Search Results

SAM is a web application used to obtain, track and trend patient data pertaining to the management of anemia, and to provide a schedule of erythropoiesis-stimulating agent (ESA) dosage recommendations to help achieve and maintain target hemoglobin (Hgb) levels in hemodialysis patients. The device is intended to help clinicians manage chronic anemia.

The device is not a substitute for, but rather intended to assist, clinical judgment. The ESA dosing regimen options calculated by this device are intended to inform the optimization of the dosage of ESAs in accordance with their approved labeling in conjunction with clinical history, symptoms, and other diagnostic measurements, as well as the clinicians' judgment. No medical decision should be based solely on the patient Hgb response to dosing regimen options calculated by this device.

SAM is a web application used to obtain, track and trend patient data pertaining to the management of anemia, and to provide a schedule of erythropoiesis-stimulating agent (ESA) dosage recommendations to help achieve and maintain target hemoglobin (Hgb) levels in hemodialysis patients. The device is intended to help clinicians manage chronic anemia.

Healthcare professionals access SAM using a web-enabled application (for example, a web browser or a web-enabled electronic health record system) communicating with the SAM web application server. Patient information is obtained by SAM from healthcare provider Electronic Medical Records. No components of SAM are required to be installed at end user or healthcare provider locations.

SAM estimates individual patient's Hgb response to ESAs. The results of this estimation are used to generate new patient-specific ESA dose recommendation to achieve target Hgb level specified by the physician. The ESA dose recommendation is reviewed by the physician, who after considering any additional relevant information about patient's condition, decides whether to follow or override the presented ESA dose recommendation.

Here's an analysis of the acceptance criteria and study detailed in the provided text for the Smart Anemia Manager (SAM) device:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state formal "acceptance criteria" for the study in a pass/fail sense with numerical targets. Instead, it describes demonstrating "non-inferiority to AMP 1 and 2 in mean hemoglobin, percent in target range and hemoglobin standard deviation." The clinical trials then provide performance measures.

Here's a table summarizing the performance metrics and implicit targets/comparisons:

2. Sample Size Used for the Test Set and Data Provenance

• Simulation Testing:

  • Test Set Sample Size: A pool of 2430 simulated patients were created. Each protocol (SAM, AMP1, AMP2) was run for 26,730 simulations (likely meaning 26,730 unique scenarios or patient instances within the pool).
  • Data Provenance: In silico simulation. The concepts for patient response to ESA were based on published pharmacodynamics (Uehlinger et al. Clin Pharmacol Ther 51:76-89, 1992). This is synthetic data, not real patient data.

• Clinical Studies:

  • Clinical Trial 1 (Randomized, Controlled, Double-Blind):
    • Test Set Sample Size: 62 subjects (52 completed the study).
    • Data Provenance: Prospective clinical trial.
  • Clinical Trial 2 (Case-Controlled & Cross-Sectional):
    • Test Set Sample Size: Not explicitly stated but compared to PhySoft AMS™ over a 45-month follow-up period.
    • Data Provenance: Retrospective, comparing long-term follow-up of SAM and PhySoft AMS™ performance.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and the Qualifications of Those Experts

• Simulation Testing: No human experts were used to establish ground truth. The ground truth was defined by the in silico model of patient response to ESA, based on published pharmacodynamics.
• Clinical Studies: The document does not specify how ground truth was established for the clinical trials beyond stating that Hgb levels were measured and safety events tracked. It implies that standard clinical measurements and outcomes served as the "ground truth." There is no mention of an expert panel to adjudicate individual cases or outcomes.

4. Adjudication Method (for the test set)

• Simulation Testing: Not applicable, as it's a simulated environment.
• Clinical Studies: Not explicitly stated. For Clinical Trial 1, safety events (CSE) were tracked, but the adjudication method for these events is not described (e.g., whether an independent clinical events committee adjudicated them). For Hgb levels, standard laboratory measurements would be considered the ground truth, not requiring expert adjudication.

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 study was not done. This device (Smart Anemia Manager, SAM) is an algorithm that provides dosage recommendations, not an imaging device or diagnostic tool that humans interpret. Therefore, the concept of "human readers improving with AI vs. without AI assistance" as typically studied in MRMC designs for diagnostic AI is not directly applicable here.
• The clinical trials assess the effectiveness of the recommendations provided by SAM compared to standard of care/predicate, rather than how humans perform with those recommendations versus without them. The device produces the recommendation; it doesn't assist human interpretation in the typical MRMC sense.

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

• Yes, the simulation testing represents a standalone evaluation of the algorithm's performance. In the simulation, SAM's dosing recommendations were directly applied to simulated patients, and the outcomes (Hgb levels, variability, ESA dose) were compared to other algorithmic protocols (AMP1, AMP2).
• The clinical trials, by definition, involve human clinicians using the SAM recommendations (or standard of care/predicate recommendations), but SAM itself is intended to generate recommendations. The study's focus is on the clinical outcomes resulting from following those recommendations.

7. The type of ground truth used

• Simulation Testing: Mathematically defined in silico model of patient pharmacodynamics (response to ESA and red blood cell life spans).
• Clinical Studies:
  • Clinical Trial 1: Measured hemoglobin concentrations, and a Composite Safety Event (CSE) based on medical outcomes (mortality, MI, CVA, CHF). This is outcomes data combined with standard clinical measurements.
  • Clinical Trial 2: Measured hemoglobin concentrations and the incidence of high hemoglobin concentrations (above 12.9 g/dL). This is standard clinical measurements.

8. The sample size for the training set

The document does not provide information regarding the sample size for the training set used to develop or train the SAM algorithm. It describes the technology as using an "individualized dose response model" but doesn't detail how this model was trained.

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

Since the document does not specify a training set or its sample size, it also does not explain how the ground truth for any training set was established.

Ask a specific question about this device