Development and validation of a model for diagnosis of obstructive sleep apnoea in primary care

Background and objective Use of in-laboratory polysomnography (PSG) to diagnose obstructive sleep apnoea (OSA) is cost resource intensive. Questionnaires, physical measurements home monitors have been studied as potential simpler alternatives. This study aimed develop a diagnostic model for OSA use in primary care. Methods Primary care practitioners were trained recognize symptoms recruited patients based on the clinical need investigate OSA. Assessment was by symptom questionnaires, anthropomorphic measurements, digital facial photography, single-channel nasal flow monitor (Flow Wizard©, DiagnoseIT, Sydney, Australia) worn at 3 nights. The PSG reference test, with defined apnoea–hypopnoea index (AHI) ≥10 events/h. Results In development phase, 25 315 whom they suspected OSA, which 57% had AHI≥10 22% AHI≥30. Published questionnaires provided low moderate prediction (area under curve [AUC] 0.53–0.73). alone yielded high accuracy predicting AUC 0.87. Sensitivity 0.87 specificity 0.77 threshold respiratory event (REI) 18 A adding age, gender, BMI REI only modestly improved (AUC 0.89), similar (0.88) confirmed validation population 114 patients. Conclusion Sleep can be diagnosed setting combination judgement portable test outcomes.