A sleep robot is a smartphone-controlled device meant to increase sleep quality. It is soft to the touch, meant to be held close to one’s body while sleeping, like a pillow. This allows the user to experience the physical sensation of the robot “breathing” and this is meant to synchronize the user’s breathing rate to a meditative state, allowing the user to fall asleep faster. Unfortunately, a new three-week study did not show any effect of the sleep robot on insomnia problems of the participants nor were there any effects on arousal, anxiety or depression. The study was published in the Journal of Sleep Research.
Insomnia is one of the most common sleep disorders in adults. People suffering from it need longer time to fall asleep, wake up early without wanting to and then feel tired, impaired and distressed throughout the day. It is not fully understood what causes it. Genetic factors are likely to play a role, but it can also be a stressful event or events.
Insomnia typically includes elevated arousal levels in people suffering from it, but thoughts accompanying this arousal may be both positive and negative. Current treatments for insomnia include relaxation techniques meant to counter the elevated arousal levels, Cognitive-Behavioural Therapy (CBT) and sleep medications. Sleep medications are used very commonly to treat insomnia, even though they carry risks of adverse effects or addiction.
“As a clinical psychologist, I am interested in non-pharmacological treatments of insomnia,” said study author Siri Jakobsson Störe, a PhD student at Karlstad University and licensed psychologist. “Sleep medication use is very high in this patient group, with risks of adverse effects and addiction with certain medicines. The gold standard treatment of CBT for insomnia is not effective in all, and there is also a shortage of CBT-therapists. Many people with insomnia test unconventional and unresearched treatments for insomnia, hence, treatments without trustworthy information about how efficacious and safe they are.”
Recently, the Somnox sleep robot started being promoted as a sleep-enhancing device. “The robot’s auditive and physical ‘breathing’ is meant to guide humans into deep breathing and relaxation, and ultimately sleep. Considering what we know about the positive effects on arousal of relaxation in general, and breathing techniques in particular, the sleep robot might help people with insomnia to sleep better,” the researchers said.
To explore whether this sleep robot really has the claimed effects, Støre and her colleagues conducted a three-week-long experiment on persons suffering from insomnia. They screened prospective participants in their study using psychological assessments for insomnia (Insomnia severity Index, ISI) and arousal level (Pre-Sleep Arousal Scale, PSAS). Forty-four applicants with scores indicating sufficiently pronounced insomnia problems were selected for the study.
Participants were randomly divided into two groups. The first group was issued Somnox sleep robots, trained to use them for 10-15 minutes and asked to sleep with them for the next 21 days. The manufacturer of the robot claimed that effects of its use can be expected after a week, so the researchers deemed that 2 weeks after that should be sufficient for effects to become detectable. The robot was set to the “sleeping” program and participants were instructed to sleep with it during the experimental period. The second group was told that it will be on the “waiting list” and received an equivalent three-week period of using the sleep robot after the experiment.
Results showed no difference in insomnia severity between the experimental and the control group after the three-week period. The same was the case with both subjective and objective sleep measures, arousal, depression and anxiety.
“A three-week at-home intervention with the Somnox sleep robot was not found to be an effective method to relieve symptoms of clinical insomnia, somatic arousal, anxiety and depression in adults with insomnia on a group level,” Støre told PsyPost. “Insomnia is, however, a heterogeneous condition, which is why the intervention might be effective in individuals or subgroups of people with clinical insomnia, or in people with subclinical symptoms.”
“Since the method is novel, and since we used a waitlist-control group (as opposed to a psychological placebo or no treatment) – which has been found to produce larger treatment effects – the null results on all outcome measures in our study were very surprising,” she said.
Although it used a rigorous research design, the study did not provide evidence for the effects of sleep robot on any of the measured psychological properties. However, it must be noted that the sample size of the study was very small, and that the effects needed to be quite large to be detectable. It might be that a study on a larger sample could detect some effects, albeit of a smaller magnitude. It could also be that the sleep robot simply does not produce the intended effects.
“The study is relatively small with 44 participants, and we only calculated a power analysis on our main outcome measure (level of insomnia symptoms), which is why a larger study might have yielded more statistically significant differences between the groups (albeit probably not clinically relevant effects),” Støre explained. “It would be interesting to study the sleep robot intervention in combination with a complete CBT program, or with certain components (e.g., psychoeducation or other relaxation techniques). Another idea for the future is to test the effects of the sleep robot in a sleep laboratory.”
“The statistical analysis plan was prospectively published, meaning that we could not ‘look for’ statistically significant results and report them as our main outcome measures,” the researcher added. “I believe that we would see more intervention studies with null results if researchers in my field preregistered not only information about their study but the actual statistical analysis plan to a greater extent, and if journals were more willing to publish null results.”
The study, “The effects of a sleep robot intervention on sleep, depression and anxiety in adults with insomnia—A randomized waitlist-controlled trial”, was authored by Siri Jakobsson Støre, Maria Tillfors, Erik Wästlund, Charlotte Angelhoff, Gerhard Andersson, and Annika Norell-Clarke.