Smartphone-Based Virtual Agents to Help the General Population Concerned by Sleep Complaints During COVID-19 Confinement: A Feasibility Study (Preprint)

Pierre Philip, Lucile Dupuy, Charles M. Morin, Etienne de Sevin, Stéphanie Bioulac, Jacques Taillard, Fuschia Serre, Marc Auriacombe, Jean-Arthur Micoulaud-Franchi
Preprint JMIR. 2020-09-11; :
DOI: 10.2196/preprints.24268


BACKGROUND
The COVID-19 crisis induces psychosocial stress and sleep complaints that require early management. KANOPEE is a smartphone-based application, providing an interaction with a virtual agent dedicated to screen and deliver behavioral interventions to fight sleep disorders.

OBJECTIVE
This paper describes the feasibility study of this application, during the context of COVID-19 confinement in France.

METHODS
2,069 users of aged 18 years and over downloaded the app during the inclusion period (between 22 April and 5 May 2020). Users first answered a screening interview based on the Insomnia Severity Index (ISI) that was conducted by the virtual agent. If participants were positive for insomniac complaints (ISI > 14), they could join a two-stage intervention program: a) complete an electronic sleep diary for one week, and b) follow personalized sleep recommendations for 10 days. Measures collected included socio-demographic information, ISI and sleep/wake schedules; and acceptance and trust of the agent.

RESULTS
Of all participants, 80% (n=1,574) completed the screening interview with the virtual agent. The virtual agent was well accepted by users regarding its usability, satisfaction, benevolence, and credibility. Of the 773 screened subjects who reported sleep complaints (ISI>14), 214 of them followed the first step of the intervention (34%). Of these, 47 (31%) followed the second step. Users who completed step one found that their insomnia complaints (mean scores: baseline ISI: 18.56; ISI after step one: 15.99; P<.001) and nocturnal sleep improved significantly after one week. Subjects who completed step 2 also showed an improvement compared to the initial measure (mean scores: baseline ISI: 18.87; ISI after step two: 14.68; P<.001). The most severely affected subjects (ISI >21) did not respond to either intervention.

CONCLUSIONS
These preliminary results show that KANOPEE is a promising solution to screen populations for sleep complaints, and that it provides practical and effective behavioral advice for subjects reporting moderately severe insomnia.

Auteurs Bordeaux Neurocampus