VetMedGuide: Building a Useful Veterinary Device Field-Feedback Form

A veterinary analyzer complaint is an intake job, not a finished investigation

When a veterinary clinic reports that an in-house hematology analyzer gave a “wrong” result, triggered an unexpected flag, or “missed” a diagnosis, the manufacturer’s first job is intake, not conclusion. Capture the observation first, and keep four things in separate fields: what the instrument or operator actually saw; the species and sample matrix; whether any clinic quality-control (QC), smear, calibration, or service evidence exists; and any clinical-effect language, which remains an unsupported causality claim until an investigation has the missing records.

This page is a manufacturer field-feedback worksheet for that split. It is not a clinic QC tutorial, not a Center for Veterinary Medicine (CVM) market-entry encyclopedia, and not a human-device complaint standard operating procedure. Adjacent MedDeviceGuide articles already cover those neighboring jobs: veterinary medical device regulation for classification and market-entry pathways; medical device complaint handling and quality investigation for human-device QMSR and ISO 13485 complaint files; home-use IVD invalid-result workflow and calibrator and control traceability for human IVD design and metrology; and CAPA, post-market surveillance, and point-of-care testing regulation for human QMS, PMS, and CLIA/POCT overviews. Those pages are related reading, not substitutes for this intake form.

VetMedGuide is a publication, not a laboratory, regulator, or proof of clearance. Its clinic-side article on veterinary hematology analyzer quality control workflows describes how a practice may organize flags, smears, and calibration during patient runs. That is clinic workflow. It is not this manufacturer form, and it is not an authority for duties, numeric limits, or investigation outcomes.

What FDA currently does and does not require for animal-use devices

A frequent error is to copy human medical-device premarket and reporting rules onto an exclusive animal-use analyzer. FDA’s literacy page “How FDA Regulates Animal Devices,” content current as of 6 April 2026, is not a regulation text and does not enumerate analyzer complaint fields. It does state the current CVM posture that this intake form has to respect.

FDA has regulatory oversight over devices intended for animal use and can take appropriate regulatory action if an animal device is misbranded or adulterated. The same page states that FDA does not require submission of a 510(k), PMA, or any premarket approval for devices intended for animal use. Device manufacturers who exclusively manufacture or distribute animal devices are not required to register their establishments or list those animal devices with FDA and are exempt from post-marketing reporting. It remains the manufacturer or distributor’s responsibility to assure that those animal devices are safe, effective, and properly labeled. That responsibility sentence is not an FDA animal-device approval, a 510(k) clearance, or a mandatory medical device reporting (MDR) clock.

The same literacy page encourages veterinarians and animal owners to report adverse drug experiences and product defects associated with animal devices to FDA using Form FDA 1932a. FDA’s companion how-to-report page, content current as of 4 February 2026, likewise encourages veterinarians and animal owners to report adverse events associated with drugs or devices used in animals, and lists medical devices used in animals among reportable products, with examples including thermometers, glucose meters, and bandage materials. Those pages are reporter instructions. They are not a manufacturer complaint SOP and they do not convert exclusive animal-device makers into 21 CFR Part 803 reporters.

FDA generally does not regulate the manner in which veterinarians use legally marketed devices in animals within the practice of veterinary medicine. The Animal Medicinal Drug Use Clarification Act of 1994 (AMDUCA) extra-label drug rules do not apply to devices. FDA also warns that, because it does not approve animal medical devices, device labeling may not be as well-developed as approved animal-drug labeling. Off-label use that causes another Federal Food, Drug, and Cosmetic Act (FD&C Act) violation, such as adulterated food from a food-producing animal, remains a separate food issue. A clinic’s extra-label use of a human-intended analyzer in animals is therefore not, by itself, an FDA animal-device approval, a completed malfunction finding, or permission to copy a human MDR SOP onto the animal report.

Human-device quality and reporting frameworks stay scoped to finished devices intended for human use. 21 CFR 820.1 states that the Quality Management System Regulation (QMSR) governs methods, facilities, and controls used for design, manufacture, packaging, labeling, storage, installation, and servicing of finished devices intended for human use, and that the provisions apply to any finished device intended for human use manufactured in, or imported into, the United States. 21 CFR 803.1, in Subchapter H, establishes medical device reporting requirements for device user facilities, manufacturers, importers, and distributors. Combined with the CVM animal-devices page, those sections are not a hidden animal-device MDR statute for exclusive animal-device makers.

Dual product lines still have to be labeled as such on the intake form. Manufacturers of a human-intended analyzer remain subject to QMSR and Part 803 for the human device. An animal-use clinic narrative is not automatically a human MDR and is not a completed investigation. Record whether the complaint names an exclusive animal-use catalog, a human-intended catalog used extra-label in animals, or is still unknown. Do not decide reportability from an incomplete file.

Form FDA 1932a already splits product problems from adverse events

Form FDA 1932a (9/23), OMB No. 0910-0284, expiration 31 August 2026, is titled Veterinary Adverse Drug Reaction, Lack of Effectiveness, or Product Defect Report and is marked for voluntary reporting. FDA CVM’s veterinary medication-errors page, content current as of 21 May 2024, likewise describes voluntary reports on Form FDA 1932a and states that the form can be used to report medication errors with or without adverse events for any animal drug or animal device. The form is a reporter instrument. A High/Medium/Low/Unknown suspicion checkbox is not a manufacturer causality determination and not a Part 803 decision.

The form’s Report Type field says to check all that apply. The three report-type boxes are Adverse Event, Product Problem, and Product Use Error. The instructions tell the sender to choose Adverse Event if reporting something that has affected an animal or a human, including lack of effectiveness; Product Problem for something associated with a product, such as crumbled tablets or peculiar appearance; and Product Use Error for something associated with a product that could have or has led to a medication error, such as look-alike or sound-alike names, similar product appearance, or error-prone packaging or labeling. Those boxes support recording an instrument or product observation without treating the same narrative as a concluded clinical-causality finding. They are not mutually exclusive, and they are not a finished investigation.

The sender is asked whether the event was previously reported to the manufacturer and, if so, for the manufacturer’s case number. The form also collects species, breed, lot number, expiration date, and the veterinarian’s level of suspicion that the product caused the adverse event (High, Medium, Low, or Unknown). Species and lot are identity fields. Suspicion on a voluntary form is still an observer judgment. Drug-oriented fields such as dose, route, dechallenge, and rechallenge will often be not applicable for an analyzer; do not force them into clinical meaning on a device file.

FDA’s Adverse Event Reports for Animal Drugs and Devices page states that CVM AER data include adverse events or product-defect reports related to animal drugs or devices used in animals, including product-quality issues, lack of effectiveness, and defective packaging. The agency states that the information is as reported to CVM and that FDA has not necessarily determined whether the product was the actual cause of the events reported. Presence of a report in that dataset is not proof of incidence or of causation. A clinic narrative that names a clinical effect is therefore an observation to record, not proof of a reportable malfunction or of completed causality.

A manufacturer intake form should keep the same split. Record the operator observation without a cause. Park clinical-effect language as unsupported pending records. Ask whether the sender already reported to the manufacturer, and keep that answer separate from whether anyone has concluded a defect. Do not collapse Form FDA 1932a’s report types, species field, lot field, and suspicion scale into one causality sentence.

Species, matrix, analyzer, lot, and software are identity fields

Veterinary hematology results are not interchangeable across species, sample matrices, analyzers, reagent or control lots, or software versions. That is why those items belong on the intake form as identity fields rather than as published numeric ranges. Missing identity fields block reconstruction of the event. They do not prove a defect.

The ASVCP 2012 hematology quality-assurance guideline (Vap et al., Veterinary Clinical Pathology 2012;41(1):8–17; PubMed 22390423) addresses control of preanalytical and analytical factors for hematology for mammalian and nonmammalian species, hemostasis, and crossmatching in veterinary laboratories. The PubMed abstract states that the guidelines are intended for veterinary diagnostic laboratories and veterinary research laboratories that are not covered by FDA Good Laboratory Practice standards (21 CFR Part 58). ASVCP’s Quality Assurance and Laboratory Standards page lists those hematology QA documents among current and archived QALS resources. The guideline is not a manufacturer complaint form. This article uses only that stated scope. It does not republish paywalled method tables, smear-review rules, allowable total error, or QC frequencies.

Cornell University’s Animal Health Diagnostic Center states that its reference intervals were established in that laboratory from healthy animals using its own equipment with specific methods and reagents, and that because results are analyzer-, method-, and reagent-dependent, those intervals are only valid for results from that laboratory. That disclaimer is the reason to request species, sample matrix, analyzer identity, reagent or control lot, and software. It is not permission to copy Cornell’s chemistry or hematology numeric tables into a manufacturer article, and it is not a universal veterinary reference range.

On the intake form, therefore, record species and sample matrix as supplied, missing, or unknown. Record analyzer identity the same way: commercial name, serial or other identification, or not supplied. Record reagent or control lot and software version, or record that they are unknown. Do not invent species reference intervals, performance-validation results, or allowable total error in order to “complete” the row. Human IVD kit traceability and invalid-result design are separate jobs, already covered in calibrators and controls traceability and home-use IVD invalid-result workflow; do not recast those encyclopedias here.

Clinic QC evidence is a requested record, not a CLIA duty

When a complaint asserts an incorrect patient result, the manufacturer may need to know whether any clinic QC, smear review, calibration, or service record exists. That request is a technical-record question. It is not a demand that a veterinary practice prove CLIA compliance.

42 CFR 493.1 sets forth the conditions that laboratories must meet to be certified to perform testing on human specimens under the Clinical Laboratory Improvement Amendments of 1988 (CLIA). 42 CFR 493.2 defines a laboratory, for that part, as a facility for examination of materials derived from the human body for diagnosis, prevention, or treatment of disease or impairment of, or assessment of the health of, human beings. Those sections do not describe how a veterinary clinic should run hematology QC, and they do not authorize copying human CLIA, CLIA-waived status, or 42 CFR 493 QC-frequency rules onto animal-specimen testing.

Clinic QC evidence on a veterinary analyzer complaint is therefore a technical record the manufacturer may request, not a CLIA legal duty copied onto animal testing. The intake form should treat the following as present, missing, or unknown, without inventing pass/fail limits or mandatory frequencies:

• Whether QC was run on the date of the disputed result, and the control lot if supplied.

• Whether a smear review was performed, if the clinic mentions one. Record the fact of review; do not import clinic smear-trigger tables.

• Whether a calibration or service event is claimed, including the date if supplied and whether the service record is attached.

• Whether those artifacts remain unknown. Leave the field unknown rather than inferring a QC miss or an in-control state.

Do not demand “CLIA-waived validation records” from a veterinary customer. Do not treat absence of a human laboratory QC log as proof of device failure. To see how clinics themselves discuss flags and smears, readers may use VetMedGuide’s veterinary hematology quality control workflow as related clinic reading with that limit: it is practice workflow, not this intake form, and it is not a source for duties or numeric limits. Human CLIA/POCT categorization remains a different article, point-of-care testing regulation.

ISO 13485 and QMSR can structure intake without creating an animal MDR clock

ISO 13485:2016, third edition, March 2016, remains the published current edition and was last reviewed and confirmed in 2025. 21 CFR 820.7(b) incorporates ISO 13485:2016(E), third edition, 1 March 2016, by reference for §§ 820.1, 820.3, 820.10, 820.35, and 820.45. Maintaining ISO 13485 does not convert exclusive animal-device makers into 21 CFR Part 803 reporters. Full clause 8.2.2 complaint-handling text is in the paid standard and is not pasted here.

ISO 13485:2016 clause 3.4, publicly available on ISO’s Online Browsing Platform, defines a complaint as a written, electronic or oral communication that alleges deficiencies related to the identity, quality, durability, reliability, usability, safety or performance of a medical device that has been released from the organization’s control or related to a service that affects the performance of such medical devices. When a veterinary customer alleges a false count or an instrument failure, that communication can meet the 3.4 definition and should enter a controlled intake record. Evaluation of whether the communication is a complaint is not the same as concluding a malfunction or opening CAPA.

For manufacturers subject to the QMSR, 21 CFR 820.35(a) adds complaint-record content on top of ISO 13485:2016 clause 8.2.2. Those manufacturers shall maintain records of the review, evaluation, and investigation for complaints involving possible failure of a device, labeling, or packaging to meet any of its specifications. If an investigation has already been performed for a similar complaint, another investigation is not necessary, but the justification must be recorded. For complaints that must be reported to FDA under 21 CFR Part 803, complaints the manufacturer determines must be investigated, and complaints the manufacturer investigated regardless of those requirements, 21 CFR 820.35(a) requires recording at least:

• Device name.

• Date the complaint was received.

• Any unique device identifier (UDI) or universal product code (UPC), and any other device identification.

• Complainant name, address, and phone number.

• Nature and details of the complaint.

• Any correction or corrective action.

• Any reply to the complainant.

Those fields are a lawful skeleton for manufacturers who already keep that standard. They do not create a Part 803 clock for exclusive animal-device makers. 21 CFR 820.10(b)(3) maps ISO 13485 clause 8.2.3, Reporting to regulatory authorities, to notification of FDA for complaints that meet 21 CFR Part 803 reporting criteria. That mapping applies to manufacturers subject to the QMSR as described in 21 CFR 820.1(a). It is not a reportability decision on an incomplete veterinary analyzer file, and it does not override the CVM statement that exclusive animal-device makers are exempt from post-marketing reporting.

Use the public 3.4 definition and the 820.35(a) list to build the veterinary worksheet: device identity, date received, identification that exists, complainant contacts, nature and details without editorializing, what remains unknown, and any reply. Do not treat logging the intake record as opening CAPA, admitting a nonconformance, or starting a human MDR clock. Human complaint encyclopedias remain at complaint handling, quality investigation, CAPA, and post-market surveillance.

Manufacturer field-feedback form and four hypothetical reports

The original deliverable is a manufacturer veterinary field-feedback form plus four distinct, clearly hypothetical reports. Required columns are at least: species; sample matrix; analyzer identity; lot and/or software; operator observation (what was seen, without a cause); QC source (present, missing, or unknown); missing records; reviewer; and next evidence request. The field table below is the worksheet. The second table applies it to four internally consistent hypothetical rows. Every row is labeled hypothetical. Unknown fields stay unknown. No row is proof of a reportable malfunction, a completed investigation, diagnostic or dosing advice, or FDA animal-device approval. No numeric clinical limits, mix-up rates, or real cases are invented.

The four hypothetical reports below are pedagogical models. They are not real clinics, animals, or proprietary files. Each row is internally consistent and labeled hypothetical. The first leaves species, matrix, and analyzer identity missing on a “wrong CBC” narrative. The second records an instrument flag claimed as analyzer failure while QC is not supplied. The third records a clinical-causality sentence with no lot or software and no QC source. The fourth records a post-service unexpected flag with software and reagent lot unknown.

Read those four rows as missing-record drills, not as case reports. Hypothetical 1 shows why a “wrong CBC” sentence cannot be triaged as a defect while species, matrix, and analyzer identity are blank. Hypothetical 2 shows why an instrument flag plus an analyzer-failure claim still needs a QC-source field and the actual flag text. Hypothetical 3 shows why a clinical-effect sentence is parked as unsupported when lot, software, and QC source are absent. Hypothetical 4 shows why a post-service flag is still an observation when software and reagent lot are unknown. None of the rows authorizes a loaner, a recall, a clinical instruction, or a Part 803 decision.

flowchart TD
    A[Incoming veterinary analyzer narrative] --> B[Record operator observation without a cause]
    A --> C[Record species and sample matrix]
    A --> D[Record QC source as present missing or unknown]
    A --> E[Park clinical-effect language as unsupported]
    B --> F[List missing records]
    C --> F
    D --> F
    E --> F
    F --> G[Next evidence request]
    G --> H[Do not close as defect QC miss or causality]

What the reviewer still cannot decide from an incomplete file

The reviewer named on the form owns the next evidence request. That person does not convert an incomplete veterinary narrative into a finished investigation. Until the missing records exist, the file does not support treating the narrative as an instrument defect, a clinic QC miss, or a clinical-causality conclusion. It also does not support a human MDR decision, a recall, or diagnostic or dosing advice.

If the clinic already used Form FDA 1932a, record that fact and any manufacturer case number the sender provides. That prior-report field is still not a causality finding. If the product line is dual human and animal, keep the animal-use narrative in its own file identity fields rather than assuming it is a human 21 CFR Part 803 event. If exclusive animal-device status is confirmed, FDA’s current literacy page still exempts that maker from post-marketing reporting; the remaining job is still to record the observation, identity fields, QC source, and what is unknown.

The worksheet is useful only if unknown stays unknown. Do not invent species ranges, QC frequencies, control standard-deviation windows, or real cases to make a row look complete. Do not recast VetMedGuide’s clinic QC article as this form. Do not recast the published veterinary regulation guide or the human complaint, investigation, CAPA, PMS, POCT, home-use invalid-result, or calibrator-traceability guides. When the requested records arrive, a later technical evaluation can begin. That later step is outside this intake page.