1. What POCT is (the short version)

POCT (point-of-care testing) is testing done outside the laboratory, close to the patient (our own working definition). According to the standards body's official page, ISO 22870:2016 set requirements for POCT carried out in a hospital, clinic or healthcare organization providing ambulatory care. It could be applied to transcutaneous measurements, analysis of expired air and in vivo monitoring of physiological parameters, and excluded patient self-testing in a home or community setting, though elements of it could apply Sourced.

• "Where" is part of the definition: the same test principle needs a different design and a different regulatory treatment depending on whether a laboratory specialist, a clinic nurse or the person at home runs it (chapter 8)
• Representative formats: the review by Koczula and Gallotta describes lateral flow assays (LFAs) as the technology behind low-cost, simple, rapid and portable detection devices, and says the most commonly used LFA is the pregnancy test Sourced. A blood glucose meter is a combination of a disposable electrochemical strip and a small meter (chapter 5)
• The ideal: for diagnostics in low- and middle-income settings, ASSURED (affordable, sensitive, specific, user-friendly, rapid, equipment-free, delivered) was coined more than ten years ago, and Land and colleagues extended it to REASSURED by adding digital elements and more Sourced

2. Two representative formats: lateral flow and electrochemical strips

3. Structure and materials of lateral flow

Koczula and Gallotta's review (2016, open access) describes the LFA as a paper-based platform that gives results within 5-30 minutes of placing the sample Sourced. The liquid sample moves by capillary action, without external force, through zones of polymeric strips on which molecules that can interact with the analyte are attached. A typical strip consists of overlapping membranes mounted on a backing card for stability, and comprises four elements.

View the original SVG schematic, source notes and numerical details

All Sourced (Koczula and Gallotta 2016 [Reference 15]). The pore size range (0.05-12 µm) is the review's statement and not a product specification.

Detection comes in two formats. In the direct (sandwich) format the analyte is captured between two antibodies; it is used for large molecules and for analytes with multiple antigenic sites, such as hCG in pregnancy tests. A positive result shows a test line, and the control line indicates proper flow. The competitive format is used for small molecules that cannot bind two antibodies at once; the analyte blocks the binding sites of the antibodies on the test line, so a positive result is shown by the absence of a line Sourced.

4. A materials engineer's view 1: flow through paper is a problem of material combinations

5. Electrochemical strips: enzyme, mediator and electrode

The disposable strip of a blood glucose meter is an electrochemical biosensor. FDA's 510(k) decision summaries state the principle as follows Sourced.

• A home-use blood glucose system (K173658): it uses a glucose-oxidizing enzyme, flavin adenine dinucleotide-glucose dehydrogenase (FAD-GDH), together with a redox mediator on a disposable test strip (an electrochemical embedded sensor), and a hand-held device measures the current generated. The current is proportional to glucose concentration, and the meter's software converts it to a concentration (mg/dL). 21 CFR 862.1345 (glucose test system), Class II, product code NBW (over-the-counter blood glucose system)
• Another home-use system (K181527): a reagent system containing FAD-GDH and a mediator reacts with glucose in whole blood from the fingertip, forearm or palm to produce a current, measured by amperometry. The meter shows a value calibrated to plasma reference

The "conditions for error" in the decision summaries

Even within one format, the conditions under which error arises differ by product. Here are the limitations on use from two FDA decision summaries Sourced.

All Sourced (FDA 510(k) decision summaries [References 12, 13]). The limits in the two products are values set under each product's own test conditions and do not rank the technology.

What FDA's guidance for home-use glucose systems (29 September 2020) asks to be evaluated

• Interferents: it gives a table of known or potential interferents that form the minimal list to test, including acetaminophen (20 mg/dL), ascorbic acid (6 mg/dL), bilirubin, cholesterol, galactose (60 mg/dL), maltose (480 mg/dL), xylose, mannitol, salicylic acid and ibuprofen, with testing at glucose values of about 50-70, 110-130 and 225-270 mg/dL
• Hematocrit: because the majority of intended users may be expected to have hematocrit between 20% and 60%, it recommends claiming that range. Because lay users have no way to determine their hematocrit, a system should measure adequately across 30-55%; if it cannot, it may not be found substantially equivalent. Testing is in 5% steps, over five glucose levels and at least three strip lots
• Strip stability: two studies, closed-vial (shelf life) and open-vial stability, across combinations of temperature and humidity

These are statements in guidance (nonbinding recommendations) and do not show the performance of any individual product Sourced.

6. A materials engineer's view 2: enzyme selectivity sets the direction of error

7. User error: designed out, estimated by study

The performance of POCT is decided not by laboratory specialists but by people who read the instructions and use the test for the first time. FDA treats this "user error" from both the design side and the study side.

Estimated by study: the user evaluation for home-use glucose systems

FDA's guidance (29 September 2020) asks for a user evaluation as part of the 510(k) and recommends this design Sourced.

• Subjects: at least 350. At least 10% of participants should be naive to glucose meters (non-diabetic subjects may be included)
• With the instructions only: subjects are given the draft labeling and take their own fingertip sample and test with no other training or prompting, and without help from a technician or healthcare provider. They are sequestered so they cannot observe others' technique
• Language and readability: the instructions are in English only, and for home use the reading level should be 8th grade or lower, assessed for example by Flesch-Kincaid
• Strips: at least 10 vials or packages covering at least 3 lots, having been through typical shipping and handling
• The target: 95% of all results within ±15% of the comparator and 99% within ±20%. FDA notes that most 510(k) studies are done in idealized conditions that may overestimate total accuracy in the hands of users, and asks for a design that reflects expected conditions of use
• Relation to ISO 15197: FDA says many manufacturers design their studies on ISO 15197 but that it believes the standard's criteria are not sufficient to protect lay users, because for example it does not adequately address the hypoglycemic range or performance across strip lots

Designed out: the CLIA waiver guidance (26 February 2020)

FDA's CLIA waiver guidance asks that a test eligible for waiver be shown to be designed so that results do not stray far despite user error Sourced.

• Risk analysis and flex studies: identify all known sources of error, such as operator error, specimen handling and reagent condition, and test the system under stress conditions beyond its normal operating limits to verify that it is insensitive to environmental and usage variation
• Examples of error sources (from the guidance's list): incorrect specimen type, incorrect application or volume, incorrect handling of reagents, placing the device on a non-level surface, wrong placement of strips or other reagent-bearing components, wrong order, amount or timing of reagents, misreading results, misreading due to color blindness, errors in specimen collection, clotted specimens, bubbles, interfering substances, and improperly stored or expired reagents
• Fail-safe and failure alert mechanisms: most risk controls should be mechanisms that give no result when something is wrong (a lock-out) or that alert the operator. Examples include guides or channels that prevent improper strip placement and messages such as "out of range high" or "out of range low"

8. The CLIA waiver: a category by where a test is used (FDA)

The US CLIA (Clinical Laboratory Improvement Amendments of 1988) sorts tests by complexity and sets laboratory requirements accordingly. FDA places in vitro diagnostic tests in one of three categories: waived, moderate complexity and high complexity Sourced.

• The statutory definition of waived: the CLIA statute (42 U.S.C. § 263a(d)(3)), as quoted on FDA's page, describes waived tests as those approved by FDA for home use, or those that the Secretary determines are simple laboratory examinations and procedures that have an insignificant risk of an erroneous result, including those that "employ methodologies that are so simple and accurate as to render the likelihood of erroneous results by the user negligible," or that "pose no unreasonable risk of harm to the patient if performed incorrectly" Sourced
• When a test is automatically waived: tests waived by regulation under 42 CFR 493.15(c), or cleared or approved for home use or over-the-counter use, are categorized as waived after clearance or approval. Otherwise FDA reads the package insert's instructions and uses a seven-criterion "scorecard" (1 to 3 points each), categorizing 12 points or fewer as moderate and above 12 as high complexity. A test that has not been categorized is treated as high complexity until it is Sourced
• CLIA Waiver by Application (CW): a manufacturer of a test categorized as moderate may ask FDA through a CW for waived status. A waived test can be run by laboratories with a Certificate of Waiver (CoW) and others. FDA's review goals are 150 FDA days for a CW alone (320 if an advisory panel is needed) and 180 FDA days for a dual 510(k) and CW Sourced
• CW study design (guidance of 26 February 2020): untrained operators (people in waived settings with limited or no laboratory training or experience) at at least three sites, with 1-3 operators per site and at least nine in total. The operators should be people who do the testing as part of daily work, multitasking with other duties Sourced

9. Rapid COVID-19 tests: emergency use authorization and ordinary review

Rapid COVID-19 antigen tests are among the most widely used POCT. What matters here is by which route they reached the market. According to FDA's antigen test page (content current as of 1 September 2026), most COVID-19 tests were issued emergency use authorizations (EUAs). An EUA is a procedure under section 564 of the Federal Food, Drug, and Cosmetic Act and is separate from ordinary 510(k) clearance or PMA approval. The page also says that, besides EUA tests, there are antigen tests authorized through traditional premarket review pathways Sourced.

• 26 August 2020: FDA's listing records Abbott Diagnostics Scarborough's "BinaxNOW COVID-19 Ag Card" as an antigen test with an EUA of 26 August, as lateral flow, visual read, single target. The same company's home version (OTC, telehealth proctor supervised) is dated 16 December Sourced
• 15 December 2020: FDA issued an EUA for Ellume's rapid lateral flow antigen test as the first OTC, fully at-home COVID-19 diagnostic test (ages two and over, including those not showing symptoms). FDA's CDRH director said that "this test, like other antigen tests, is less sensitive and less specific than typical molecular tests run in a lab" Sourced
• 11 May and 6 June 2023: the public health emergency expired. On 6 June FDA granted marketing authorization to the Cue COVID-19 Molecular Test (a molecular nucleic acid amplification test) as the first at-home OTC COVID-19 test through a traditional premarket review pathway. Existing EUAs, however, remained in effect under section 564 Sourced
• 29 June to 26 December 2026: FDA's page says the HHS Secretary determined that circumstances no longer exist justifying emergency use of COVID-19 diagnostics, personal respiratory protective devices and medical devices, and that these declarations will be terminated effective 26 December 2026. On that date the EUAs listed on the page will no longer be in effect. FDA encourages manufacturers to review the transition plan Not yet confirmed

Performance: averages from a systematic review

The Cochrane review by Dinnes and colleagues (2022; literature to March 2021) pooled 152 evaluations of single test applications (100,462 samples) of rapid antigen tests Sourced.

• Sensitivity (average): 73.0% in symptomatic people (95% CI 69.3-76.4%) and 54.7% in asymptomatic people (47.7-61.6%). It was 80.9% in the first week after symptom onset and 53.8% in the second, and fell as viral load fell
• Specificity (average): 99.1% in symptomatic and 99.7% in asymptomatic people
• Differences between brands: used according to manufacturer instructions, average sensitivity by brand ranged from 34.3% to 91.3% in symptomatic participants (20 assays) and from 28.6% to 77.8% in asymptomatic participants (12 assays). Seven assays reached 80% or more, the WHO acceptable criterion
• The authors' conclusion: sensitivity varies between tests and a negative result may still be an infection, so the tests are more suitable as triage to RT-PCR. Many commercial tests have not been evaluated in independent studies

These are averages from literature up to March 2021. They do not show the performance against later variants, or the present performance of any individual product.

10. Our calculation: predictive value and permitted counts in a clinical study

11. Related standards: the withdrawal of ISO 22870 and ISO 15189:2022

The international standards on POCT have changed considerably in recent years. Here is what we could confirm on the official pages of the standards body EVS (the Estonian Centre for Standardisation and Accreditation) Sourced.

All Sourced (EVS standard pages [References 1-6, 22]. The ISO site itself could not be fetched automatically, so the pages of an ISO member body were used. The text of the standards is paid and was not checked).

The relationship between ISO 15197 and FDA also matters. FDA's home-use glucose guidance (2020) says many manufacturers design studies on ISO 15197 but that the standard does not adequately address the hypoglycemic range or performance across strip lots, and that FDA believes it is not sufficient to protect lay users. FDA's 2016 explanatory slides state "FDA does not recognize ISO 15197:2013" and describe the studies in its guidance as "compatible with ISO 15197:2013, but not identical" Sourced (2016 material; the present recognition status was not checked). For glucose testing by professionals (prescription POCT), a separate guidance sets goals of 95% within ±12% (above 75 mg/dL) or ±12 mg/dL (below 75 mg/dL), and 98% within ±15% or ±15 mg/dL Sourced. Even for the same glucose measurement, the standard changes with who uses it (the home or a professional).

12. A materials engineer's view 3: "simple" is a performance that design creates

13. Open issues and what could not be confirmed

(1) Japanese and EU systems were not checked

The regulatory statements here rest on US sources (FDA, CMS). We did not check Japan's in vitro diagnostic reagents under the Pharmaceuticals and Medical Devices Act, or the self-test and near-patient-test classes of the EU's IVDR (Regulation (EU) 2017/746). The latter is covered in this series' article on medical device categories and home use.

(2) The text of standards was not checked

The text of ISO 22870, ISO 15189, ISO 15197 and others is paid. We confirmed only titles and scope on the standards body's pages, and did not check the accuracy criteria of ISO 15197:2013 or the concrete POCT requirements of ISO 15189:2022.

(3) Actual error rates in the home could not be confirmed

FDA's guidance shows the design of user studies and a pass target, but we found in the sources consulted no primary source giving the frequency of mistakes in real home use.

(4) The performance of COVID-19 tests is given only as averages

The Cochrane review pools literature up to March 2021. We did not check performance against later variants, the current performance of individual products, or the handling after 26 December 2026. For an individual product, one must refer to the conditioned values in its EUA documents and labeling.

(5) Other POCT formats are not covered

Small analyzers (cartridge immunoassays, blood gas), nucleic acid amplification POCT and smartphone readers are not covered here.

(6) The EXTRANEAL label is the 2016 version

The label warning cited in chapter 6 is from the 2016 version published by FDA. We did not check the current version.

14. Glossary

15. References

1. EVS (Estonian Centre for Standardisation and Accreditation) ISO 22870:2016 Point-of-care testing (POCT) — Requirements for quality and competence (withdrawn 6 December 2022) — evs.ee
2. EVS ISO 15189:2022 Medical laboratories — Requirements for quality and competence — evs.ee
3. EVS ISO 15197:2013 In vitro diagnostic test systems — Requirements for blood-glucose monitoring systems for self-testing in managing diabetes mellitus — evs.ee
4. EVS ISO 18113-1:2022 In vitro diagnostic medical devices — Information supplied by the manufacturer (labelling) — Part 1 — evs.ee
5. EVS ISO 23640:2011 In vitro diagnostic medical devices — Evaluation of stability of in vitro diagnostic reagents — evs.ee
6. EVS ISO 20916:2019 In vitro diagnostic medical devices — Clinical performance studies using specimens from human subjects — Good study practice — evs.ee
7. U.S. Food and Drug Administration (FDA) "CLIA Waiver by Application" (CLIA categories, statutory text, review times, dual submission) — fda.gov
8. FDA "CLIA Categorizations" (categorization criteria, scorecard, uncategorized tests default to high complexity; content current as of 17 July 2023) — fda.gov
9. FDA Guidance "Recommendations for Clinical Laboratory Improvement Amendments of 1988 (CLIA) Waiver Applications for Manufacturers of In Vitro Diagnostic Devices," 26 February 2020 (nonbinding recommendations) — fda.gov
10. FDA Guidance "Self-Monitoring Blood Glucose Test Systems for Over-the-Counter Use," 29 September 2020 (nonbinding recommendations) — fda.gov
11. FDA (CDRH) Webinar slides on the final blood glucose guidances (OTC and prescription POC), 21 November 2016 — fda.gov
12. FDA 510(k) decision summary K173658, Confidence Blood Glucose Monitoring System (Applied Biomedical LLC) — accessdata.fda.gov
13. FDA 510(k) decision summary K181527, On Call Sure / On Call Sure Sync Blood Glucose Monitoring Systems — accessdata.fda.gov
14. FDA (Drugs@FDA) EXTRANEAL (icodextrin) peritoneal dialysis solution, prescribing information (2016 version, NDA 021321) — accessdata.fda.gov
15. Koczula KM, Gallotta A Lateral flow assays. Essays Biochem 60(1): 111-120 (2016). DOI 10.1042/EBC20150012 (open access, PMC4986465) — doi.org
16. Posthuma-Trumpie GA et al. Lateral flow (immuno)assay: its strengths, weaknesses, opportunities and threats. A literature survey. Anal Bioanal Chem 393(2): 569-582 (2009). DOI 10.1007/s00216-008-2287-2 (abstract only checked, PMID 18696055) — doi.org
17. Land KJ et al. REASSURED diagnostics to inform disease control strategies, strengthen health systems and improve patient outcomes. Nat Microbiol 4(1): 46-54 (2019). DOI 10.1038/s41564-018-0295-3 (abstract checked on PubMed, PMID 30546093) — doi.org
18. FDA "In Vitro Diagnostics EUAs - Antigen Diagnostic Tests for SARS-CoV-2" (content current as of 1 September 2026) — fda.gov
19. FDA Press release "Coronavirus (COVID-19) Update: FDA Authorizes Antigen Test as First Over-the-Counter Fully At-Home Diagnostic Test for COVID-19," 15 December 2020 — fda.gov
20. FDA Press release "FDA Permits Marketing of First COVID-19 At-Home Test Using Traditional Premarket Review Process," 6 June 2023 — fda.gov
21. Dinnes J et al. Rapid, point-of-care antigen tests for diagnosis of SARS-CoV-2 infection. Cochrane Database Syst Rev 7: CD013705 (2022). DOI 10.1002/14651858.CD013705.pub3 (abstract checked on PubMed, PMID 35866452) — doi.org
22. EVS ISO 14971:2019 Medical devices — Application of risk management to medical devices — evs.ee

16. Claim-to-source audit

Last updated 4 October 2026. Sources are limited to primary material (FDA guidance, decision summaries, press releases and EUA listings, official pages of standards bodies, peer-reviewed papers, a Cochrane review, and prescribing information published by FDA). Because the article includes structural readings and materials-design interpretations, those are marked as Commentary and kept separate from sourced fact. Japanese and EU systems, the text of ISO standards, actual home error rates, the current performance of COVID-19 tests, and other POCT formats are not stated, because no published primary source could be confirmed or they are outside this article. The end of the declarations on 26 December 2026 is a date scheduled on FDA's page. All figures are explanatory concept graphics. Figs. 1, 2 and 4 to 7 are vector drawings, Fig. 8 is a drawing that includes our calculation, and the hero image and Fig. 3 are AI-generated images; none of them shows a real product, strip or procedure. This article explains the materials, evaluation and regulation of tests and is not medical advice.