Analysis Population Identification in Clinical Study Report Displays
ICH E3 emphasizes that analyses, tables, and figures should clearly identify the set of patients from which the displayed result was generated. This is a results-display identification rule, not a substitute for prospectively defining analysis populations in a protocol or statistical analysis plan.
ReviewedEvidence1 sourceSectionWriting & Style
Quick answer
Label each CSR analysis, table, and figure so the analysis set or patient population behind the result is unambiguous.
Key details
Core IssueICH E3 emphasizes that analyses, tables, and figures should clearly identify the set of patients from which the displayed result was generated.
Registerscholarly, professional, or research/report writing
Important caveats
Scope Boundary
This is a results-display identification rule, not a substitute for prospectively defining analysis populations in a protocol or statistical analysis plan.
Further guidance
Content
Put the patient-set identification in the text or directly with the display so reviewers can tell which population generated each result.
Purpose
Label each CSR analysis, table, and figure so the analysis set or patient population behind the result is unambiguous.
Sources and evidence
Sources are shown with the role they play in this guide. Historical or style-sensitive claims are kept within the evidence boundary described above.
ICH E3 section 12.3.3 calls for analysis and discussion across deaths, other serious adverse events, and other significant adverse events after the relevant case listings and narratives. This is an interpretive synthesis, not a duplicate death/serious-event listing or a replacement for individual patient narratives.
ICH E3 section 12.2.3 calls for analysis of adverse events beyond their tabular display, focusing on patterns important to safety interpretation. Do not treat the analysis as a substitute for the event display or patient-level listing, and do not imply causality beyond what the study evidence supports.
ICH E3 section 11.4.6 identifies by-patient displays as a distinct part of efficacy evaluation for showing individual response information when useful. A by-patient efficacy display is not the same as the appendix listing of adverse events or a narrative case report; keep efficacy and safety purposes distinct.
ICH E3 states that estimated or derived values used in a table, figure, or data listing should be conspicuously identified and their derivation and underlying assumptions explained. This display rule is broader than the patient-level listing guide: it applies across tables, figures, and data listings wherever estimated or derived values appear.
ICH E3 section 9.3 identifies Selection of Study Population as the investigational-plan section that frames how the study population was selected and leads into inclusion, exclusion, and removal criteria. This is the ICH E3 investigational-plan population-selection section, not the later results description of who was actually enrolled or their demographic/baseline characteristics.
ICH E3 section 12.4.1 points to a by-patient listing of abnormal laboratory values in section 14.3.4 and distinguishes that safety display from the broader listing of individual laboratory measurements. This page is about the dedicated abnormal-value listing under ICH E3 14.3.4; it is distinct from the full individual-measurement appendix under 16.2.8 and from analytical discussion in section 12.4.2.
ICH E3 section 11.4.2.7 identifies special efficacy-analysis considerations for active-control studies intended to support equivalence. This is a CSR efficacy-results reporting issue for active-control equivalence designs; it does not define a universal equivalence margin or replace protocol-level design justification.
ICH E3 section 11.4.2.1 calls for covariate or prognostic-factor selection and adjustment to be explained, including adjustment methods, analysis results, supportive statistical documentation, and differences from planned analyses when relevant. This results-stage statistical issue is narrower than the general statistical-methods section and should not be used as a substitute for documenting the full analysis plan.
