Weighted STOP-Bang and screening for sleep-disordered breathing

Area of Science:

• Sleep Medicine
• Respiratory Medicine
• Diagnostic Tool Development

Background:

• The STOP-Bang tool is used to assess the risk of sleep-disordered breathing (SDB).
• Conventional STOP-Bang scoring uses dichotomous variables.
• There is a need to improve the diagnostic characteristics of the STOP-Bang tool.

Purpose of the Study:

• To evaluate if modifying the STOP-Bang tool by weighting variables can enhance its predictive performance for SDB.
• To compare the test characteristics of a modified STOP-Bang score with the conventional score.

Main Methods:

• Analysis of data from the Sleep Heart Health Study (SHHS) with derivation (n=1667) and validation (n=4774) datasets.
• Linear regression was used to determine coefficients for variable weighting.
• Continuous variables (BMI, age, neck circumference) were incorporated into modified scoring models: weighted STOP-Bang (wSTOP-Bang) and continuous STOP-Bang (cSTOP-Bang).
• Receiver operating characteristic (ROC) curves were constructed to assess performance.

Main Results:

• The continuous STOP-Bang (cSTOP-Bang) model demonstrated a higher area under the curve (AUC) of 0.738 compared to conventional STOP-Bang (0.706) and weighted STOP-Bang (0.69).
• Sensitivities were similar across all models (approximately 93.2%).
• cSTOP-Bang exhibited significantly higher specificity (31.8%) than conventional STOP-Bang (23.2%) and wSTOP-Bang (23.6%).
• cSTOP-Bang also showed a higher positive likelihood ratio (1.36) compared to the other models.

Conclusions:

• Weighting STOP-Bang variables and utilizing continuous data for BMI, age, and neck circumference improves diagnostic accuracy for SDB.
• The modified continuous STOP-Bang score maintains high sensitivity while enhancing specificity and positive likelihood ratio.
• This refined scoring method offers improved predictive value for sleep-disordered breathing.