Search Results

BT-001 is a prescription-only digital therapeutic device intended to provide cognitive behavioral therapy to patients 18 years or older with type 2 diabetes. The device targets behavior to aid in the management of type 2 diabetes in patients who are under the care of a healthcare provider. BT-001 provides cognitive behavioral therapy as a treatment that should be used adjunctively with standard of care.

BT-001 is a digital diabetes device that delivers cognitive behavioral therapy (CBT) to a patient with Type 2 Diabetes through an application on the patient's personal Android smartphone. BT-001 is prescription use only and is completely self-directed. The device is designed to allow patients to complete CBT without human support or intervention and is intended to be used in 90-day increments adjunctively to standard of care.

The cognitive behavioral therapy delivered by BT-001 is intended to aid users in making behavioral modifications to better adhere to behaviors associated with the management of diabetes. BT-001 focuses on users understanding barriers to their adhering to known diabetes management behaviors such as eating habits. nutrition, and exercise. The device also provides a way for users to self-report their meals, exercise, medications, and biometrics. When relevant, BT-001 will supply additional notifications for users to contact a healthcare provider, such as when a user logs a blood glucose reading.

This medical device has functions subject to FDA premarket review as well as functions that are not subject to FDA premarket review. For this De Novo request, if the product has functions that are not subject to FDA premarket review. FDA assessed those functions only to the extent that they either could adversely impact the safety and effectiveness of the functions subject to FDA premarket review or they are included as a labeled positive impact that was considered in the assessment of the functions subject to FDA premarket review.

Here's a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided text:

Acceptance Criteria and Device Performance

Study Details

2. Sample Size for Test Set and Data Provenance

• Sample Size: The pivotal study enrolled 726 participants, with 668 downloading the assigned app and completing the onboarding process.
  • Control Group: 343 participants
  • Intervention Group (BT-001): 325 participants
  • Analysis populations: The Intention-to-Treat (ITT) population for day 90 was 313 (control) and 297 (BT-001). For day 180, ITT was 271 (control) and 244 (BT-001). Per Protocol (PP) populations were also evaluated.
• Data Provenance:
  • Country of Origin: Fourteen US clinical sites.
  • Retrospective or Prospective: Prospective, randomized, controlled, parallel-group multicenter clinical trial.

3. Number of Experts Used to Establish Ground Truth and Qualifications

• This device is a digital therapeutic for behavioral therapy and diabetes management. The primary endpoint is HbA1c reduction, which is a biochemical measure directly obtained from laboratory tests, not subjective expert interpretation of images or other data.
• Therefore, the concept of "experts establishing ground truth for the test set" (as typically applied to image recognition or diagnostic AI where human readers provide labels) is not directly applicable in the same way. The ground truth (HbA1c levels) is an objective clinical measurement.
• Qualifications of Experts (for study design/oversight): While not explicitly stated as "experts establishing ground truth," the study involved "healthcare providers" at "fourteen US clinical sites" who had the ability to change glycemic control medications. It is implicit that these were qualified medical professionals overseeing the patients and data collection. The sponsor's medical team would have overseen the study design and data interpretation.

4. Adjudication Method for the Test Set

• Ground Truth Adjudication: As mentioned above, the primary endpoint (HbA1c) is an objective lab measurement, so no adjudication method (e.g., 2+1, 3+1) for establishing ground truth was required for this specific outcome.
• Adverse Event Adjudication: The sponsor reported that four (4) TEAEs in three subjects were adjudicated to be possibly or probably associated with the intervention. The method of this adjudication (e.g., blinded committee, specific criteria) is not detailed in the provided text, but it confirms that some form of "adjudication" was performed for adverse events.

5. If a Multi Reader Multi Case (MRMC) Comparative Effectiveness Study was done

• No, an MRMC comparative effectiveness study was not done. MRMC studies are typically used to evaluate diagnostic imaging devices where multiple human readers interpret cases with and without AI assistance. This study focused on the effectiveness of a digital therapeutic for diabetes management, measured by a change in a physiological biomarker (HbA1c), rather than human interpretation performance.
• Effect size of how much human readers improve with AI vs without AI assistance: Not applicable, as it was not an MRMC study.

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

• The study evaluated the device with a human-in-the-loop (adjunctive use), not in a standalone (algorithm only) capacity. The device is indicated to "be used adjunctively with standard of care" and is "intended to provide cognitive behavioral therapy to patients... who are under the care of a healthcare provider." The clinical trial compared BT-001 plus standard of care against standard of care alone. The device itself is designed for patient self-direction without human support or intervention on the device side, but it is explicitly not a standalone therapy in the broader patient management context.

7. The Type of Ground Truth Used

• Objective Clinical Measurement (HbA1c): The primary ground truth for effectiveness was the change in Glycated Hemoglobin (HbA1c) percentage, a direct and widely accepted clinical biomarker for long-term blood glucose control in diabetes. This is obtained through laboratory testing.
• Self-Reported Data / Clinical Assessment (Adverse Events, PHQ-9): For safety outcomes, adverse events were observed and reported by investigators and PHQ-9 scores (a patient-reported outcome measure for depression) were tracked. The adjudication of adverse events indicates a clinical assessment of their relatedness to the device.

8. The Sample Size for the Training Set

• The document does not specify a separate "training set" for the BT-001 device as it's described in the context of machine learning. The device delivers pre-programmed cognitive behavioral therapy content.
• The clinical trial described is a pivotal validation study, evaluating the effectiveness of the already developed device/therapy. If there was any data-driven iterative development or "training" of components within the CBT program, that information is not provided here. The sample sizes provided (726 participants total, 668 onboarded) refer to the clinical trial population used to validate the device's efficacy and safety.

9. How the Ground Truth for the Training Set Was Established

• As there is no explicitly defined "training set" in the machine learning sense for this device based on the provided text, this question is not fully applicable.
• The "ground truth" for the content of the therapeutic program itself would have been established through established principles of Cognitive Behavioral Therapy (CBT) and clinical guidelines for diabetes management, developed by experts in psychology, diabetes, and behavior science. This foundational knowledge forms the basis of the therapeutic content delivered by the device, rather than being "trained" on a specific dataset in the way an AI algorithm might be.

Ask a specific question about this device