The Results section should be organized to follow the order of comparisons and outcomes specified in the protocol so that it explicitly addresses the objectives of the review. The text should present the overall results in a logical and systematic way: it should not have to rely too heavily on the tables or figures, or constantly refer to them to get a clear picture of the review findings. Rather, tables should be used as an additional resource that might provide further details. However, excessive repetition of data in the text that are also provided in tables or figures should be avoided.
Answers to post hoc analyses and less important questions for which there happen to be plentiful data should not be overemphasized. Post hoc analyses should always be identified as such. Authors should make clear in the Results section the method of analysis used for each quoted result (in particular, the choice of effect measure, the direction of a beneficial effect and the meta-analysis model used), although the analytic methods themselves should be described in the Methods section. Results should always be accompanied by a measure of uncertainty, such as a 95% confidence interval. The abstract should summarize findings for only the most important comparisons and outcomes, and not selectively report those with the most significant results. It is helpful also to indicate the amount of information (numbers of studies and participants) on which analyses were based.
Each figure and Additional table should be referred to, explicitly, in the text. When referring to results in a figure, table or ‘Data and analysis’ forest plot that has not been selected as figures, the figure, table or analysis should be referenced in the text.
Authors should consider presenting results in formats that are easy to interpret. For example, odds ratios and standardized mean differences do not lend themselves to direct application in clinical practice but can be re-expressed in more accessible forms. See Chapter 12 (Sections 12.5 and 12.6).