Guidelines for setting cut-off scores in AUC (AUC-GUIDE): balancing sensitivity, specificity, and purpose
- 1Clinical Research Centre, Institute For Clinical Research, National Institutes of Health Ministry of Health Malaysia, Sarawak General Hospital
Abstract
Determining ideal cut-off scores is a critical step in screening and diagnostic test evaluation. While the Area Under the Curve (AUC) with Receiver Operating Characteristic (ROC) curve are frequently reported, these do not directly indicate the best threshold for decision-making. In addition, pursuing an optimal solution might not always be practical. This paper presents a practical guideline for selecting cut-off scores based on AUC values in conjunction with sensitivity, specificity, and intended scientific use. The study outline recommended strategies for different AUC ranges, incorporating realistic considerations and scientific priorities. The framework offers flexibility, including when to prioritize sensitivity, specificity, or balanced performance via Youden's Index and other techniques. This guideline aims to support researchers and clinicians in making evidence-based, transparent, and context-sensitive decisions when interpreting screening or diagnostic accuracy results.
Related
- A general-purpose multi-path stacked coupling for stochastic reaction networks
- Balancing tautological intersection numbers
- The infinite sum of the cubes of reciprocal balancing numbers
- An optimal constant for vector balancing with permutations
- Upper Bounds and Fekete–Szegö Problem for Bi-Univalent Functions Involving Bazilevič and λ-Pseudo Functions Related to Lucas-Balancing Polynomials
BibTeX
@article{bujang2026guidelines,
title = {{Guidelines for setting cut-off scores in AUC (AUC-GUIDE): balancing sensitivity, specificity, and purpose}},
author = {Mohamad Adam Bujang},
journal = {Frontiers in Research Metrics and Analytics},
year = {2026},
month = jun,
doi = {10.3389/frma.2026.1808675},
url = {https://doi.org/10.3389/frma.2026.1808675},
note = {CC BY},
}
RIS
TY - JOUR
TI - Guidelines for setting cut-off scores in AUC (AUC-GUIDE): balancing sensitivity, specificity, and purpose
AU - Mohamad Adam Bujang
PY - 2026
DA - 2026/06/05
JO - Frontiers in Research Metrics and Analytics
DO - 10.3389/frma.2026.1808675
UR - https://doi.org/10.3389/frma.2026.1808675
AB - Determining ideal cut-off scores is a critical step in screening and diagnostic test evaluation. While the Area Under the Curve (AUC) with Receiver Operating Characteristic (ROC) curve are frequently reported, these do not directly indicate the best threshold for decision-making. In addition, pursuing an optimal solution might not always be practical. This paper presents a practical guideline for selecting cut-off scores based on AUC values in conjunction with sensitivity, specificity, and intended scientific use. The study outline recommended strategies for different AUC ranges, incorporating realistic considerations and scientific priorities. The framework offers flexibility, including when to prioritize sensitivity, specificity, or balanced performance via Youden's Index and other techniques. This guideline aims to support researchers and clinicians in making evidence-based, transparent, and context-sensitive decisions when interpreting screening or diagnostic accuracy results.
SN - 2504-0537
ER -