# Pre-specified vs post-hoc analysis

Source: https://onco.cc/terms/prespecified-vs-post-hoc/  
OnCo record `prespecified-vs-post-hoc` (Term). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Whether an analysis was planned and written down before the data were seen (pre-specified) or invented afterwards (post-hoc). Post-hoc findings, especially in subgroups, are prone to be chance and count as hypotheses for a new trial, not as evidence.

## Summary

Trial protocols and statistical analysis plans, registered in advance (ClinicalTrials.gov, EU CTR), list the endpoints, populations and tests that will count; regulators approve on pre-specified analyses and treat post-hoc results as exploratory with 'nominal' P values. Post-hoc subgroup analyses have produced both misleading claims (retrospective PD-L1 cut-offs) and genuine discoveries later confirmed (KRAS wild-type for cetuximab). Pre-registration is standard in clinical trials and the site's ideas argue for extending it to animal studies and to 'silent trial' evaluations of AI, where the lack of pre-registration is a known source of false positives.

## Fields

- Kind: Term
- Last checked: 2026-09-09
- Also known as: pre-specified; prespecified; pre-planned; prespecified analysis; pre-specified analysis; post-hoc; post hoc; post-hoc analysis; post-hoc subgroup; retrospective analysis; unplanned analysis; exploratory; exploratory analysis; hypothesis-generating; hypothesis generating; nominal; nominal P value; nominally significant; data dredging; p-hacking; statistical analysis plan; protocol-specified; pre-registered; pre-registration; preregistered; registered report; registered reports; statistical analysis plans

## Sources

- Wikipedia: https://en.wikipedia.org/wiki/Preregistration_(science)
- Wikipedia: https://en.wikipedia.org/wiki/Preregistration_(science)

## Connected records

- terms: [Enrichment and biomarker-stratified designs](https://onco.cc/terms/biomarker-stratified-design/), [Estimands and intercurrent events (ICH E9(R1))](https://onco.cc/terms/estimand/), [Intention-to-treat (ITT) and per-protocol analysis](https://onco.cc/terms/intention-to-treat/), [P-value](https://onco.cc/terms/p-value/), [Primary, secondary and co-primary endpoints](https://onco.cc/terms/primary-endpoint/), [Statistical significance (P values, alpha, multiplicity)](https://onco.cc/terms/statistical-significance/), [Subgroup analysis (forest plots)](https://onco.cc/terms/subgroup-analysis/), [Trial protocol and statistical analysis plan](https://onco.cc/terms/trial-protocol/), [Trial registration and results reporting (ClinicalTrials.gov, EU CTR)](https://onco.cc/terms/trial-registration/), [Why trials fail: underpowered, wrong endpoint, control arm drift, subgroup fishing, crossover](https://onco.cc/terms/trial-failure-modes/)
- fronts: [Drug Discovery Platforms](https://onco.cc/fronts/drug-discovery/)
- trials: [MAGNITUDE](https://onco.cc/trials/magnitude/), [MINDACT](https://onco.cc/trials/mindact/), [Single-injection depot progesterone before breast surgery (Tata Memorial)](https://onco.cc/trials/progesterone-preop-tmh/)

---
JSON: https://onco.cc/api/v1/entities/prespecified-vs-post-hoc.json