In recent times there has been an increased interest in the use of blood or milk to test for pregnancy in dairy cattle. A number of our clients have used the test, or have asked us whether it is useful in their herds. Although we would be accused of having a vested interest, we think there are very few situations where blood or milk testing have any advantage over an experienced veterinarian using ultrasound or manual techniques to diagnose pregnancy. Many Warrnambool Veterinary vets are accredited by the Australian Cattle Veterinarians which means they have passed a strict examination, and can be audited to assess their accuracy over time. At Warrnambool Veterinary we do strive to be 100% accurate, and if we are unsure we will let you know!

The table below lists some comparisons between blood and milk testing and veterinary pregnancy testing.


Both the milk and blood pregnancy test detect a set of proteins which are produced by the placenta when a cow becomes pregnant. These proteins are called Pregnancy Associated Glycoproteins (PAG). The proteins are first detectable at 24 days of pregnancy, but only reach diagnostic usefulness at about 28-30 days post AI. They are still present after calving for a period of about 60 days.

How good is the test?
At first glance, the statistics which are provided by the company to support using the blood pregnancy test sound excellent with 99% sensitivity (ability to detect a pregnant cow) and 95% specificity (ability to detect an empty cow).

But what does that mean in a real life scenario?
Take a 200 cow herd where around 60% of cows are in calf (e.g. a normal first round pregnancy test in a seasonal herd). The blood test will come up with 5 incorrect results when compared with ultrasound.

  • Out of the 120 pregnant cows, the test will call 1 empty.
  • Out of the 80 empty cows, the test will call 4 pregnant.
  • Would you accept 5 errors out of 200 cow herd or 15 errors in a 600 cow herd from your vet?

What are the other limitations of the test?

  1. Can’t be used for ageing.
  2. The cost of the test is between $4 to $6, plus the cost of herd testing.
  3. Results take 2-3 days.
  4. Errors can occur with sample handling (Errors are really common when people are herd testing).
  5. Can’t detect other problems such as damage to the uterus, foetal mummies, pyometras (pus in the uterus).

What about using the test for early pregnancy diagnosis and re-synchrony?
The test performs slightly different depending on the stage of pregnancy. The highest percentages of errors occur when the test is used for early pregnancy diagnosis.

When used between 5 and 8 weeks of pregnancy, the test will call a higher percentage of empty cows pregnant than at other stages of pregnancy (8 incorrect calls out of 100 empty cows). This occurs when the foetus has died but the test is still able to detect the hormone from the placenta – in most dairy herds early foetal loss is relatively common especially in high producing animals.

What are the scenarios where the test could be used?
Where there are no facilities to safely perform pregnancy diagnosis. Where a skilled veterinarian is not readily available (a list and contact details of accredited PREgCHECKTM veterinarians is available on the website: