When you read “98% accuracy” on a rapid test box or product page, it is easy to think: so it tells the truth 98 times out of 100. The reality is more interesting – and much more useful for your decisions. In this article we explain in plain language what sensitivity, specificity and agreement mean, why a test result can be falsely positive or falsely negative, and how to use a home rapid test to get the most reliable answer possible. We illustrate all of this with the real validation data from the manufacturer of MancsLab tests.
In short: the three numbers worth knowing
- Sensitivity: the percentage of truly infected animals in which the test gives a positive result. The higher it is, the rarer false negative results are.
- Specificity: the percentage of truly healthy (uninfected) animals in which the test gives a negative result. The higher it is, the rarer false positive results are.
- Agreement (overall accuracy): in how many of all the samples tested the rapid test result matched a laboratory reference method (for example PCR or ELISA).
We did not invent these concepts: the international standard for validating diagnostic tests, the manual of the World Organisation for Animal Health (WOAH), also uses them to describe how reliable a test is.
A real example: what a validation study looks like
Take the CPV Ag parvovirus rapid test. The manufacturer tested 300 dog faecal samples with both the rapid test and PCR. PCR is a laboratory method that detects the virus's genetic material, so it served as the “gold standard”. The result:
| Infected according to PCR | Not infected according to PCR | |
|---|---|---|
| Rapid test positive | 147 (true positive) | 2 (false positive) |
| Rapid test negative | 3 (false negative) | 148 (true negative) |
This is where the figures shown on the product page come from:
- Sensitivity: the test detected 147 of the 150 infected dogs → 98.00%.
- Specificity: it correctly showed 148 of the 150 uninfected dogs as negative → 98.67%.
- Agreement: the two methods agreed in 295 of the 300 samples → 98.33%.
As you can see, the test can be wrong in two ways, and the two kinds of error have different consequences. A false negative result “clears” a sick animal, while a false positive “flags” a healthy one. That is why looking at a single number is not enough.
Why “98%” does not mean the same for everyone: the key is who you test
This is the most important idea in this article, and most people are not aware of it. How likely a positive result is to be true depends not only on the test, but also on how likely infection was before testing. The literature calls this the positive predictive value. Let's look at the same CPV Ag test in two situations, with 1,000 dogs in each.
Situation 1: puppies with symptoms
Suppose we test unvaccinated puppies with sudden bloody diarrhoea and vomiting, in whom parvo is strongly suspected – say every second one is actually infected.
- Of 500 infected puppies, the test shows about 490 as positive (98%).
- Of 500 uninfected puppies, about 7 will be false positives (1.33%).
- So of the approx. 497 positive results, approx. 490 are true → approx. 99%.
In a symptomatic animal where infection is suspected, a positive result is therefore a very reliable signal.
Situation 2: routine screening of dogs without symptoms
Now suppose we screen adult, vaccinated dogs that look completely healthy. Infection is rare among them – say 20 in 1,000 are affected.
- Of the 20 infected dogs, the test finds about 20.
- But of the 980 healthy dogs, about 13 will be false positives – the same 1.33%, just calculated from many more healthy animals.
- So of the approx. 33 positive results, only about 20 are true → approx. 60%.
Same test, same specificity, yet a positive result means something quite different. If infection is rare in the group tested, even a few false positives become “concentrated” alongside the true positives. A negative result, on the other hand, is almost certainly correct in this situation (here in more than 99.9% of cases).
What does this mean in practice?
- Test based on symptoms. A rapid test is at its strongest when there is a specific suspicion: diarrhoea, vomiting, coughing, lethargy, or known exposure (for example a new dog has arrived, or your dog has met an infected animal).
- Always have a vet confirm a positive result in an animal without symptoms. Not because the test is bad, but because in that situation a positive result is less likely to be true.
- A negative result in an animal without symptoms is good news – in an animal with symptoms, however, you need to look at the whole picture (more on this below).
Where can false negative results come from?
A false negative result means the animal is infected, but the test does not show it. The cause is usually not the quality of the test, but biology and timing:
- Intermittent shedding. Parvovirus, coronavirus and especially Giardia are not shed in the faeces in the same amount every day. A sample taken on a “bad day” may contain little detectable antigen. That is why, if symptoms persist, we recommend repeating the test within 2–3 days.
- Testing too early. Some pathogens can only be detected weeks or months after infection. Heartworm antigen, for example, appears in the blood of most dogs only about 6–7 months after infection, and antibodies against FIV take weeks to develop.
- Unsuitable sample. Too little sample, a sample that is not fresh, or the wrong sample type (for example a blood sample for a faecal test) worsens the result. On every product page and in every instruction leaflet we describe exactly what sample is needed.
- Storage and temperature. The cassettes should be stored at 4–30 °C, and the test itself must be performed at 15–30 °C. Freezing or direct sunlight can damage the reagents.
Where can false positive results come from?
- Recent vaccination. After a vaccine containing live, attenuated virus (parvo, panleukopenia), the vaccine virus may also be shed in the faeces for a few days and give a positive result. If the animal has been vaccinated recently, always tell your vet.
- Maternal or vaccine-induced antibodies. Antibody tests (for example FIV Ab) cannot distinguish antibodies produced because of infection from those received from the mother or induced by vaccination. In kittens up to 6 months of age this can be misleading.
- Reading too late. Most MancsLab tests must be read between 5 and 10 minutes. After 15 minutes a faint line may appear that is not a true positive – which is why a result read after 15 minutes is invalid.
Antigen or antibody? What the test measures matters
The abbreviations Ag and Ab in the test names fundamentally determine what question the test answers:
- Antigen test (Ag): detects the pathogen itself or part of it. It shows whether the pathogen is present now. Examples are the CPV, Giardia, FPV, FeLV and heartworm tests.
- Antibody test (Ab): detects antibodies the body has produced against the pathogen. It shows whether the animal has ever encountered the pathogen. Examples are the FIV and FIPV Ab tests.
The two FIP tests are a good example. The FIPV Ag shows from faeces whether the cat is shedding coronavirus. The FIPV Ab shows from blood whether it has ever encountered it. Neither diagnoses FIP on its own – this is always determined by a vet based on the clinical picture and further tests.
The reference method follows the same logic: during validation, antigen tests are typically compared with PCR, and antibody tests with ELISA (a laboratory antibody measurement).
Rapid test or laboratory test?
The two methods are not competitors but complement each other. Laboratory tests (PCR, ELISA) detect even very small amounts of pathogen or antibody, and in some cases give quantitative results, but they require sampling, transport and processing, so the result takes hours or days and is available through a vet. A rapid test, by contrast, gives a yes/no answer in 5–10 minutes, at home. It is most useful when a decision has to be made quickly: for example with a puppy with sudden diarrhoea, where every hour counts because parvo is suspected. So the right order is often: rapid test at home → vet → laboratory confirmation if needed.
Validation data for MancsLab tests in one place
The table below shows the results of the validation studies carried out by the manufacturer (Chongqing Safvet Biotech Co., Ltd.). All data can also be found on each product's page, together with the PDF instructions for use.
| Test | Sensitivity | Specificity | Agreement | Samples / reference |
|---|---|---|---|---|
| CPV Ag (parvo) | 98.00% | 98.67% | 98.33% | 300 / PCR |
| Giardia Ag | 98.57% | 99.13% | 99.00% | 300 / PCR |
| FPV Ag (panleukopenia) | 98.00% | 97.50% | 97.71% | 350 / PCR |
| FIPV Ag | 96.67% | 97.33% | 97.00% | 300 / PCR |
| FIPV Ab | 95.88% | 96.25% | 96.06% | 330 / ELISA |
| FIV/FeLV – FIV component | 96.88% | 98.13% | 97.50% | 320 / ELISA |
| FIV/FeLV – FeLV component | 97.77% | 97.05% | 97.42% | 350 / PCR |
| CPV/CDV – CDV (distemper) component | 96.67% | 97.33% | 97.00% | 300 / PCR |
| 3-in-1 – CCV (coronavirus) component | 98.50% | 97.33% | 98.00% | 350 / PCR |
| 3-in-1 – Giardia component | 97.50% | 96.00% | 96.67% | 360 / PCR |
The combination tests (2-in-1, 3-in-1, 4-in-1) detect each pathogen on a separate test line, so the manufacturer validated each one separately; all component data can be found on the product pages. For two tests – Lyme Ag and heartworm (HW Ag) – we deliberately do not publish percentage figures while we clarify the details of the validation table with the manufacturer. We would rather give no figure than publish an inaccurate one.
How to read a “faint line”
One of the most common questions: what does it mean if the T (test) line is only faintly visible? The answer is simple: any visible T line is a positive result, provided the C (control) line has also appeared. The intensity of the line roughly reflects the amount of antigen or antibody in the sample, but not the severity of the disease. If, however, the C line does not appear, the test is invalid and must be repeated with a new cassette and a new sample.
Five rules for the most reliable result
- Read the instructions for the product. The sample type (faeces, blood, eye discharge, tick) and amount differ from test to test.
- Use a fresh sample, and perform the test immediately after opening the foil pouch.
- Stick to the timing: read between 5 and 10 minutes; after 15 minutes the result is invalid.
- Negative result, but the symptoms persist? Repeat the test within 2–3 days, or see a vet.
- Positive result? Take the cassette (or a photo of it) to your vet. They can carry out a confirmatory test and start treatment.
So what is a home rapid test good for – and what is it not?
A rapid test is a pre-screening tool. It uses the same immunochromatographic technology as the rapid tests used in veterinary practices, and based on the validation data it gives a reliable picture if you choose well when and whom to test. It helps you decide how urgent the situation is, and lets you arrive at the vet with concrete information. What it cannot do: it does not replace a clinical examination, the medical history or laboratory confirmation. If you are unsure when a test is enough and when you need to go to the vet straight away, read this too: Home rapid test: when to use it and when to go to the vet
Frequently asked questions
If the test is 98% accurate, does that mean it is wrong 2% of the time?
It is not that simple. Sensitivity and specificity show separately how often the test is wrong in infected and in healthy animals. How likely your particular result is to be correct also depends on how likely infection was before testing. In an animal with symptoms, a positive result is very reliable; in routine screening of an animal without symptoms, it is worth having a positive result confirmed.
What is PCR, and why is the rapid test compared with it?
PCR is a laboratory method that detects the pathogen's genetic material even in very small amounts. That is why it serves as the reference (the “gold standard”) when rapid tests are validated. Antibody tests are typically compared with ELISA, a laboratory antibody measurement.
Why was the test negative if my dog has diarrhoea?
Diarrhoea can have many causes, and the test does not check for all of them. In addition, pathogens may be shed intermittently, so a sample may contain little detectable antigen. If symptoms persist, repeat the test within 2–3 days or see a vet.
Can the test be positive after vaccination?
Yes, this can happen. After a vaccine containing live, attenuated virus (for example against parvo or panleukopenia), the vaccine virus may be detectable in the faeces for a few days. If the animal has been vaccinated recently, always tell your vet.
Does a faint T line also count as positive?
Yes. If the control line (C) has appeared, any visible test line (T) indicates a positive result, provided you read it within 5–10 minutes. A line that appears after 15 minutes cannot be interpreted.
Read more in our Knowledge Base
- Parvovirus in dogs – what every owner needs to know
- Giardia in dogs – the hidden parasite that is easily misdiagnosed
- Diarrhoea in dogs: when is it harmless and when is it a warning sign?
- FIV and FeLV – why every cat owner should test
- FIP – no longer a death sentence: what you need to know
Sources
- Chongqing Safvet Biotech Co., Ltd. – manufacturer validation data, in the instructions for use of each product (PDFs downloadable from the product pages).
- World Organisation for Animal Health (WOAH): Terrestrial Manual, Chapter 1.1.6 – Principles and methods of validation of diagnostic assays for infectious diseases.
- Kansas State Veterinary Diagnostic Laboratory: Canine Heartworm Antigen Tests – on the timing of when heartworm antigen appears.
Important: MancsLab rapid tests are intended for pre-screening at home and do not replace a veterinary diagnosis. In the case of a positive or uncertain result, or if symptoms get worse, always contact your vet.