Any patient withdrawn from the study will be included in the statistical intention-to-treat analysis.
The perspective of all statistical analyses was intention to treat.
Statistical analysis, in intention to treat, used parametric tests (chi-2, ANOVA).
Two statistical analyzing methods, intention to treat (ITT) and per protocol (PP), were applied to assess the main curative indexes, including Hamilton Depression Scale (HAMD), Self-rating depression scale (SDS), and Rating Scale for Side Effects (SERS) scores.
One patient in the intervention group did not continue through the whole treatment schedule, but is included in the material in accordance with the statistical principle of intention to treat [ 14].
The statistical analysis was intention-to-treat, that is, the invitation group was allocated as randomised.
All statistical analyses were intention-to-treat based and performed using PC-SAS (Version 9.1; SAS Inc., Cary, NC) with P-values less than 0.05 considered statistically significant.
Consistent with CONSORT requirements for analysis of randomized controlled trials, we undertook statistical evaluation using "intention-to-treat" analysis, where outcomes were evaluated among all randomized practices, regardless of their uptake of CME interventions.
Strengths of this study include the randomised design, blinding of the primary outcome assessors, sample size determined from power calculation achieved, appropriate statistical methods including intention-to-treat analysis and linear mixed models applied, moderate withdrawal and reporting according to CONSORT guidelines.
Statistical methods employed an intention to treat and repeated-measures design.
This approach also provides a methodology that transforms a current set of lighting intentions into a single lighting intention with statistical parameters.
