(65 days)
The Healgen URS Test Strips are for the qualitative detection of Leukocytes (white blood cells) and nitrite in urine as an aid in the screening of a Urinary Tract infection (UTI). It is intended for over-the-counter home use only.
Healgen URS Test Strips are in vitro diagnostic test devices for qualitative detection of leukocyte and nitrite in urine. The device is composed of two color pads aligned on a strip. One pad is employed for testing leukocyte and the other for nitrite by visually reading the color change of the pad and comparing with the corresponding blocks on a color chart.
The document describes the performance of the Healgen URS Test Strips for the qualitative detection of Leukocytes and Nitrite in urine to aid in UTI screening for over-the-counter home use.
Here's a breakdown of the requested information:
1. A table of acceptance criteria and the reported device performance
The document does not explicitly state "acceptance criteria" for each performance characteristic in a formal table. However, it presents various study results that implicitly serve as an indication of meeting performance expectations for a Class I device intended for OTC use. The implied acceptance is based on the presented results demonstrating sufficient sensitivity, precision, and user-friendability.
Here's a table summarizing the reported device performance for key characteristics:
| Performance Characteristic | Analyte | Reported Device Performance |
|---|---|---|
| Limit of Detection (LOD) | Leukocyte | LOD verified at 15 cells/uL (100% detection at 30 and 15 cells/uL, 47% at 10 cells/uL, 7% at 5 cells/uL). |
| Nitrite | LOD verified at 0.05 mg/dL (100% detection at 0.1 and 0.08 mg/dL, 83% at 0.06 mg/dL, 50% at 0.05 mg/dL). | |
| Precision/Reproducibility | Leukocyte | 100% agreement with expected results across Negative, Trace (15 cells/uL), Small (70 cells/uL), Moderate (125 cells/uL), and Large (500 cells/uL) controls (N=45 for each control level). |
| Nitrite | 100% agreement with positive results for 0.1 mg/dL and 0.3 mg/dL controls (N=45 for each control level). | |
| Analytical Specificity | Leukocyte | No interference from various substances (Albumin, Ammonium Chloride, Bilirubin, Ciprofloxacin, Creatinine, Fructose, Galactose, Glycine, Hemoglobin, Lactose, Oxalic Acid, Phenazopyridine, Phenolphthalein, Potassium Chloride, Riboflavin, Sodium Nitrate, Sodium Nitrite, Sodium Phosphate, Sulfamethoxazole, Theophylline, Urea) at specified concentrations. |
| Potential interference: High glucose (≥1000 mg/dL), high ascorbic acid (≥150 mg/dL) may decrease readings. pH >8.0 may cause false positive readings. Specific gravity >1.035 may cause false negative readings. | ||
| Nitrite | No interference from various substances (same list as Leukocyte). | |
| Potential interference: High ascorbic acid (≥150 mg/dL) may cause false negative readings. No effect of pH or specific gravity on nitrite testing. | ||
| Stability | Both | Stable for 24 months at 2-30℃ based on real-time stability studies. |
| Lay-user Study (Method Comparison) | Leukocyte | +% Agreement (Exact Match): Large (90.9%), Moderate (92%), Small (89.7%), Trace (88.5%), Negative (98.3%). |
| +% Agreement (+/- Color Block): 100% for all levels. | ||
| Nitrite | +% Agreement (Exact Match): 100% for both Positive and Negative, with 100% overall agreement. | |
| Lay-user Instruction Survey | Instructions | All lay users indicated that the device instructions can be easily followed. Flesch-Kincaid reading analysis showed a reading grade level less than 7. |
2. Sample size used for the test set and the data provenance
- Test Set Sample Sizes:
- Sensitivity Study (LOD): 30 samples/tests per concentration for each analyte (Leukocyte and Nitrite). Each sample was tested in duplicates.
- Precision/Reproducibility Study: 45 assays results for each of eight levels of control (total 3 lots x 3 sites x 2 operators x 5 days x 1 rep/day = 90 tests per control, though results are reported as N=45 for each control).
- Analytical Specificity: Not explicitly stated as a single test set size, but multiple urine samples (negative urine spiked with interfering substances) were tested with three device lots by three different operators for each substance.
- Lay-user Studies: 150 lay users with UTI symptoms.
- Data Provenance: The document does not explicitly state the country of origin for the data. However, for the precision study, it mentions "three clinical sites," implying clinical data from a real-world setting. The lay-user study involved recruitment of "150 lay users with UTI symptoms," suggesting prospective collection of real-world user data. The sensitivity and analytical specificity studies appear to be laboratory-based. It is not specified if any data was retrospective.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
- LOD and Precision Studies: "Known concentrations" for spiked samples and "expected ranges/values" for controls are used as ground truth. The expertise involved in preparing these precisely calibrated samples and controls is implied but not explicitly detailed (e.g., analytical chemists, laboratory technicians). Operators are mentioned (5 for LOD, 2 at each of 3 sites for precision), but their role is to perform the tests, not establish ground truth.
- Analytical Specificity: Ground truth is established by "negative urine with different leukocyte and nitrite concentrations" which were likely precisely prepared laboratory samples.
- Lay-user Studies: The "comparison testing by healthcare professionals" served as the ground truth. The number and qualifications of these healthcare professionals are not specified beyond the general term "healthcare professionals."
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set
- The document does not describe a formal adjudication method (like 2+1 or 3+1) for resolving discrepancies in the test results.
- For the lay-user study, the comparison is made between the layperson's results and the results obtained by healthcare professionals, implying the healthcare professional's result is the definitive ground truth for comparison.
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 device is a test strip for visual interpretation by lay-users, not an AI-assisted diagnostic device evaluated by professional readers. The study focuses on direct comparison of lay-user performance to healthcare professional results for the test strip itself, not on the impact of AI assistance on human readers.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done
- No, this device is not an algorithm. It is a visual assay (colorimetric test strip) that requires human interpretation. Therefore, a standalone algorithm-only performance study is not applicable. The "standalone" performance here refers to the device's inherent chemical reaction and color development, which is then interpreted by a human.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
- LOD, Precision, Analytical Specificity: Ground truth was established using known concentrations of spiked analytes in urine samples and control materials, presumably confirmed by established laboratory methods and validated reagents.
- Lay-user Study: Ground truth was established by results obtained by healthcare professionals using the same test strips. This implies the healthcare professionals' interpretation served as the 'expert' reference. It's not explicitly stated whether these healthcare professionals confirmed results with a gold standard like culture or microscopic analysis, or if their reading of the strip was considered the definitive truth for the purpose of the study.
8. The sample size for the training set
- The document describes performance studies for the Healgen URS Test Strips, which are a chemical assay (test strips) with visual interpretation. This is not a machine learning or AI-based device that typically requires a "training set." Therefore, the concept of a training set size is not applicable here.
9. How the ground truth for the training set was established
- As stated above, this is not an AI/ML device, so there is no training set or ground truth establishment for a training set in the context of machine learning. The studies described are for analytical and lay-user performance of a diagnostic test strip.
§ 864.7675 Leukocyte peroxidase test.
(a)
Identification. A leukocyte peroxidase test is a device used to distinguish certain myeloid cells derived from the bone marrow, i.e., neutrophils, eosinophils, and monocytes, from lymphoid cells of the lymphatic system and erythroid cells of the red blood cell series on the basis of their peroxidase activity as evidenced by staining. The results of this test are used in the differential diagnosis of the leukemias.(b)
Classification. Class I (general controls). This device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 864.9.