400 Utility of the STOP Questionnaire in Predicting Sleep Disordered Breathing in Older Women Veterans

Abstract

Abstract Introduction Sleep disordered breathing (SDB) is underdiagnosed in older women, despite a significant increase in SDB prevalence post-menopause. Few studies have assessed the diagnostic accuracy of SDB screening questionnaires in older women, particularly older Women Veterans (WV). WV have higher rates of SDB compared to non-Veteran women and are particularly vulnerable to sleep disorders in general. We examined the diagnostic accuracy of the STOP questionnaire compared to home sleep apnea testing (HSAT) that includes sleep time estimation (i.e., WatchPAT) in older WV. Methods Cross-sectional baseline data obtained from chart review were combined from two behavioral sleep intervention studies targeting WV with sleep difficulties (i.e., insomnia symptoms) or SDB risk factors (e.g., hypertension, obesity). A total of 136 older WV (50-91y; age=60.0±7.8y) completed the STOP questionnaire (yes/no: snoring, tiredness, observed breathing pauses, and high blood pressure [BP]) and had an apnea-hypopnea index (AHI) available from their baseline HSAT (WatchPAT). Sensitivity, specificity, and positive and negative likelihood ratios (+LR/-LR) were calculated to characterize the diagnostic accuracy of STOP≥2 for AHI≥5 (mild SDB) or AHI≥15 (moderate SDB). Results 70.6% (n=96) of participants endorsed a STOP≥2, 83.8% (n=114) demonstrated an AHI≥5 and 46.3% (n=63) demonstrated an AHI≥15. For AHI≥5, sensitivity was 73.7% (95% CI=64.6,81.5%), specificity was 45.5% (95% CI=24.4,67.8%), +LR was 1.35 (95% CI=0.91, 2.01), and -LR was 0.58 (95% CI=0.33,1.00). For AHI≥15, sensitivity was 76.2% (95% CI=63.8,86%), specificity was 34.2% (95% CI=23.5,46.3%), +LR was 1.16 (95% CI=0.93,1.44), and -LR was 0.70 (95% CI=0.30,1.20). Conclusion The likelihood ratios for STOP≥2 limited the utility of the STOP vs. an HSAT system with sleep scoring in determining AHI. While the STOP correctly identified 3/4 of older WV with SDB on WatchPAT, it correctly identified <50% of older WV without SDB. Screening measures that better capture predictors of moderate SDB in women at risk for SDB are needed, especially in older women who may not present clinically with the common SDB symptoms (i.e. snoring, tiredness, observed breathing pauses, and high BP). STOP compared to polysomnography studies are also needed. Support (if any) VA HSR&D IIR-13–058, IIR 16–244 and RCS 20–191; NIH/NHLBI K24 HL143055, VAGLAHS GRECC, VA Office of Academic Affiliations, and AASM Foundation.