Digital Phenotyping in Insomnia: Real-Time Smartphone Assessment Detects Daytime Benefits of Suvorexant Missed by Traditional Scales

The Diagnostic Challenge of Daytime Insomnia Symptoms

Insomnia disorder is defined not only by nocturnal sleep disturbance but also by significant daytime impairment. Diagnostic criteria, including those in the DSM-5-TR and ICSD-3, emphasize negative mood, fatigue, and cognitive dysfunction as core features of the condition. Historically, clinical research has struggled to quantify these daytime symptoms accurately. Most trials rely on retrospective questionnaires, such as the Insomnia Severity Index (ISI) or the Epworth Sleepiness Scale, which require patients to summarize their experiences over the previous week or month. This approach is inherently susceptible to recall bias, where recent or extreme experiences disproportionately influence the reported data, potentially masking subtle but clinically meaningful therapeutic effects.

Overcoming Recall Bias with Ecological Momentary Assessment

To address the limitations of retrospective reporting, researchers have turned to Ecological Momentary Assessment (EMA). EMA involves repeated sampling of a participant’s current behaviors and states in real-time within their natural environment. By utilizing smartphone technology, clinicians can capture high-resolution longitudinal data that reflects the dynamic nature of symptoms throughout the day. A recent randomized clinical trial led by Wickwire et al., published in JAMA Network Open, investigated whether this “digital phenotyping” approach could detect the effects of the insomnia medication suvorexant on daytime symptoms more effectively than traditional methods.

Study Design and Methodology

This double-blind, placebo-controlled randomized clinical trial was conducted entirely remotely between October 2023 and August 2024. The study recruited 40 older adults (mean age 67.9 years; 90% women) who met the criteria for chronic insomnia. Participants were required to have an Insomnia Severity Index (ISI) score of 15 or higher and own a smartphone. Following a baseline assessment, participants were randomized 1:1 to receive either suvorexant or a placebo. The dosing regimen for suvorexant was 10 mg for the first two nights, followed by 14 nights at 20 mg.

The primary innovation of the trial was the use of the Daytime Insomnia Symptoms Scale (DISS), administered via smartphone four times per day—at 10:00 AM, 1:00 PM, 4:00 PM, and 7:00 PM. This resulted in 64 total administrations over the 16-day treatment period. These real-time assessments were compared against traditional post-treatment retrospective questionnaires evaluating insomnia severity, sleepiness, fatigue, anxiety, and depression.

Key Findings: The Superiority of Real-Time Assessment

The results of the trial highlighted a stark contrast between traditional metrics and real-time EMA. Both groups showed high engagement, with an overall EMA survey completion rate of 93.3%, demonstrating the feasibility of intensive digital monitoring in a geriatric population.

General Insomnia Severity

Suvorexant demonstrated clinical efficacy in reducing overall insomnia severity. The mean change in the ISI for the suvorexant group was -9.6 (SD 5.4) compared to -5.5 (SD 6.8) for the placebo group. The estimate of the treatment effect was 4.1 (SE 1.9), which was statistically significant (P = .04) with a moderate-to-large effect size (0.66).

Daytime Symptoms: Retrospective vs. EMA

When looking at daytime symptoms through the lens of retrospective questionnaires, no statistically significant differences were found between the suvorexant and placebo groups regarding fatigue, sleepiness, or mood. Had the trial relied solely on these traditional endpoints, it might have concluded that suvorexant had no impact on daytime functioning.

However, the EMA data provided a different picture. The real-time DISS scores revealed significant between-group differences in subjective cognition (χ24 = 11.12; P = .03) and fatigue (χ24 = 21.43; P = .003) at various points during the day. Specifically, the EMA was sensitive enough to detect that suvorexant improved cognitive clarity and reduced fatigue levels at specific intervals where retrospective recall failed to register an effect. This suggests that the medication’s benefits on daytime vitality are nuanced and may vary by time of day—a granularity that is lost when symptoms are averaged over a week in a questionnaire.

Expert Commentary: Mechanistic Insights and Clinical Utility

The findings underscore the unique pharmacological profile of suvorexant, a dual-orexin receptor antagonist (DORA). Unlike traditional GABAergic hypnotics, which act as general central nervous system depressants and can cause residual “hangover” effects, DORAs target the wake-promoting orexin system. By blocking the signals that keep a patient awake, DORAs facilitate a more physiological transition to sleep. The evidence from this trial suggests that by improving nocturnal sleep without residual daytime sedation, suvorexant may enhance daytime cognitive function and energy levels, provided those levels are measured with high-frequency tools like EMA.

From a methodological standpoint, this study serves as a proof-of-concept for the integration of digital health tools in clinical trials. The discrepancy between the ISI results and the EMA results suggests that many previous insomnia trials may have underreported the daytime benefits of pharmacotherapy due to insensitive measurement tools. For clinicians, this reinforces the importance of asking patients about their symptoms at specific times of the day rather than asking for a general weekly summary.

Study Limitations

While the study provides compelling evidence for EMA, several limitations must be considered. The sample size was relatively small (n=40), and the population was predominantly female, which may limit generalizability. Additionally, the study period was short (16 days), and further research is needed to determine if these daytime benefits are sustained over long-term treatment. Finally, because all assessments were remote, there was no objective sleep measurement (such as polysomnography), though the study’s focus was specifically on subjective daytime experience.

Conclusion

This randomized clinical trial provides critical evidence that smartphone-based ecological momentary assessment is a superior tool for capturing the daytime impact of insomnia treatments. In older adults, suvorexant not only improved nocturnal insomnia severity but also yielded detectable improvements in daytime fatigue and cognition that traditional retrospective scales missed. As the field of sleep medicine moves toward personalized care, the use of real-time digital monitoring may become a standard for both evaluating treatment efficacy and managing patient outcomes in clinical practice.

Funding and Trial Registration

This study was supported by grants and resources from the academic center involved. ClinicalTrials.gov Identifier: NCT05908526.

References

Wickwire EM, Zhou J, Chen S, Wilckens KA, Steenbergh TA, Buysse DJ, Verceles AC. Smartphone-Based Real-Time Assessment of Daytime Insomnia Symptoms With Suvorexant: A Randomized Clinical Trial. JAMA Netw Open. 2026;9(1):e2550186. doi:10.1001/jamanetworkopen.2025.50186.