Validity of a New Prediction Model to Identify Patients at Risk for Obstructive Sleep Apnea Hypopnea Syndrome

Area of Science:

• Sleep Medicine
• Clinical Prediction Modeling
• Respiratory Physiology

Background:

• Obstructive sleep apnea-hypopnea syndrome (OSAHS) poses significant health risks.
• Accurate and sensitive screening tools are crucial for early detection and intervention.
• Existing screening methods have limitations in sensitivity for identifying at-risk individuals.

Purpose of the Study:

• To develop and validate a novel clinical prediction model for OSAHS screening.
• To enhance the accuracy of identifying patients requiring polysomnography (PSG).
• To provide a more sensitive tool for clinical decision-making in sleep evaluations.

Main Methods:

• Development of a prediction model using multivariate logistic regression on an 892-patient dataset undergoing PSG.
• Validation of the model on an independent 374-patient dataset using PSG.
• Performance evaluation via receiver operating characteristic analysis, sensitivity, and specificity, compared to Epworth Sleepiness Scale and STOP-Bang scores.

Main Results:

• The developed model identified six significant predictors for OSAHS (apnea-hypopnea index ≥15): male sex, choking/apnea events, hypertension, neck circumference, waist circumference, and BMI.
• The model achieved an area under the curve of 0.753 during development.
• In validation, the model demonstrated a sensitivity of approximately 93% and specificity of 26% for OSAHS detection, outperforming the Epworth Sleepiness Scale (42.26%) and STOP-Bang (56.23%) in sensitivity.

Conclusions:

• The new prediction model offers superior sensitivity for identifying patients at risk of OSAHS compared to the Epworth Sleepiness Scale and STOP-Bang score.
• This model can aid clinical decision-making by prioritizing patients for comprehensive sleep evaluations and PSG.
• The findings suggest a valuable role for this model in improving OSAHS screening efficiency.