Quick facts
- A glucose meter used outside a laboratory meets the Medical Devices Regulations definition of a near patient in vitro diagnostic device, and a near patient IVDD is classified as Class III (Schedule 1, Part 2, Rule 6).
- The definition does not stop at the clinic: it names "testing at home" as well, so a self-testing meter is also a near patient IVDD. A glucose system used inside a laboratory is the version that falls to Class II, under the patient-management rules.
- The accuracy question is usually answered with ISO 15197:2013: at least 95% of results within ±15 mg/dL below 100 mg/dL, and within ±15% at or above 100 mg/dL, evaluated per reagent lot.
- ISO 15197 is under revision: the 2013 edition remains the published text, while the revision project is registered and the record moved to stage 90.92 ("to be revised") in December 2024.
- In Canada, point-of-care testing is pulled into laboratory accreditation; IQMH's programme covers medical laboratories, specimen procurement and point-of-care testing, and references ISO 15189 and ISO 22870:2016.
- In the United States, the same glucose system is a Class II device under 21 CFR 862.1345, and testing with a home-use cleared meter can be performed under a CLIA certificate of waiver (42 CFR 493.15(c)(7)).
A blood glucose monitor kit looks like the simplest item in the clinic: a meter, a box of strips, a lancing device and lancets. What an inspection sees is not the kit. It sees where the reading was taken, who took it, which lot the strip came from, and whether the result was ever compared with a laboratory value. That is the whole audit, and most of it is paperwork the clinic already generates — if it keeps it. The kit discussed here is the blood glucose monitor kit with 50 test strips, and the parts that matter to an inspection are the strip lot on the vial and the record beside the meter.
Does a home-use meter licence cover testing inside a clinic?
The Medical Devices Regulations define the category first, and the definition is about location rather than technology: a near patient in vitro diagnostic device is "an in vitro diagnostic device that is intended for use outside a laboratory, for testing at home or at the point of care, such as a pharmacy, a health care professional's office or the bedside" (SOR/98-282, section 1). A clinic treatment room is a health care professional's office or a bedside in that language.
Then the classification rules do the rest. Under Rule 6, "a near patient IVDD is classified as Class III." Rule 9 keeps a short table of carve-outs — near-patient pregnancy and fertility devices and near-patient cholesterol devices are Class II, microbiological media and cultured-organism identification are Class I — and glucose is not in that table. The laboratory-use version of a glucose system takes the Rule 3 route instead, where an IVDD intended for patient management is Class II unless a life-threatening condition or an imminent risk applies.
For a buyer the consequence is narrower than it first looks, and it is worth stating plainly because it is widely got wrong: the class does not flip because a meter is small, cheap or sold to patients. What flips it is whether the device is used inside a laboratory. A clinic that buys the home-use version and runs it on patients in the office has not escaped Rule 6 — it has bought a device whose declared intended use still has to match the way the clinic uses it, and whose operator and quality-control obligations now sit with the clinic rather than with the patient.
Verifying what is actually licensed is a two-minute step: the licence number and the manufacturer's intended use are in Health Canada's Medical Devices Active Licence Listing, and the intended use wording is what an inspector compares against how the clinic uses the meter.
How does the glucose meter licence differ between Canada and the United States?
The two systems ask different questions. Canada asks what class the device is in and whether a licence exists for that class and intended use. The United States asks whether the test is waived or non-waived, and then applies the CLIA requirements attached to that answer.
| Item | Canada | United States |
|---|---|---|
| Governing instrument | Medical Devices Regulations, SOR/98-282 | 21 CFR Part 862 (device) and 42 CFR Part 493 (testing) |
| Device class for a glucose test system | Class II under Rule 3 for patient management; Class III under Rule 6 when it is a near patient device | Class II, special controls, 21 CFR 862.1345 |
| Pre-market route | Medical device licence for the applicable class | 510(k) premarket notification with the special controls |
| Who may perform the test | The licensed setting, under its laboratory or point-of-care quality programme | A CLIA-certified laboratory, or a site holding a certificate of waiver for waived tests |
| Home-use meter used in a clinic | Still a near patient device in either setting; Rule 6 applies | Waived testing is available for blood glucose by devices "cleared by the FDA specifically for home use" (42 CFR 493.15(c)(7)) |
| Where the quality rules come from | Accreditation against ISO 15189 and ISO 22870 through IQMH, plus CLSI POCT12 practice | CLIA subpart K, including the control-procedure standard at 42 CFR 493.1256 |
The practical consequence is asymmetric. An American clinic can build a defensible glucose programme around a waived home-use meter and a waiver certificate, because the regulation explicitly names that configuration. A Canadian clinic cannot borrow that logic: the near-patient classification does not depend on the meter being "simple", and the waiver concept does not exist in the Canadian instrument.
What accuracy does an auditor expect the meter to hold?
Accuracy is not a marketing claim in this context; it is an acceptance criterion with a number attached. ISO 15197:2013 sets it: for each reagent system lot, at least 95% of results must fall within ±15 mg/dL of the comparison method at glucose concentrations below 100 mg/dL, and within ±15% at concentrations at or above 100 mg/dL.
Two details in that sentence matter more than the percentages. The first is "per lot". A meter that passed a system evaluation with the lot the manufacturer submitted is not automatically cleared for the next strip lot that arrives in the clinic, which is why lot-to-lot verification appears in point-of-care quality programmes and in CLSI POCT12-A3, the approved guideline for point-of-care blood glucose testing in acute and chronic care facilities.
The second is that the standard is in motion. ISO 15197:2013 is still the published edition, but the revision project is registered and the ISO record carries stage 90.92, "International Standard to be revised", recorded 19 December 2024. A clinic writing a validation file this year is writing it against the 2013 text; a clinic writing one next year should check the record again rather than assume the document has not moved.
Canadian clinical guidance adds a second, simpler check that an inspector can ask about directly. Diabetes Canada's chapter on monitoring glycemic control states that at blood glucose levels above 4.2 mmol/L, a difference of less than 15% between a self-monitoring result and a simultaneous venous fasting plasma glucose is considered acceptable, and recommends comparing meter results with a laboratory measurement of fasting plasma glucose at least annually, and whenever A1C does not match the meter readings.
How is the accuracy claim checked in Canada compared with the United States?
In Canada the checking machinery is accreditation. IQMH's programme covers medical laboratories, specimen procurement and point-of-care testing, and its requirements reference ISO 15189 for laboratory quality and competence and ISO 22870:2016 for point-of-care testing specifically. ISO 22870 exists because point-of-care testing has failure modes that a central laboratory does not: the operator is not a laboratory technologist, the environment is not controlled, and the device is maintained by whoever is on shift.
In the United States the machinery is statutory. A laboratory holding a certificate of waiver is limited to the tests listed in 42 CFR 493.15(c), which include blood glucose by glucose monitoring devices cleared by the FDA specifically for home use. Anything outside that list pulls the site into the full CLIA requirements, and 42 CFR 493.1256 requires control procedures that "monitor the accuracy and precision of the complete analytic process", with a number, type and frequency of control testing that the laboratory establishes.
Neither system inspects the meter in isolation. Both inspect the programme around it — which is why a clinic that buys a good meter and never writes down a control result fails the same question as a clinic that bought a poor one.
Which records does the inspection ask for?
Four artefacts carry almost the entire conversation, and clinics typically have two of them.
| Record | Who owns it | How often it is produced | What it proves |
|---|---|---|---|
| Operator authorisation | Clinic lead or laboratory director | At hire and at re-authorisation | The person who tested was trained on this meter and this procedure |
| Control results | The operator, reviewed by the lead | Each testing day, or per the programme's stated frequency | The meter and the strip lot were performing when patient results were released |
| Lot verification | The lead | Each new strip lot before patient use | The new lot performs like the verified lot |
| Comparison with a laboratory result | The lead | At least annually, and when A1C and meter readings disagree | The meter still tracks the reference method in this clinic's hands |
Add the licence record — the device licence for the meter as used, and the accreditation certificate for the setting — because those two documents answer the first question an inspector asks and cannot be reconstructed after the fact.
What goes wrong most often in a point-of-care glucose audit?
Four failures account for most of the findings, and none of them is about the device.
The first is a meter in daily use that no one can tie to an authorisation record, usually because the device was bought for a physician's office and absorbed into clinic use without a training step. The second is control material that is expired, or a log that shows control results for days when the clinic was closed. The third is a strip lot that entered service with no verification, which is invisible until someone asks to see the transition. The fourth is the missing annual comparison: the clinic has a year of meter readings and no single laboratory glucose value to compare them against.
Each of these is cheap to fix and expensive to discover during an inspection, because the fix requires reconstructing history that was never recorded.
What should a clinic do when a strip lot or meter changes?
Treat the change as a small validation rather than an inventory event. Verify the new lot against the current one or against a laboratory result before patient testing, record the date and the person who did it, and keep the certificate of analysis or lot documentation that arrived with the shipment. When the meter itself changes, add operator training on the new device to the file before the first patient result, because the authorisation is specific to the device.
The same discipline applies to purchase. A kit that arrives with strips, lancing device and lancets keeps the clinic running for the first weeks, but the audit trail depends on what happens at the second order: whether the new strip lot is verified, whether the vial date is written on the box, and whether the lancets and alcohol prep pads used for site preparation are stored and rotated on the same schedule.
For clinics that run glucose testing alongside other near-patient work, the same evidence pattern repeats across devices, which is why the record set is worth standardising once: one operator list, one control log format, one lot-verification page. The diabetes care range and the laboratory consumables range cover the consumable side of that programme, and the sterile blood lancets, 33G, 100 per box are the single-use companion to the meter rather than a reusable alternative to it.
Sources
- Medical Devices Regulations, SOR/98-282 — consolidated text (definition of near patient IVDD, Schedule 1 Parts 1 and 2) — read 2026-11-08; page date modified 2026-11-02
- Health Canada — Medical Devices Active Licence Listing (MDALL)
- 21 CFR 862.1345 — Glucose test system
- 42 CFR 493.15 — Laboratories performing waived tests
- 42 CFR 493.1256 — Standard: control procedures
- ISO 15197:2013 — In vitro diagnostic test systems: blood-glucose monitoring systems for self-testing
- ISO/AWI 15197 — revision project record
- CLSI POCT12-A3 — Point-of-Care Blood Glucose Testing in Acute and Chronic Care Facilities
- ISO 22870:2016 — Point-of-care testing: requirements for quality and competence
- ISO 15189:2022 — Medical laboratories: requirements for quality and competence
- IQMH — accreditation requirements and guidance information for medical laboratories
- IQMH — accreditation programme information
- Diabetes Canada — Clinical Practice Guidelines, Monitoring Glycemic Control
- Laboratory and Specimen Collection Centre Licensing Act, R.S.O. 1990, c. L.1
- FDA — Device classification database
- CDC — Disinfection and sterilization: sterilizing practices
- Public Health Ontario — laboratory test information index
The free compliance log pack includes dated record blocks for operator authorisation, control results and lot verification, and the log generator on the same page prints a version for the current month. A clinic that wants its point-of-care record set reviewed before an inspection can ask for an institutional account review rather than rebuild the file from memory. Reselling? become a distributor.
Related reading
- Blood glucose test systems in Canada vs the US: which class applies?
- Blood glucose monitor procurement for Canadian clinics: testing kits and compliance
- Blood glucose monitoring systems for Canadian hospitals and home care
- How much do specimen containers and labels cost a Canadian clinic per test?
CliniEco Medical supplies glucose monitoring consumables and laboratory consumables in Canada under MDEL #35334. This page describes what inspections ask for; the requirements that bind a clinic are those of its regulator, accreditation body and manufacturer instructions for use. Institutional account enquiries go through the account enquiry page.
Frequently Asked Questions
Is a clinic glucose meter a Class III device in Canada?
A near patient in vitro diagnostic device is Class III under Rule 6 of Schedule 1. The definition reaches any device intended for use outside a laboratory, including testing at home, so a meter used in a treatment room or at the bedside meets it. Glucose is not among the Rule 9 carve-outs. A glucose system used inside a laboratory takes the Rule 3 patient-management route and lands in Class II.
Does a meter licensed for home use cover clinic testing?
The class is the same, because the near-patient definition covers both settings, but the licence is tied to the intended use the manufacturer declared. A clinic should check that the declared use covers professional or point-of-care operation, and should expect to supply the operator authorisation, control results and lot verification that the patient at home does not.
How often must a clinic compare its meter with the laboratory?
Diabetes Canada recommends comparing meter results with a laboratory measurement of fasting plasma glucose at least annually, and whenever A1C does not match the meter readings. A point-of-care programme may set a tighter interval, and any comparison that fails should be treated as a finding rather than a re-test.
Is a control result required on every testing day?
The programme decides the frequency and must be able to defend it. CLSI POCT12-A3 is written around a quality-control plan, operator authorisation and administrative responsibility rather than a single universal interval, and 42 CFR 493.1256 requires a laboratory to establish the number, type and frequency of control testing it performs.
What does ISO 15197 measure, and does it cover accuracy in clinic hands?
ISO 15197:2013 sets system accuracy requirements for blood glucose monitoring systems, including the 95% acceptance criteria evaluated per reagent lot. It is a manufacturer's design and validation standard; a clinic still has to verify each lot it puts into service and compare results with a laboratory method.
What should be kept when a new strip lot arrives?
The lot number, the expiry date, the date the lot entered service, the verification result and the name of the person who released it. Keeping the shipment documentation with the verification record turns a routine question at inspection into a page that can be produced in a minute.