Handling of Dropouts or Missing Data in Clinical Study Report
ICH E3 section 11.4.2.2 calls for the methods used to address dropouts and missing observations in efficacy analysis to be described and justified. This is the CSR results-analysis treatment of missing efficacy data; it does not replace the broader protocol or statistical-analysis-plan description of prespecified methods.
ReviewedEvidence1 sourceSectionWriting & Style
Quick answer
Describe the rule or statistical method used for missing observations and dropouts, explain its rationale, and make sensitivity or alternative handling clear when relevant.
Key details
Core IssueICH E3 section 11.4.2.2 calls for the methods used to address dropouts and missing observations in efficacy analysis to be described and justified.
Registerscholarly, professional, or research/report writing
Important caveats
Scope Boundary
This is the CSR results-analysis treatment of missing efficacy data; it does not replace the broader protocol or statistical-analysis-plan description of prespecified methods.
Further guidance
Content
Describe the rule or statistical method used for missing observations and dropouts, explain its rationale, and make sensitivity or alternative handling clear when relevant.
Purpose
Describe the rule or statistical method used for missing observations and dropouts, explain its rationale, and make sensitivity or alternative handling clear when relevant.
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 9.8 calls for changes made after the study began to the conduct of the study or planned analyses to be described, including timing, reasons, decision process, responsibility, available data, and interpretive implications as applicable. This section is broader than a statistical-methods subsection: it covers material study-conduct changes as well as planned-analysis changes, while detailed statistical methods remain in their dedicated section.
ICH E3 describes different levels of report detail: important summary figures and tables may appear in the text, other summary material in section 14, specified patient listings in Appendix 16.2, and U.S. archival individual-patient data in Appendix 16.4. This is an ICH E3 report-organization principle; the exact submission package and regional requirements can differ by authority.
ICH E3 section 9.3.3 identifies Removal of Patients from Therapy or Assessment as a planned study-population subsection of the investigational plan. This subsection states planned removal criteria or procedures; it is not the same as the results-stage disposition of patients or a listing of patients who actually discontinued.
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 9.5.2 calls for discussion of the appropriateness, reliability, or relevance of measurements used to assess efficacy and safety when explanation is needed. This subsection justifies or qualifies measurement choices; it does not replace the section that defines the variables, methods, and timing of assessments.
ICH E3 section 12.2.1 calls for a brief summary of adverse events within the safety evaluation before more detailed displays and analyses. Keep this section a high-level summary; it is not a substitute for detailed adverse-event displays, patient-level listings, serious-event treatment, or narratives.
ICH E3 section 9.2 calls for discussion of the chosen control and study design, including design problems or biases that may matter for interpretation. This guide concerns the rationale and limitations of the control/design choice; the Study Design guide owns the descriptive overview of the study configuration itself.
