Statistics in Medicine — Reporting of Subgroup Analyses in Clinical Trials.

Wang, R., Lagakos, S. W., Ware, J. H., Hunter, D. J., & Drazen, J. M. (2007). Statistics in Medicine — Reporting of Subgroup Analyses in Clinical Trials. New England Journal of Medicine, 357(21), 2189-2194.

Type

Journal article

Intellectual base

Primary research

Year

2007

Volume

357

Issue

21

Pages

2189-2194

DOI

10.1056/nejmsr077003

DOI registrant

Massachusetts Medical Society

Link

https://doi.org/10.1056/NEJMsr077003

Cited from

organizational_management

Cited by

3 citations across 3 artifacts.

Each citation links to the sentence it supports in the citing article.

Primes

• Objective Creep
  • In research, a study designed for one hypothesis adds secondary aims and sub-group analyses until the primary aim is under-powered.

    This sourceShows that accumulating secondary and subgroup analyses inflates error and that such analyses are typically under-powered relative to the primary aim.

    Supported in partVerified against the work's full text

    Wang et al. show that many subgroup analyses inflate false-positive risk and that trials are often underpowered for subgroup interactions, but not that added aims erode the primary aim's power.

    “Many trials lack the power to detect heterogeneity in treatment effect; thus, the inability to find significant interactions does not show that the treatment effect seen overall necessarily applies to all subjects.”

    Read by an automated reader; not a human review. How support was checked

Mechanisms

• Intersectional Stratification Table
  • The discipline is to pre-specify the strata and cell-size floor before looking, and to treat the worst cell as a lead to validate, not a verdict.

    This sourceRecommends distinguishing and reporting subgroup analyses specified before data inspection from post hoc analyses.

• Subgroup Disaggregation Audit
  • This is the multiple-comparisons trap

    This sourceWarns that post hoc subgroup analyses are vulnerable to multiplicity and inflated false-positive findings.

Verification

Does it exist? Confirmed. This work's DOI resolves to a registered record, which fixes its identity. That is all it fixes.

Does it back the claim? Read against the text for 1 of 3 citations: 1 supported in part. Each verdict is shown under its citation below, with what in the work backs the sentence.

Support is checked per citation rather than per work — the same source can be cited soundly in one article and wrongly in another. Per-citation recording began recently, so a citation with no recorded check is a gap in the record rather than evidence it went unchecked.

See how references were verified.

Registry ID ref:b476d67917dd · see in the full table