Statistical issues in randomised controlled trials: a narrative synthesis

Low-Level Nighttime Light Therapy for Age-Related Macular Degeneration: A Randomized Clinical Trial

“…2014; 15:246. accordance with the consensus relating to randomized controlled trials, differences in baseline characteristics between groups were not assessed. 42,43 Descriptive statistics were used to summarize participant demographics, as well as the proportion of individuals showing disease progression in each group (695% CIs using ''exact'' calculation) and the magnitude of changes in secondary outcome measures. The primary outcome measure (proportion demonstrating disease progression at 12 months) was assessed using a stratified Mantel-Haenszel test.…”

Implementing a larviciding efficacy or effectiveness control intervention against malaria vectors: key parameters for success

“…This situation usually occurs when a two-arm parallel random control trial design is used [ 64 ]. One way of solving the imbalance is to compare all baseline characteristics between the groups and to consider variables with statistically significant differences [ 78 , 79 ]. This approach enables anticipation on statistical methods that can be used to control for differences or on ways of considering these differences when drawing conclusions about the study.…”

“…This approach enables anticipation on statistical methods that can be used to control for differences or on ways of considering these differences when drawing conclusions about the study. Because this method ignores variables that are strongly correlated to the primary outcome but which are not significantly different between groups [ 78 , 79 ], it is recommended in a first instance to consider variables displaying high correlation with the primary outcome as the most important [ 79 ]. A variable displaying a correlation coefficient greater or equal to 0.3 with the primary outcome is considered as suitable for statistical adjustment [ 79 ].…”

“…Because this method ignores variables that are strongly correlated to the primary outcome but which are not significantly different between groups [ 78 , 79 ], it is recommended in a first instance to consider variables displaying high correlation with the primary outcome as the most important [ 79 ]. A variable displaying a correlation coefficient greater or equal to 0.3 with the primary outcome is considered as suitable for statistical adjustment [ 79 ]. Methods used to control for imbalance between groups include study design adjustments (minimisation or stratification), matching of similar cluster in pairs or statistical adjustments [ 64 , 69 , 80 ].…”

Evaluation of surgical methods for treatment of cubital tunnel syndrome - statistical perspectives

“…5,6 In addition, a well conducted observational study can be more valuable than randomized controlled trials with distorted randomization, as statistical adjustment for strong predictors of outcome and overall interpretations usually take bias in nonexperimental studies into consideration. [7][8][9] A sub-group analysis 10,11 to compare the difference between the treatments for the sub-groups of interest should be performed; namely, for moderate and severe cubital tunnel syndrome that have previously shown differences in efficacy of surgical options. 12 To estimate differences in treatment effect within subgroups (a subgroup effect), 13 the treatment interaction effect could be examined using multivariate analysis within an appropriate regression model.…”