriate. Student t-test or Mann–Whitney U test was used to compare continuous variables between groups. Categorical data were summarized as percentages and analyzed with the chi-square test. A repeated measures ANOVA was used to evaluate differences at baseline, middle, and endpoints in time-dependent variables within patient groups. ITT analysis specified, but unclear because demographic information from only 60 patients  +
Patients who received at least one dose of assigned therapy were defined as evaluable for the primary endpoint.  +
ITT Analysis planned, but no results of it reported. Sample size calculation was based on previous studies investigating the effect of L-Carnitine on inflammatory markers, with TNFα level differences as the primary endpoint, and resulted in a recruitment goal of 90 patients (45 per treatment arm) for a statistical power of 90% with an error probability of <5%. After a prescheduled interim analysis for sample size recalculation of 72 blinded datasets showed a wide variation of the standard errors for inflammatory markers, a recruitment of 554 patients (277 per group) would have been necessary. Since this goal was unattainable, the study was closed after enrolment of 72 patients and the data were unblinded for statistical analysis.  +
L
Occurrence, duration and severity of oral mucositis (WHO: grade 0-4, weekly check of oral status during radiotherapy, then monthly by doctor and an author); log rank analysis  +
Baseline characteristics, QoL, and severity of acute and delayed nausea and vomiting were compared between the 2 groups using independent samples t/nonparametric tests for continuous variables and Pearson’s χ2/Fisher’s exact tests for categorical variables. Incidence of acute and delayed nausea and vomiting was calculated as a binary variable (“yes” if participants had nausea or vomiting or “no” if participants had no nausea or vomiting) and was compared with a Pearson χ2 test. Two-tailed tests with a significance level of .05 were used for all analyses.  +
One patient in the green tea group did not complete the seven days of treatment as she died of advanced cancer on day 5, the malodorous scores for day 5 to day 7 were conservatively assumed as per scored on day 4. Frequencies and percentages were calculated for categorical variables; Descriptive statistics were calculated for continuous variables; Wilcoxon ranksum test was used for statistical analysis to compare the malodorous scores between the two arms  +
Number of participants evaluated in PP n=291 For the primary efficacy endpoint, that is, percent improvement in average pain NRS score from baseline to end of treatment, the comparison was analyzed using the Wilcoxon rank-sum test. Estimates of the median difference between nabiximols and placebo, together with approximate 95% CI, were calculated using the Hodges-Lehmann approach, and P-values were used for the hierarchical gate-keeping procedure. Other sensitivity analyses for the primary efficacy endpoint included the Wilcoxon rank-sum test based on the PP analysis set, Van der Waerden test, and analysis of covariance with the corresponding baseline value as a covariate and treatment group as a factor, based on the ITT analysis set. Mixed-effect model repeat measurement was also applied with baseline NRS average pain score as a covariate, treatment group as a fixed factor, and the interaction terms for treatment-by-time and baseline-by-time included. For the key secondary efficacy endpoints (average pain score, worst pain score, and sleep disruption score), analysis of covariance was applied, similar to the primary efficacy endpoint analysis. P-values from these analyses were used for the hierarchical gate-keeping procedure. The time course of the treatment effect on the key secondary endpoints was also evaluated in a similar fashion to the primary efficacy endpoint using model repeat measurement on the ITT analysis set. Analysis of variance was applied on the other secondary endpoints, including PGIC, SGIC, or PSQ, daily total/maintenance/breakthrough opioid dose, except NRS constipation score with ordinal logistic regression. Subgroup analyses for region (U.S. and rest of the world (ROW)) were performed for the primary and key secondary efficacy endpoints using the ITT set at the 0.05 level, without formal adjustment for multiplicity.  +
The specific analysis method is not stated in the study, however, according to the authors, all included patients of the subgroup analysis were evaluated at the end of the study.  +
The specific analysis method is not stated in the study, however, according to the authors, all included patients in the subgroup analysis were evaluated at the end of the study  +
The specific analysis method is not stated in the study, however, according to the authors, all included patients were evaluated at the end of the study. Chi-square-test; Student’s t test and an analysis of variance; survival curves were plotted according to the Kaplan-Meier method; the differences between curves were evaluated by the log-rank test.  +
The specific analysis method is not stated in the study, however, according to the authors, all included patients were evaluated at the end of the study. Chi-square test, Student’s t-test and analysis of variance; survival curves were plotted by the Kaplan-Meier method and statistically evaluated by the log-rank test.  +