Development and validation of a simple clinical nomogram for predicting obstructive sleep apnea

Area of Science:

• Sleep Medicine
• Medical Diagnostics
• Biostatistics

Background:

• Obstructive sleep apnea (OSA) is a prevalent sleep breathing disorder.
• Accurate risk prediction is crucial for timely diagnosis and management.
• Existing screening tools may lack objectivity or ease of use.

Purpose of the Study:

• To develop and validate an objective, user-friendly nomogram for predicting OSA risk.
• To identify key clinical factors associated with OSA.
• To enhance clinical decision-making for sleep disorder referrals.

Main Methods:

• Retrospective data collection from polysomnography patients.
• Random assignment into training (1035 participants) and validation (1078 participants) cohorts.
• Logistic regression and Lasso regression for factor selection and nomogram construction.
• Evaluation using C-index, calibration curves, and decision/clinical impact analyses.

Main Results:

• Identified age, gender, diastolic blood pressure, BMI, neck circumference, and Epworth Sleepiness Scale as significant predictors.
• The nomogram demonstrated good identification and calibration in both cohorts (C-index ≈ 0.74).
• Nomogram performance (AUC=0.741) exceeded the STOP-Bang score (AUC=0.728) and proved clinically useful.

Conclusions:

• A concise and practical nomogram for OSA risk prediction has been established.
• The nomogram aids clinicians in prioritizing patients for sleep center evaluation.
• This tool supports more efficient and effective sleep medicine workflows.