Introduction: Excessive daytime sleepiness is an increasingly frequent condition among older adults with comorbidities and living in nursing homes (NH). This study investigated associations between participants' characteristics and excessive daytime sleepiness (EDS); the ability of the Epworth Sleepiness Scale (ESS) scores, EDS and EDS severity levels to predict mortality at 12 months of follow-up; and the optimal cut-off for ESS to predict mortality among NH residents.
Methods: Prospective and cross-sectional analysis in a prospective study including older adults permanently residing in 12 NHs from South Australia. Baseline characteristics including the ESS were collected and mortality at 12 months was assessed. Logistic regression analysed associations between participants' characteristics and EDS (ESS>10). Kaplan-Meier cumulative survival estimates followed by log-rank and adjusted Cox proportional hazards models explored associations of ESS scores, EDS and EDS severity levels with time-to-incident death. Receiver operator (ROC) analysis assessed the best cut-off for ESS to predict mortality risk.
Results: 550 participants (mean [SD] age, 87.7 [7.2] years; 968 [50.9%] female) were included in the final analysis. Malnutrition (adjusted odds ratio [aOR] 2.02, 95% confidence interval [CI] 1.13-3.61), myocardial infarction (aOR 1.91, 95% CI 1.20-3.03), heart failure (aOR 2.85, 95% CI 1.68-4.83), Parkinson’s disease (aOR 2.16, 95% CI 1.04-4.47) and severe dementia (aOR 8.57, 95% CI 5.25-14.0) were associated with EDS. Kaplan-Meier analyses showed reduced survival among participants with EDS (log-rank test: χ2 = 25.25, p < 0.001). EDS predicted increased mortality risk (HR=1.63, 95% CI 1.07-2.51, p=0.023). ESS score of 10.5 (>10) was the best cut point predicting mortality risk (area under the curve (AUC)= 0.62).
Conclusions: Mortality among older adults living in NHs can be predicted by increasing daytime sleepiness, which is associated with commonly observed conditions in institutionalised individuals, such as malnutrition and neurodegenerative and cardiac comorbidities. Monitoring sleep-wake disturbances by simple methods can be an effective strategy to identify NH residents who are more susceptible to EDS-related poor quality of life and reduced survival. Measuring EDS levels as a general health indicator can sustain improved care practices by triggering the assessment and management of potentially treatable EDS risk factors and reducing the impact of reversible EDS consequences. In the context of non-reversible associated conditions, such as advanced dementia, increasing EDS can precipitate discussions and considerations about end-of-life care. Future studies targeting interventions to reduce EDS can result in new evidence-based strategies to improve quality of life and reduce the morbimortality observed in NHs.