Search Results

Sleepio is a digital therapeutic intended for the treatment of chronic insomnia disorder as an adjunct to usual care in patients aged 18 and older. Sleepio is a prescription device delivering Cognitive Behavioral Therapy for Insomnia (CBT-I) and can be made available on the order of a licensed healthcare provider.

Sleepio is a digital therapeutic for the treatment of chronic insomnia / insomnia disorder. Sleepio treats chronic insomnia disorder by delivering evidence-based techniques targeting the cognitive and behavioral factors that maintain insomnia and chronic sleep problems. Patient experience is tailored based on symptoms and daily sleep tracking. In addition to core therapeutic components, there is in-the-moment therapeutic content for help falling asleep. Content is delivered via smartphone and tablet applications (iOS and Android), as well as via web. Sleepio is intended as an adjunct to usual care treatment for chronic insomnia / insomnia disorder by a healthcare provider. Healthcare providers have access to a dashboard to track patient engagement with Sleepio.

Based on the provided text, the device in question is Sleepio, a digital therapeutic for the treatment of chronic insomnia disorder. Here's a breakdown of the acceptance criteria and the study proving the device meets them:

Disclaimer: The provided document is an FDA 510(k) summary, which focuses on demonstrating substantial equivalence to a predicate device rather than setting explicit acceptance criteria for this specific device's performance in a new, de novo fashion. The clinical data presented serves to support the substantial equivalence claim by showing similar or superior performance to standard care (Sleep Hygiene Education, SHE) and implicitly, to the predicate. Therefore, the "acceptance criteria" table below is inferred from the goals of the clinical study and the claims made about the device's efficacy in the context of demonstrating substantial equivalence for an FDA 510(k) submission.

1. Table of inferred acceptance criteria and the reported device performance

2. Sample size used for the test set and the data provenance

• Test set sample size: The clinical effectiveness was evaluated in the CrEDIT trial, a two-arm, parallel group, randomized controlled trial (RCT) with 336 adults.
  • Sleepio group (n=168)
  • Online sleep hygiene education (SHE) group (n=168)
• Data provenance: Participants were recruited from across the United States via social media. The study was prospective (Randomized Controlled Trial).

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

The document does not specify the use of human experts to establish ground truth for the test set in the conventional sense of image or diagnostic adjudication. The ground truth for insomnia was established clinically based on self-reported patient assessments using standardized questionnaires (Insomnia Severity Index, ISI) and sleep diaries (SOL, WASO), and a DSM-5 diagnosis of insomnia disorder as criteria for study inclusion. The study evaluated the effectiveness of a therapeutic intervention, not the diagnostic accuracy of an AI model against expert consensus.

4. Adjudication method for the test set

Not applicable. As noted above, this study measures the therapeutic effectiveness of Sleepio based on patient-reported outcomes, not the performance of a diagnostic or classification algorithm that would require an adjudication method for ground truth.

5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

No, an MRMC comparative effectiveness study was not done. This was a clinical trial evaluating a digital therapeutic directly delivered to patients for treatment, not an AI-assisted diagnostic tool for human readers. Therefore, the concept of "human readers improve with AI vs without AI assistance" does not apply.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

Yes, in a way. Sleepio is a digital therapeutic that delivers Cognitive Behavioral Therapy for Insomnia (CBT-I) directly to patients via smartphone/tablet apps and web. While it's a "prescription device" and "can be made available on the order of a licensed healthcare provider" who "have access to a dashboard to track patient engagement," the core therapeutic delivery and patient interaction are directly from the algorithm/software. The study design directly assesses the effect of this standalone digital intervention compared to sleep hygiene education.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

The ground truth or clinical endpoints were based on patient-reported outcomes data derived from:

• Insomnia Severity Index (ISI) scores: A standardized, self-reported questionnaire.
• Sleep diary data: Self-reported sleep onset latency (SOL) and wake after sleep onset (WASO).
• DSM-5 diagnosis of insomnia disorder: Used as an inclusion criterion for participants.
• Sleep Condition Indicator (SCI-8): A secondary outcome measure.

8. The sample size for the training set

The document does not specify a training set sample size in the context of an AI/ML model for this device. Sleepio delivers Cognitive Behavioral Therapy for Insomnia (CBT-I), which is an evidence-based therapeutic approach. The "training" for such a device would typically relate to the development of the CBT-I content and algorithms based on established clinical guidelines and research in psychology and sleep medicine, rather than a machine learning training dataset in the typical sense for image recognition or diagnostic AI. The software verification and validation are noted, but this refers to engineering and quality assurance processes, not a machine learning training curriculum.

9. How the ground truth for the training set was established

Not applicable directly, as this is not described as a device that uses a trained AI model in the sense of predictive or diagnostic capabilities. Sleepio is a digital delivery mechanism for an established therapy (CBT-I). The "ground truth" for its therapeutic approach is the established principles and efficacy of CBT-I as validated in clinical literature and practice over many years. The clinical study (CrEDIT trial) served as a validation study to demonstrate that Sleepio's digital implementation of CBT-I is effective, rather than a study to establish ground truth for a novel AI algorithm's training.

Ask a specific question about this device

Somryst is a prescription-only digital therapeutic intended to provide a neurobehavioral intervention (Cognitive Behavioral Therapy for Insomnia - CBT-I) in patients 22 years of age and older with chromic insomnia. Somryst treats chronic insomnia by improving a patient's insomnia symptoms.

Somryst is a 9-week prescription digital therapeutic (computerized behavioral therapy) used in the treatment of chronic insomnia. Somryst is based on principles of Cognitive Behavioral Therapy (CBT) for Insomnia, Sleep Restriction, and other proven psychosocial treatment elements, which are delivered in a sequence of "cores" of patient education, training and skill building. The therapy is delivered via a mobile application intended to be used on a patient's mobile device and consists of text, video, animation and graphics. Clinicians, as part of a patient's general treatment program, have access to a clinician dashboard that shows patient utilization and engagement with the application.

Here's an analysis of the acceptance criteria and the studies that prove Somryst meets these criteria, based on the provided document:

1. Table of Acceptance Criteria and Reported Device Performance

The FDA clearance document for Somryst does not explicitly list "acceptance criteria" in a typical quantitative pass/fail format with predefined thresholds for performance metrics. Instead, it demonstrates the device's efficacy by comparing its performance to a control group in several clinical studies. The implicit acceptance criterion is that Somryst should significantly improve insomnia symptoms compared to usual care or a control.

Below is a summary of the reported device performance from the two principal clinical studies, showing the statistically significant improvements.

2. Sample Size for Test Set and Data Provenance

The "test set" in this context refers to the participants in the two clinical trials where the effectiveness of Somryst (referred to as SHUTi in the studies) was evaluated.

• "The GoodNight Study" (ANU Study):

  • Sample Size (Test Set):
    • UC+SHUTi group: N=574
    • UC+Control group: N=575
    • Total participants mentioned in some tables (e.g., ISI, Week 9): UC+SHUTi (250), UC+Control (342). These are "available patient data," possibly reflecting those who completed the assessments at that specific timepoint.
  • Data Provenance: Australia (Australian National University, Black Dog Institute, University of Sydney). Prospective randomized controlled trial.

• Pivotal Trial (UVA Study - NCT01438697):

  • Sample Size (Test Set): Total N=303 adults with chronic insomnia.
    • UC+SHUTi group: N=151
    • UC+Control group: N=152
    • Total participants mentioned in some tables (e.g., ISI, Week 9): UC+SHUTi (133), UC+Control (142).
  • Data Provenance: United States (University of Virginia). Prospective randomized controlled trial.

3. Number of Experts and Qualifications for Ground Truth

The document does not describe the use of experts to establish a "ground truth" for the test set in the traditional sense of image annotation or diagnostic consensus. Instead, the ground truth for measuring insomnia severity and outcomes was based on:

• Patient-reported outcomes: Insomnia Severity Index (ISI) questionnaires. The ISI is a validated scale.
• Patient-reported sleep diaries: These were administered online and collected for 10 days within a 2-week window at each assessment time point. These diaries were used to calculate objective-like sleep parameters such as Sleep Onset Latency (SOL) and Wake After Sleep Onset (WASO).

The ISI itself incorporates a defined scoring system for severity and remission, which acts as a standardized "ground truth" established by clinical consensus (as referenced by published psychometric studies of the ISI).

4. Adjudication Method for the Test Set

No explicit adjudication method is mentioned. The primary outcome measures (ISI scores, sleep diary data for SOL and WASO) are quantitative and collected directly from patients. Clinical intervention studies like these typically rely on these validated instruments rather than expert adjudication of individual case outcomes after the trial.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No, an MRMC comparative effectiveness study was not done. Somryst is a computerized behavioral therapy device designed to be used by patients, not an AI diagnostic tool that assists human readers in interpreting medical images or data. Therefore, the concept of assessing how much human readers improve with AI assistance is not applicable to this device.

6. Standalone Performance

Yes, a standalone performance study was done. The entire evaluation of Somryst (SHUTi in the studies) was based on its standalone performance in improving insomnia symptoms in patients directly. The comparator was either "Usual Care" plus an attention-matched digital control or an un-tailored digital education program, allowing for the isolation of Somryst's therapeutic effect. The device (SHUTi) itself delivered the intervention without human-in-the-loop assistance for the core therapy delivery. Clinicians had access to a dashboard, but this was for monitoring and management, not direct therapeutic intervention via the device.

7. Type of Ground Truth Used

The ground truth used was primarily patient-reported outcomes and validated psychometric scales:

• Insomnia Severity Index (ISI): A validated questionnaire for assessing insomnia severity. The definitions of "responder" (>=7-point ISI reduction) and "remitter" (ISI score < 8) are based on established clinical criteria derived from psychometric studies of the ISI.
• Sleep diaries: Patient-reported logs of sleep parameters like sleep onset latency (SOL) and wake after sleep onset (WASO). These provide quantitative, if subjective, measures of sleep behavior.

8. Sample Size for the Training Set

The document does not provide a specific "training set" for the Somryst algorithm in the context of machine learning. Somryst is described as a digital therapeutic based on "principles of Cognitive Behavioral Therapy (CBT) for Insomnia, Sleep Restriction, and other proven psychosocial treatment elements." This implies that the therapy logic and content are based on established medical and psychological knowledge (CBT-I protocols) rather than being developed through a data-driven machine learning training process.

The clinical studies described (ANU and UVA studies) are effectiveness studies, essentially serving as a "test set" to validate the established therapeutic approach delivered by the device.

9. How Ground Truth for the Training Set Was Established

As noted above, there isn't a traditional "training set" with ground truth in the machine learning sense. The "ground truth" for the content and principles of Cognitive Behavioral Therapy for Insomnia (CBT-I) and other psychosocial treatment elements, which form the basis of Somryst, would have been established over decades of clinical research, expert consensus, and efficacy studies in the field of sleep medicine and psychology. It represents established medical knowledge and best practices for treating insomnia.

Ask a specific question about this device