{"id":6766,"date":"2026-05-12T17:56:24","date_gmt":"2026-05-12T15:56:24","guid":{"rendered":"https:\/\/mediconomics.com\/glossar\/post-hoc-analysis\/"},"modified":"2026-05-12T17:56:24","modified_gmt":"2026-05-12T15:56:24","slug":"post-hoc-analysis","status":"publish","type":"glossary","link":"https:\/\/mediconomics.com\/en\/glossar\/post-hoc-analysis\/","title":{"rendered":"Post-hoc Analysis"},"content":{"rendered":"<p>A post-hoc analysis is a statistical evaluation of clinical study data performed after the completion of the study or after reviewing the results, without the research question or methodology having been predefined in the study protocol or Statistical Analysis Plan (SAP). The term is derived from the Latin &#8220;post hoc&#8221; (after this). Post-hoc analyses include, among others, subgroup analyses, exploratory endpoint evaluations, and retrospective comparisons between treatment groups. They are widely used in clinical research but are subject to significant methodological and regulatory limitations, as they are characterized by multiple testing issues and selection bias.   <\/p>\n<h2>Distinction: Confirmatory vs. Exploratory<\/h2>\n<p>The fundamental difference between confirmatory and exploratory analyses is central to understanding post-hoc evaluations:<\/p>\n<ul>\n<li><strong>Confirmatory analyses<\/strong> are pre-planned, defined in the SAP, and serve to formally prove a hypothesis with a controlled Type I error rate. Their results are considered conclusive for approval and reimbursement decisions. <\/li>\n<li><strong>Exploratory\/post-hoc analyses<\/strong> are not pre-planned or are formulated after reviewing the data. They generate hypotheses for future studies but do not provide confirmatory evidence. Their Type I error rate is uncontrolled, as false positive findings are inevitable with many parallel tests.  <\/li>\n<\/ul>\n<p>The boundary between exploratory and post-hoc analyses is fluid. A pre-planned but not primary subgroup analysis is considered exploratory, not confirmatory. A completely unforeseen evaluation after data review is a classic post-hoc analysis with the strongest limitations regarding evidential value.  <\/p>\n<h2>Multiple Testing and Statistical Error Inflation<\/h2>\n<p>The biggest methodological problem with post-hoc analyses is the inflation of the Type I error rate due to multiple testing. If many subgroups or endpoints are tested in parallel, the probability of a randomly significant result increases significantly. With 20 independent tests at the 0.05 level, a false positive result is statistically almost unavoidable. Corrections such as the Bonferroni method or the Benjamini-Hochberg procedure can control the error rate but are rarely applied in post-hoc analyses. The main problem remains: without predefined analysis plans, the number of tests actually performed is often not transparently documented, making a retrospective multiplicity correction impossible. Regulators and reviewers therefore cannot assess how many tests were performed and how great the actual risk of a randomly significant result is \u2013 a serious problem for evaluating the quality of evidence.     <\/p>\n<h2>Regulatory Assessment and Acceptance<\/h2>\n<p>The EMA, BfArM, and G-BA generally consider post-hoc analyses as hypothesis-generating, not hypothesis-confirming. In the context of early benefit assessment according to \u00a7 35a SGB V, post-hoc subgroup analyses are regularly critically questioned and often not recognized as robust evidence of efficacy for specific patient subgroups. The ICH-E9 guideline requires that subgroup analyses be pre-planned and that their results be labeled as exploratory. The EMA Guideline on the Investigation of Subgroups (2019) further specifies the requirements: subgroup analyses must be biologically plausible, predefined, and secured by adjusted multiplicity control to gain regulatory relevance. Subgroup analyses based solely on a single post-hoc finding, without replication in independent datasets or biological plausibility, are regularly not accepted as a basis for indication extensions or differentiated reimbursement decisions.    <\/p>\n<h2>Relevance for clinical trials<\/h2>\n<p>Despite their methodological limitations, post-hoc analyses are indispensable in clinical research for generating new hypotheses, characterizing subpopulations, and planning follow-up studies. They can provide important clinical signals that were not discernible in the primary analysis. Crucial is transparent labeling: post-hoc analyses must be clearly identified as such to prevent misinterpretation as confirmatory evidence. Full-service CROs like mediconomics support sponsors in the precise planning of subgroup analyses in the SAP, the methodologically correct execution of exploratory analyses, the transparent labeling of post-hoc findings, and regulatory-compliant reporting in the Clinical Study Report. A clearly structured analysis plan that separates confirmatory and exploratory levels is the most effective means of minimizing regulatory discussions about the evidential value of subgroup analyses.    <\/p>\n<h2>Frequently Asked Questions (FAQ)<\/h2>\n<p><strong>Can a post-hoc analysis serve as a basis for approval?<\/strong><\/p>\n<p>Generally no. Regulators do not accept post-hoc analyses as primary evidence of efficacy. Exceptionally, a post-hoc analysis may be used as supportive evidence if it is biologically plausible, has been consistently replicated across multiple studies, and is confirmed by a prospective study. It is not accepted as the sole basis for approval.   <\/p>\n<p><strong>What is the difference between a post-hoc analysis and a sensitivity analysis?<\/strong><\/p>\n<p>A sensitivity analysis is pre-planned in the SAP and examines the robustness of the primary analysis under alternative assumptions (e.g., different imputation methods). A post-hoc analysis is not pre-planned and formulates new questions after data review. Sensitivity analyses are considered part of the confirmatory strategy; post-hoc analyses are purely exploratory.  <\/p>\n<p><strong>Must post-hoc analyses be reported in the Clinical Study Report?<\/strong><\/p>\n<p>Yes. ICH E3 (Structure and Content of Clinical Study Reports) requires that all analyses performed be documented in the Clinical Study Report, regardless of whether they were pre-planned. Post-hoc analyses must be identified as such, and the exploratory nature of the results must be unambiguously communicated in the report.  <\/p>\n<h2>Regulatory References<\/h2>\n<ul>\n<li>ICH E9 \u2013 Statistical Principles for Clinical Trials (1998): Planning of Subgroup Analyses, Exploratory vs. Confirmatory Analyses<\/li>\n<li>ICH E9(R1) \u2013 Addendum on Estimands (2019): Consistency of Analysis Strategies with Estimands<\/li>\n<li>EMA Guideline on the Investigation of Subgroups in Confirmatory Clinical Trials (EMA\/CHMP\/539146\/2013, rev. 2019)<\/li>\n<li>ICH E3 \u2013 Structure and Content of Clinical Study Reports (1995): Documentation Requirements for All Analyses<\/li>\n<li>G-BA Rules of Procedure \u00a7 35a SGB V: Assessment of Subgroup Analyses in Early Benefit Assessment<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>A post-hoc analysis is a statistical evaluation of clinical study data performed after the completion of the study or after reviewing the results, without the research question or methodology having been predefined in the study protocol or Statistical Analysis Plan (SAP). The term is derived from the Latin &#8220;post hoc&#8221; (after this). Post-hoc analyses include, [&hellip;]<\/p>\n","protected":false},"author":10,"featured_media":0,"parent":0,"template":"","meta":{"_acf_changed":false,"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"glossary-cat":[],"class_list":["post-6766","glossary","type-glossary","status-publish","hentry"],"acf":[],"related_terms":"","external_url":"","internal_reference_id":"","_links":{"self":[{"href":"https:\/\/mediconomics.com\/en\/wp-json\/wp\/v2\/glossary\/6766","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/mediconomics.com\/en\/wp-json\/wp\/v2\/glossary"}],"about":[{"href":"https:\/\/mediconomics.com\/en\/wp-json\/wp\/v2\/types\/glossary"}],"author":[{"embeddable":true,"href":"https:\/\/mediconomics.com\/en\/wp-json\/wp\/v2\/users\/10"}],"version-history":[{"count":0,"href":"https:\/\/mediconomics.com\/en\/wp-json\/wp\/v2\/glossary\/6766\/revisions"}],"wp:attachment":[{"href":"https:\/\/mediconomics.com\/en\/wp-json\/wp\/v2\/media?parent=6766"}],"wp:term":[{"taxonomy":"glossary-cat","embeddable":true,"href":"https:\/\/mediconomics.com\/en\/wp-json\/wp\/v2\/glossary-cat?post=6766"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}