Abstract:Background Obstructive sleep apnea (OSA) is highly prevalent among older adults and has been associated with cognitive decline and dementia risk. The suitability of screening tools for detecting OSA in memory clinic settings is unclear. Objective To evaluate the utility and validity of the STOP-Bang questionnaire (SBQ) and pulse oximeter as a screening tool, compared to gold-standard polysomnography (PSG) in older adults attending a memory clinic. Methods Participants aged over 50 with new onset cognitive/mood concerns attended a memory clinic, then completed the SBQ, oximetry, and PSG. The SBQ and oximetry's accuracy in detecting moderate-severe and severe OSA was evaluated using receiver operating curves. Intraclass correlation and Bland-Altman plots compared the oximeter's adjusted oxygen desaturation index (ODI-Ox) and PSG's apnea-hypopnea index (AHI-PSG). Results Of 194 participants (mean age = 65.6, 64 males) who completed PSG, 184 completed the SBQ, and 138 completed oximetry. SBQ demonstrated limited performance for moderate-severe OSA (sensitivity = 52%, specificity = 62%, AUC = 0.600) and severe OSA (sensitivity = 18%, specificity = 87%, AUC = 0.577). Oximetry was satisfactory for moderate-severe OSA (sensitivity = 67%, specificity = 73%, AUC = 0.769) and severe OSA (sensitivity = 50%, specificity = 88%, AUC = 0.730). The diagnostic performance was improved with new cut-offs at ODI-Ox ≥ 11 for AHI-PSG ≥ 15 and ODI-Ox ≥ 20 for AHI-PSG ≥ 30. Bland-Altman plots and intraclass correlation indicated acceptable agreement for oximetry. Conclusions The findings suggest that while the SBQ may be unsuitable to detect moderate or severe OSA for older adults with cognitive impairment, oximetry may be a viable screening tool. Given OSA treatment can optimize sleep and may slow cognitive decline, routine screening for OSA should be part of memory clinic assessments.

Screening for obstructive sleep apnea in the memory clinic: A comparison of questionnaires, pulse oximetry, and polysomnography