Baseline data is important for comparing treatment groups and assessing success of randomization. It can also help readers understand how generalisable the results of trial are to their clinical situations (Burgess 2003).
Main results: For each primary and secondary outcome there needs to be a summary of results for each group and the estimated effect size (usually risk ratio or mean difference) and its precision (usually 95% confidence intervals) (based on CONSORT Criteria number 17: Outcomes and estimation).
Secondary analyses: Subgroup analyses should always be prespecified. If presenting both unadjusted and unadjusted analyses, specify which are prespecified and which are exploratory.
Burgess D, Gebski V, Keech A. Baseline data in clinical trials. MJA 2003; 179:105-7.