{"entity":{"id":"idea-tr1-shrinkage-subgroup-analysis","kind":"idea","name":"Bayesian shrinkage for subgroup claims to stop false 'works in this group' stories","aka":[],"tldr":"Trials look at dozens of patient subgroups and some will look good by chance. A statistical method that pulls extreme subgroup results toward the overall result would make these claims more honest.","summary":"Forest plots of subgroup effects are replaced or accompanied by hierarchical Bayesian shrinkage estimates, in which each subgroup effect is partially pooled toward the overall effect according to its precision. Regulators require shrunken estimates for any subgroup claim in a label and journals require them in publications.","asOf":"2026-09-08","links":[{"label":"Bottleneck evidence (Trial design, endpoints and cost): Davis et al., Availability of evidence of benefits on survival and quality of life of cancer drugs approved by EMA 2009-13 (BMJ 2017)","url":"https://doi.org/10.1136/bmj.j4530"}],"tags":[],"related":[],"cancers":[],"sections":[],"technologies":[],"targets":[],"drugs":[],"companies":[],"institutions":[],"pathways":[],"terms":["hazard-ratio"],"trials":[],"people":[],"bottlenecks":["b-trial-design","b-negative-results"],"keyPapers":["paper-davis-bmj"],"journals":[],"dependsOn":[],"notes":[],"hypothesis":"Shrinkage-based subgroup reporting will reduce the number of subgroup-driven label restrictions or expansions that are later reversed, and will improve replication of subgroup effects in subsequent trials.","rationale":"Naive subgroup estimates are noisy and biased toward extremes when selected; shrinkage is the standard remedy in other fields and would have avoided several well-known oncology subgroup reversals.","test":"Re-analyse subgroup claims from past label decisions with shrinkage and check which would have survived; compare with later replication data.","maturity":"speculative","actor":"regulator","cost":"small","horizonYears":1},"route":"/ideas/idea-tr1-shrinkage-subgroup-analysis/","neighbours":{"term":[{"id":"hazard-ratio","kind":"term","name":"Hazard ratio (HR)","route":"/terms/hazard-ratio/"}],"bottleneck":[{"id":"b-negative-results","kind":"bottleneck","name":"Failures are hidden","route":"/bottlenecks/b-negative-results/"},{"id":"b-trial-design","kind":"bottleneck","name":"Trial design, endpoints and cost","route":"/bottlenecks/b-trial-design/"}],"paper":[{"id":"paper-davis-bmj","kind":"paper","name":"Availability of evidence of benefits on overall survival and quality of life of cancer drugs approved by European Medicines Agency: retrospective cohort study of drug approvals 2009-13","route":"/key-papers/paper-davis-bmj/"}]}}