General Practice Sleep Scale – The “GPSS” – A proposed new tool for use in General Practice for risk assessment of Obstructive Sleep Apnoea

Abstract

Background

This pilot study investigated a new simplified OSA screening tool that could be used in primary care/GP settings – the “GPSS” tool – “General Practice Sleep Scale” and compared against common existing OSA screening tools.

Methods

A convenience sample of patients attending the respiratory and sleep clinic in the Northern Territory of Australia were included if they completed the GPSS prior to undergoing a diagnostic polysomnography. The GPSS contained 9 questions to provide information on: sex, age, body mass index, neck circumference, snoring, witnessed apnoeas, morning tiredness, daytime sleepiness and presence of hypertension/diabetes/heart disease/depression. Presence of OSA was defined as an apnoea-hypopnoea index of ≥15/hour. The GPSS scoring was developed via log odds of regression predictions for each GPSS question upon OSA.

Results

159 patients (65 % male, median age 45 years) were enrolled. A minimum score of 1 was assigned to GPSS questions, up to 5 for the strongest predictor (neck circumference). The median total GPSS score was 13 (IQR 9, 16) (maximum 22) and correlated strongly with OSA (AUC 0.812 (95 % CI 0.744, 0.881)). Categorised into low (0–7), moderate (8–13) or high risk (>13), a moderate or severe score had sensitivity 100, specificity 34.9 %. The GPSS significantly outperformed the Epworth Sleepiness Scale, Berlin questionnaire and OSA-50, and was comparable but slightly improved against the STOP-Bang.

Conclusions

The proposed GPSS tool could be of use in general practice settings. Further prospective research is warranted to test the applicability and adaptability of the GPSS tool in wider population settings.