O045 Anxiety and daytime sleepiness scores have a weak but statistically significant correlation with apnoea-hypopnoea index

Abstract

Abstract Background Previous datasets demonstrate inconsistent relationships between apnoea-hypopnoea index (AHI) and questionnaire measures of daytime sleepiness, anxiety and depression. Methods 1149 consecutive diagnostic polysomnograms at a quaternary hospital were retrospectively analysed (2020–2021). Relationships between age, sex, AHI, Epworth Sleepiness Scale (ESS) and Hospital Anxiety and Depression Scales (HADS-A and HADS-D) were reviewed. Progress to date:Mean age was 47.8+/-15.7 years, with male gender bias (59%). 49.9% had elevated HADS-A (>7; mean 8.1+/-4.5). 33.7% had elevated HADS-D (>7; mean 6.2+/-4.0). 29.6% had sleepiness (ESS >10; mean 7.9+/-5.0). Mean AHI was 23.5+/-27.9 events/hour. 69.7% had obstructive sleep apnoea (OSA); 45.0% were moderate-severe (AHI >14/h). HADS-A (9.1 versus 7.3; P<0.001; 95% CI [-2.32, -1.25]), HADS-D (6.9 versus 5.8; P<0.001; 95% CI [-1.62, -0.67]) and ESS (8.2 versus 7.7; P=0.039; 95% CI [-1.21, -0.03]) were higher in females. Males had greater OSA severity (AHI 27.0/h versus 18.6/h; p<0.001; 95% CI [5.15, 11.67]). Pearson’s tests demonstrated a statistically significant but weak positive correlation between AHI and HADS-A (P=0.021, R=0.07, N=1096), and AHI and ESS (P=0.042, R=0.06, N=1135). AHI and HADS-D showed no correlation. ESS weakly correlated with HADS-A (P<0.001, R=0.237, N=1104). Intended outcome and impact:Severity of sleep apnoea as defined by AHI only explains a small proportion of the variance in daytime sleepiness and anxiety as measured by the ESS and HADS-A, respectively, with weak linear relationships demonstrated. Neither ESS nor HADS-A were helpful in predicting the presence of OSA. Further study is required to determine optimal polysomnographic correlates of sleep apnoea symptoms.