Search Results

The Home Access® A1C Test is an in vitro test method for the quantitative measurement of Hemoglobin A1c using capillary blood collected from the fingertip, collected onto filter paper via the Home Access collection cassette. The Home Access AIC Test is for measurement of HbA1c on blood specimens which can be collected at the patient's home or in a healthcare professional setting and delivered to the laboratory by mail. Measurements obtained through this method can be used for monitoring the long-term control of blood sugar (glucose) in people with diabetes.

This test is not to be used to diagnose or screen for diabetes. Not for use on neonates.

The Home Access® A1C Test includes a micro-blood specimen collection kit. The collection kit is intended to facilitate in vitro laboratory testing of finger stick blood samples for a variety of clinical chemistry assays.

This collection kit includes components needed to self-collect, package and mail a dried micro-blood sample to the certified clinical laboratory for testing. The collection kit is comprised of:

• Blood Sample Collection Cassette
• Sample Pouch
• Sterile Safety Lancets (2)
• Gauze Pad
• Bandage (2)
• Instructions for use
• Prepaid Return Mailer
• Patient Info Card
• Outer Packaging

Here's an analysis of the acceptance criteria and the studies performed for the Home Access® A1C Test, based on the provided document:

Acceptance Criteria and Reported Device Performance

The document does not explicitly present a table of "acceptance criteria" in a pass/fail format with numerical targets. Instead, it describes various performance studies and their findings, which implicitly establish that the device meets certain performance expectations. The closest representation of acceptance criteria for linearity and bias are inferred from the study findings and standard guidelines.

Inferred Acceptance Criteria and Reported Device Performance:

• Abnormal High (12.52%): SD 0.18, CV% 1.40%
• High (8.92%): SD 0.16, CV% 1.81%
• Diabetic in control (6.83%): SD 0.11, CV% 1.64%
• Normal (5.85%): SD 0.06, CV% 1.09%
  Total Imprecision:
• Abnormal High: SD 0.19, CV% 1.50%
• High: SD 0.27, CV% 3.00%
• Diabetic in control: SD 0.27, CV% 3.93%
• Normal: SD 0.16, CV% 2.69% |
  | Linearity | Linear over the claimed assay range. | Linear over the assay range of 4.5% - 14.5% A1c. Linear fit: y = -0.0553x + 1.02, r=0.995. |
  | Interference | Bias from tested interfering substances 97.1%). |

Study Details:

2. Sample sizes used for the test set and the data provenance:

• Precision Study: N=80 per HbA1c level (total N not explicitly stated but based on 2 replicates/level, 2 runs/day for 20+ days per level).
• Linearity Study: 76 individually prepared dry blood samples (19 levels of HbA1c x 4 samples per level). 52 of these were capillary specimens collected from 13 volunteers. The remaining were six commercially available venous whole blood specimens, each spotted onto four Cassettes.
• Interference Study: The number of samples for each interferent is not specified, only the concentration levels tested.
• Product Stability (Cassette/Pouch): 53 levels of HbA1c.
• Product Stability (Mailed Sample): 128 self-collected capillary blood samples.
• Flex Studies:
  • Blood Volume: 420 cassettes (3 HbA1c levels x 14 blood volumes/level x 10 cassettes/volume).
  • Interruption: 90 cassettes (3 HbA1c levels across control and two time interval groups).
  • Hematocrit: Not explicitly stated but 39 HbA1c levels with six specimens per level (2 control, 2 high Hct, 2 low Hct).
  • Coating Solution: 53 levels of HbA1c.
• Method Comparison: 256 patient samples (128 self-collected, 128 professionally-collected).
• Data Provenance: The document does not explicitly state the country of origin for the patient data. It refers to "volunteers" and "patient samples" but no geographic location or demographic details. The studies are clinical laboratory performance studies, thus are likely for a U.S. regulatory submission. It is retrospective for the clinically collected samples (venous blood) used for comparison, and prospective for the collection of capillary blood (self-collected and professionally-collected) as part of the study design.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience):

The concept of "experts" for establishing ground truth in this context is primarily related to the analytical measurement from established, FDA-cleared laboratory systems, rather than human interpretation.

• The ground truth for HbA1c values was established using FDA-cleared laboratory reagent and analysis systems, specifically the Beckman Coulter AU640e (K961274) and Beckman Coulter A1c reagents (K031380), which served as the reference method. These systems are inherently considered "expert" in their field for HBA1c measurement due to their regulatory clearance and established accuracy.
• For the Human Factors Studies, the "experts" were the lay-users themselves, as the study assessed their ability to understand and use the product. The number and qualifications of these lay-users are not specified beyond being "lay-users."

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set:

No adjudication method is described. The ground truth is based on objective laboratory measurements from a reference system.

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:

This is not applicable to this device. The Home Access® A1C Test is an in vitro diagnostic device for quantitative measurement of HbA1c, not an AI-assisted diagnostic imaging or interpretation tool that involves human readers in that sense. There is no "AI" component, nor "human readers" interpreting images/data in a similar fashion to an MRMC study setup.

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

The system is a standalone algorithm/device in terms of its analytical performance. The "algorithm" here is the chemical assay and instrument processing. The testing described (precision, linearity, interference, stability, flex studies) directly assesses the performance of the device itself (collection cassette and subsequent lab analysis), which functions without human intervention during the measurement phase within the lab. However, the overall "test" involves a human (patient) for sample collection, and a laboratory for processing.

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

The ground truth for the analytical performance studies (precision, linearity, interference, method comparison) was established by measurements obtained from an FDA-cleared predicate laboratory system (Beckman Coulter AU640e and A1c reagents) on venous blood samples. This is a reference standard measurement from a legally marketed device.

8. The sample size for the training set:

There is no explicit "training set" or "validation set" described in the context of an algorithm or AI. This is a medical device for chemical analysis. The studies described are performance verification studies for the device.

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

As there is no distinct "training set" in the AI/machine learning sense, this question is not applicable. The ground truth for the performance evaluations was established by the reference laboratory method as explained in point 7.

Ask a specific question about this device