Abstract
Insomnia and excessive daytime sleepiness (EDS) often co-occur, despite involving distinct physiological mechanisms. The KANOPÉE application, a fully autonomous virtual agent that collects sleep-related data and delivers personalised behavioural recommendations over a 17-day period, offers a unique opportunity to better understand this unexpected phenotype. Our primary aim was to characterise these 'sleepy insomniacs', and our secondary aim was to evaluate their response to this digital sleep intervention. Among 21,590 participants, individuals with an Epworth Sleepiness Scale score ≥ 11 and an Insomnia Severity Index score ≥ 15 were classified as 'sleepy insomniacs'. Comorbidities (i.e., obstructive sleep apnea syndrome, restless legs syndrome, depression, and sleep medication use) were first described and then excluded for further analyses. At baseline, 4843 (47.9%) of the 10,114 participants with insomnia also reported EDS and were categorised as 'sleepy insomniacs'. Half of this subgroup reported at least one comorbidity, with depression being the most common. After excluding participants with comorbidities, 3239 individuals (44.3%) remained in the 'sleepy insomniacs' category. These individuals were more likely to experience middle or late insomnia symptoms compared to those with insomnia without EDS but responded similarly to the digital sleep intervention. In conclusion, EDS is highly prevalent among individuals with insomnia symptoms. While comorbidities, particularly depression, explained the co-occurrence in approximately half of the sample, a substantial proportion of participants without comorbidities also exhibited this unexpected phenotype. The association with specific insomnia subtypes highlights the need for further investigation. Notably, a 17-day digital sleep intervention proved effective in treating 'sleepy insomniacs'.