Engaging Refugees With a Culturally Adapted Digital Intervention to Improve Sleep: A Randomized Controlled Pilot Trial

Abstract

Refugees are exposed to multiple stressors affecting their mental health. Given various barriers to mental healthcare in the arrival countries, innovative healthcare solutions are needed. One such solution could be to offer low-threshold treatments, for example by culturally adapting treatments, providing them in a scalable format, and addressing transdiagnostic symptoms. This pilot trial examined the feasibility, acceptance, and preliminary effectiveness of a culturally adapted digital sleep intervention for refugees. Sixty-six refugees participated, with 68.2% of them seeking psychological help for the first time. Only three participants did not show clinically significant insomnia severity, 93.9% reported past traumatic experiences. Participants were randomly assigned to the intervention group (IG) or the waitlist control group (CG). Insomnia severity, measured by the Insomnia Severity Index, and secondary outcomes (sleep quality, fear of sleep, fatigue, depression, wellbeing, mental health literacy) were assessed at baseline, 1 and 3 months after randomization. The self-help intervention included four modules on sleep hygiene, rumination, and information on mental health conditions associated with sleep disturbances. 66.7% of the IG completed all modules. Satisfaction with the intervention and its perceived cultural appropriateness were high. Linear multilevel analyses revealed a small, non-significant intervention effect on insomnia severity of Hedge's g = 0.28 at 3-months follow-up, comparing the IG to the CG [F2, 60 = 0.88, p = 0.421]. This non-confirmatory pilot trial suggests that low-threshold, viable access to mental healthcare can be offered to multiple burdened refugees by culturally adapting an intervention, providing it in a scalable format, and addressing a transdiagnostic symptom.

Smartphone-delivered mental health care interventions for refugees: A systematic review of the literature

“…Furthermore, the identified interventions were at different stages of their evaluation process. We included two RCTs (Nickerson et al, 2019 ; Röhr et al, 2021 ) and two pilot RCTs (Lindegaard et al, 2021b ; Spanhel et al, 2022 ), three feasibility studies (Holmes et al, 2017 ; Lindegaard et al, 2021a , 2022 ), two qualitative studies on the development process (Burchert et al, 2019 ), and the cultural adaptation of interventions (Spanhel et al, 2019 ). Moreover, two peer-reviewed study protocols (Böge et al, 2020 ; Fischer et al, 2021 ) and three interventions without any published study reports were identified.…”

“…According to the respective stage of evaluation, different clinical and nonclinical outcomes were reported: Four of the identified study reports collected data on clinical outcomes and tested efficacy in the context of RCTs (Nickerson et al, 2019 ; Röhr et al, 2021 ) or pilot RCTs (Lindegaard et al, 2021b ; Spanhel et al, 2022 ). One of the included study protocols described a planned RCT on the efficacy of a stepped and collaborative care model including the smartphone-delivered intervention as one of several low-threshold stand-alone interventions (Böge et al, 2020 ).…”

“…One study reported a significant difference between the intervention and control group in the pre–post comparison (Lindegaard et al, 2021b ). Three studies did not report significant results regarding their primary outcomes (Nickerson et al, 2019 ; Röhr et al, 2021 ; Spanhel et al, 2022 ). Additionally, most of the studies reported several secondary clinical outcomes, with only few significant results.…”

“…Additionally, most of the studies reported several secondary clinical outcomes, with only few significant results. At the same time, the control conditions in the studies differed greatly: Three studies applied waitlist control conditions (Nickerson et al, 2019 ; Lindegaard et al, 2021b ; Spanhel et al, 2022 ), one used an active control condition (Röhr et al, 2021 ), and the study protocol reported a planned treatment-as-usual control condition (Böge et al, 2020 ).…”

“…Dropout rates also varied widely between studies. While most reported low dropout rates (2.9–18.5%; Nickerson et al, 2019 ; Röhr et al, 2021 ; Spanhel et al, 2022 ), some reported high rates (37.9–80%; Lindegaard et al, 2021b , 2022 ). Lastly, few participants worked on all modules offered throughout the interventions or used the interventions as recommended.…”