13. Discussion and overall conclusions
Section 13 of a clinical study report says what the study means. It brings together what Section 11 shows about the intended benefits and what Section 12 shows about unwanted effects: how the benefits and the risks stand in relation to one another, what in the findings was new or unexpected, and what follows for particular participants and for future studies. Two further questions shape the section: whether the discussion is made in the report or reserved for the integrated summaries of the whole dossier, and how the conclusions are written so that they can be copied into the synopsis.
Discussion and conclusions, merged or separate
The discussion and the overall conclusions may be merged if preferred, but CORE Reference suggests keeping them separate.
Summary of results and the relationship of risks and benefit
E3 asks for a brief summary and discussion of the efficacy and safety results and of the relationship of risks and benefit, referring to the tables, figures and earlier sections as needed. It sets two constraints on the presentation: it should not simply repeat the description of results, and it should not introduce new results.
CORE Reference details the inclusion of a benefit–risk assessment and of possible drug interactions in the discussion. It also clarifies that the discussion should not present a review of the therapeutic area, and that any post-hoc results in the discussion should be accompanied by an explanation of the relevance of that post-hoc analysis to this study.
New, unexpected and inconsistent findings
E3 asks that the discussion and conclusions clearly identify any new or unexpected findings, comment on their significance, and discuss any potential problems such as inconsistencies between related measures.
The results in the light of other existing data
E3 also asks that the clinical relevance and importance of the results be discussed in the light of other existing data.
Implications for individual subjects, at-risk groups and future studies
E3 asks that the discussion identify any specific benefits or special precautions required for individual subjects or at-risk groups, and any implications for the conduct of future studies.
Discussion in the report or in the integrated summaries
E3 allows the alternative of reserving these discussions for the summaries of safety and efficacy that refer to the entire dossier — the integrated summaries.
The discussion section is often a source of doubt among medical writers. The usual advice is to keep it as short as possible, without deep analysis and without any attempt to put the trial into context. Two arguments support this advice. The best place to really discuss the findings is the integrated summaries, where pooled data are presented and the focus is on the big picture. And because company positions may change over time, a discussion that is too detailed and assertive may cause problems later. This shorter-is-better approach is sound and widely applicable, particularly to reports that will generally be read as part of a submission.
E3 is not, in itself, against actually including discussion in the discussion section. The last sentence of its passage suggests, in line with the shorter-is-better advice, that summaries and overviews may be more appropriate places to compare the trial's results with other trials, but its phrase "such discussions" is ambiguous and might not refer to all the preceding points. Detailed comparisons of the trial's results with other results from the programme or from the literature are best left to the summaries, but questions about trial design and conduct may still be worth mentioning. The first sentence of the passage points the same way: identifying any new or unexpected findings means saying whether the interpretation of the data carries caveats and whether any findings bring the proper conduct of the study into question. It is legitimate to consider addressing such study-specific issues in the discussion, and the guideline would seem to encourage it.
Not every report is submitted as part of an initial dossier. A report that forms part of a follow-up measure, for a study the health authorities required as a condition for marketing approval, may well be read largely as a stand-alone document rather than as part of an overall dossier. Its discussion should then certainly address the peculiarities of the trial and, where the outcomes are not as expected, justify why that might be. If the findings are not properly justified, the health authorities are likely to demand an explanation anyway. The discussion should certainly not be a dissertation, but in some contexts certain study issues should be considered for discussion, perhaps even at length.
Shape of the conclusions
CORE Reference suggests that the conclusions be stated concisely in one or two short paragraphs, with no repetition of data. Their content may then be copied directly into the conclusions of the synopsis, which is already common practice.