(35 days)
ON•SITE Alcohol is recommended for professional use only and is not intended for overthe-counter sale to the general public. Professional use includes applications in healthcare, corrections and drug treatment.
ON•SITE Alcohol yields only a "qualitative" result. A positive result indicates that alcohol (ethanol) is present at a concentration of approximately 0.02% when performed according to the instructions.
Confirmation of ON•SITE Alcohol results using gas chromatography is recommended if quantitative results are desired.
ON•SITE Alcohol is a self contained, disposable enzymatic test device that provides rapid detection of alcohol in saliva or urine. It includes all reagents and supplies necessary to perform the test.
Here's an analysis of the provided text, focusing on the acceptance criteria and the study details:
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria are not explicitly stated in a quantitative form, but rather implied by the design intention and the performance characteristics observed in the 'modified' device compared to the 'current' device. The key goal was to increase the reaction time and slightly raise the limit of detection.
| Criteria (Implied) | Current Device Performance | Modified Device Performance | Target/Acceptance (Implied) |
|---|---|---|---|
| Accuracy & Precision (Detection at 0.008% w/v EtOH) | 13/20 positives in 2 mins, Mean Reaction Time: 111 sec | 0/20 positives in 2 mins, Mean Reaction Time: 147 sec | Reduced detection at 0.008% (i.e., fewer positives at 2 mins, increased reaction time) |
| Accuracy & Precision (Detection at 0.02% w/v EtOH) | 20/20 positives in 2 mins, Mean Reaction Time: 81 sec | 12/20 positives in 2 mins, Mean Reaction Time: 115 sec | Maintain detection at 0.02% (or slightly reduced as a side effect of shifted LOD), increased reaction time |
| Accuracy & Precision (Detection at 0.032% w/v EtOH) | 20/20 positives in 2 mins, Mean Reaction Time: 61 sec | 20/20 positives in 2 mins, Mean Reaction Time: 100 sec | Maintain detection at 0.032%, increased reaction time |
| Sensitivity (Reaction Times across various EtOH concentrations) | Range of mean reaction times from 1 min 6 sec to 1 min 47 sec | Range of mean reaction times from 1 min 29 sec to 2 min 41 sec | Consistently increased reaction times across all tested concentrations |
2. Sample Size Used for the Test Set and Data Provenance
- Test Set Sample Size:
- Accuracy and Precision: 20 replicates for each of the three ethanol controls (0.008%, 0.02%, 0.032% w/v). This means 20x3 = 60 tests for the current device and 60 tests for the modified device, totaling 120 tests for this part.
- Sensitivity: 10 replicates for each of the eight ethanol controls (0.005% to 0.0225% w/v). This means 10x8 = 80 tests for the current device and 80 tests for the modified device, totaling 160 tests for this part.
- Data Provenance: Not specified within the document (e.g., country of origin, type of samples used - saliva/urine). Given the context of a 510(k) summary, it's implied to be laboratory testing of controlled ethanol solutions, not human samples. It is a prospective laboratory study comparing a modified device to its predicate.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications
Not applicable. The ground truth was established by laboratory-prepared ethanol solutions of known concentrations. The "results were read" likely implies visual interpretation by a technician, but their number or qualifications are not specified as it's not an expert-driven ground truth.
4. Adjudication Method for the Test Set
Not applicable. The outcomes were objective measurements (time to positive result) and counts of positive/negative results at a specific time point, based on predefined ethanol concentrations. No adjudication process for differing interpretations of subjective data would be needed.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done
No, an MRMC study was not done. This study is a technical performance comparison between two versions of a diagnostic device in a laboratory setting, not a study involving human readers interpreting cases.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done
Yes, this was a standalone performance study. The device's reaction to known concentrations of alcohol was evaluated directly, without human interpretation of results beyond simple visual observation of a color change, and timing that change. There is no "algorithm" in the modern sense relevant to AI.
7. The Type of Ground Truth Used
The ground truth used was objectively known concentrations of ethanol in laboratory-prepared control solutions.
8. The Sample Size for the Training Set
Not applicable. This device is an enzymatic test, not an AI/machine learning model, so there is no training set in the computational sense. The "training" for the device's development would have been implicit in chemical formulation and calibration, not a data-driven training set.
9. How the Ground Truth for the Training Set Was Established
Not applicable, as there is no training set in the context of an AI/ML model. If interpreted very broadly as the development process for the chemical formulation, the ground truth would still be based on known chemical reactions and desired performance characteristics to achieve the intended "shift in reaction time and detection levels."
§ 862.3040 Alcohol test system.
(a)
Identification. An alcohol test system is a device intented to measure alcohol (e.g., ethanol, methanol, isopropanol, etc.) in human body fluids (e.g., serum, whole blood, and urine). Measurements obtained by this device are used in the diagnosis and treatment of alcohol intoxication and poisoning.(b)
Classification. Class II.