Jump to content

Property:Specifications on analyses

From CAMIH

This is a property of type Text.

Showing 20 pages using this property.
N
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.  +
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.  +
The time to first recurrence was estimated using the Kaplan-Meier technique and compared using a 2-sided log rank test. The recurrence rate was compared using 2-sided permutation test. Adjustment for prognostic factors was carried out by means ofretrospective stratification or, in the case of time to first recurrence, by means of the Cox proportional hazards regression model.  +
Exploratory analysis, 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.  +
Per-protocol evaluation of the endpoints; Intention-to-treat evaluation of side effects; 2-sided Wilcoxon rank-sum test: test whether the change in each of the endpoints differed by the intervention arm, because the distributions for some of the endpoints were not symmetrical; percentage of patients with positive or negative changes for each of the endpoints was compared between the intervention arm using a Fisher exact test of proportions at a 2-sided 0.05 level of significance; secondary analyses were not corrected for multiple comparisons, but the results were interpreted cautiously, given the multiple markers being explored.  +
93 patients were evaluable for efficacy analysis; because six patients did not start treatment with vitamin D, only 87 patients were evaluable for safety  +
O
ITT-analysis not specified, but no drop-out occured. The frequency, severity, clinical course, and consequences of HFS were determined by descriptive methods. Between-group differences were assessed using χ2 test for categorical variables. Inverted Kaplan-Meier curves were constructed to determine the relationship between cumulative dose of capecitabine and occurence of HFS, with differences assessed by the log-rank test.  +
P
Between-arm (ginger vs control) comparisons were performed using independent-samples t test (for quantitative variables) or χ2 and Fisher’s exact test (for categorical variables)  +
The number of samples analyzed per endpoint differs slightly since some samples were not available for analysis. Pre-intervention values and PSA changes during the intervention were not normally distributed and were thus tested statistically using Kruskal Wallis test and Mann-Whitney test. Fishere-Freemane-Halton test was used to identify statistical differences in categorical data and contingency tables. For tests of correlation, Spearman's Correlation Coefficient was calculated.  +
Missing data for the efficacy endpoints were imputed using the last observation carry forward (LOCF) method. The proportions of responders were compared between the treatments using logistic regression, with region (North America/Rest of the World) and treatment used as factors. The cumulative response to treatment was shown by plotting cumulative response rates against increasing thresholds for response, ie, percentage changes from baseline in the mean 11-point NRS pain score that defined a response. The cumulative response curves for each of the active treatment groups were compared with placebo using pairwise Wilcoxon rank-sum tests. The Hodges-Lehmann estimates and 95% CI for the median also were performed. The analysis of all the secondary efficacy assessments was considered supportive and no formal adjustments for multiple comparisons were made. The change in mean pain NRS scores, BPI-SF, sleep disruption NRS, PAC-QoL questionnaire, and MADRS were all analyzed using analysis of covariance (ANCOVA), with the baseline value as a covariate and region and treatment group as factors. An analysis also was performed on the mean pain NRS scores to assess the time course of the treatment effect using repeated measure analysis. Additionally, the difference in time required to establish baseline was investigated as a possible moderator of treatment effect by using the number of days until the patient became eligible for randomization and total number of days in the baseline period as covariates in the analysis of change in the mean daily NRS score for average pain.  +
According to authors ITT analysis but 1 drop-out (non-compliance); Chi-square/t-test; logistic regression analysis; Kaplan-Meier test; log-rank test; multivariate analysis was performed using the Cox proportional hazard model  +
The data was analyzed using SPSS Software v10.0 (SPSS corporation Chicago IL) by applying Student’s t-test, ANOVA F test, Chi-square test and Karl Pearson Correlation Coefficient.  +
R