Type 1 Error on Type 2 Inflammation: Circular Analysis in Asthma Clustering

Patchett and Schulman discuss Type 1 error, refers to incorrectly rejecting a true null hypothesis, which can lead to circular analysis in asthma clustering. This occurs when two clusters are falsely concluded to differ in a certain variable when they actually do not. To avoid this, an acceptable pr...

Full description

Saved in:
Bibliographic Details
Published in:American journal of respiratory and critical care medicine 2023-09, Vol.208 (6), p.655-656
Main Authors: Patchett, Brian J, Schulman, Edward S
Format: Article
Language:English
Subjects:
Asthma
Cluster Analysis
Humans
Hypotheses
Inflammation
Viewpoint: Turning the Air Blue
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Patchett and Schulman discuss Type 1 error, refers to incorrectly rejecting a true null hypothesis, which can lead to circular analysis in asthma clustering. This occurs when two clusters are falsely concluded to differ in a certain variable when they actually do not. To avoid this, an acceptable probability of Type 1 error is typically set at 0.05. Eight influential papers on asthma clustering were considered, and all of them utilized the circular analysis method to some extent. However, not all of the hypothesis testing involved double dipping. Cluster analysis, a powerful form of machine learning, remains crucial for identifying asthma phenotypes and endotypes. The challenge lies in performing valid inference when data gathering precedes hypothesis generation. Nevertheless, valid inference is essential for reproducible medical research.
ISSN:1073-449X
1535-4970
1535-4970
DOI:10.1164/rccm.202306-0995VP