Figure 1.
Flowchart of included studies.
Table 1.
Characteristics of the identified RCTs (n = 413).
Table 2.
Alpha level, power level and type of outcome for RCTs where recalculation of the sample sizes were feasible (n = 121).
Table 3.
Median, range and interquartile range for standardized percentage difference where sample recalculation was feasible (n = 121).
Table 4.
Standardized percentage difference per subgroup for RCTs where recalculation of the sample sizes was feasible (n = 121).
Figure 2.
Predicted probabilities of adequate sample size reporting with 95% confidence intervals (CIs) derived from the adjusted model for reporting sample size calculation details based on the journal and year of publication.
Table 5.
Univariable and multivariable logistic regression derived ORs and 95% confidence intervals (CI) for the feasibility of sample size recalculation for the identified Randomized Controlled Trials (n = 413).
Figure 3.
Boxplots of percentage standardized difference between assumed and observed variance for continuous outcome from the RCTs where sample size recalculation was feasible (n = 92/121) based on journal of publication.
The horizontal line at zero indicates perfect agreement between assumed and observed variance. Median values below zero indicate optimistic assumptions of variance (smaller than observed) and vice versa.
Figure 4.
Boxplots of ratio of Odds Ratios (ORs) of assumed compared to observed ORs or HRs for binary, ordinal and time-to-event outcomes from the RCTs where sample size recalculation was feasible (n = 29/121) based on journal of publication.
The horizontal line at 1 indicates perfect agreement between assumed and observed ORs. Median values below 1 indicate optimistic assumptions of variance (smaller than observed) and vice versa.