What is the best way to compare lab results?

Same units, same method, same timing

AMES, Iowa — The best way to compare lab results is to check that the tests really match: same method, same units, same sample type and similar timing. Once those line up, compare results over time instead of leaning on one number. Two labs can both be right and still report different numbers, because they measured slightly different things.

Most confusion comes from comparing results that look alike but aren't. A "bacteria count" from one lab and a "bacteria count" from another may use different methods and different reporting units.

Check the method and the unit

Standard plate count, Petrifilm aerobic count, preliminary incubation count, lab pasteurized count and coliform count measure different things. Look at the report for the method name. Units matter too: colony-forming units per milliliter, individual bacteria counts per milliliter and counts reported on a log scale are not the same. Somatic cell counts may be reported per milliliter in thousands, or as a linear score. Convert before you compare.

If the report doesn't say which method was used, call the lab and ask. A decent lab answers that in a minute.

Compare like with like

Check the sample. Was it a bulk tank sample, a cow composite, a quarter sample? Was it taken at the start of a milking or after the tank was agitated? Was it kept cold, and how long did it take to reach the lab? A sample that sat warm for hours will read higher for bacteria. A sample with preservative may be handled differently than a fresh one.

Compare samples taken at similar times. Milk from the first day after a tank cleaning isn't the same as milk from a tank with three days in it.

Know that numbers have noise

Counts of bacteria vary even in a single sample. Split a sample in two and send one to each lab, and the results will usually differ somewhat. A tiny difference isn't a trend. Look for changes that are big and repeat. Ask the lab about its method variability, or whether it takes part in proficiency testing programs where labs test unknown samples against peers.

Build a simple tracking sheet

Make a spreadsheet with columns for date, sample source, lab, method, units, result and any notes like weather, bedding change or new wash chemical. Add a chart. A line over several months tells you more than any single result. If you change labs, send a few split samples to both for a while to learn the difference between them, and mark the switch on the chart.

Compare against standards carefully

Regulatory limits are written for specific tests and specific sample types. A processor's premium program might use a different scale than the regulatory limit. Don't compare your result to a limit that applies to another method. Ask your milk buyer or state dairy regulator for the current limits and the method each one applies to.

For methods and interpretation, the National Mastitis Council at https://www.nmconline.org/ is a good source. Iowa State's ag decision resources at https://www.extension.iastate.edu/agdm/ are useful for spreadsheet tracking and cost thinking, and the Agricultural Marketing Service at https://www.ams.usda.gov/ has dairy program information.

Numbers don't lie, but they do come in different languages. Translate first, argue later.