Did my varroa treatment work?

The way to find out is a second mite count. Enter the mites per 100 bees from the wash before the treatment and from the wash after it. The page gives the percent drop, puts the after level next to the Honey Bee Health Coalition threshold and, if you give the date the treatment ended, the day to take the second wash. It says nothing about products or doses.

Mites per 100 bees before Mites per 100 bees after Colony phase (for the threshold) Treatment ended on (optional)

Each count from 0 to 100 mites per 100 bees. Only have mites and bees? The varroa mite calculator converts them.

Change in mite level

92.5 % drop

From 4.0 to 0.3 mites per 100 bees.

Below the threshold: the after level of 0.3 per 100 bees is under the Honey Bee Health Coalition threshold of 2 for Population Decrease (1.7 under).

When to take the second count

NOD Apiary Products writes in its guide to mite checks: "Your post-treatment monitoring should be done 21 days after treatment ends." It also says to capture mite levels before the treatment begins, since "the before and after mite levels can be compared to assess the efficacy of your treatment." That is the comparison this page does. Leave the end date empty if you only want the percent.

Worked example

A wash before the treatment shows 4 mites per 100 bees. A wash after shows 0.3. The difference is 3.7, and 3.7 divided by 4 is 92.5 percent. In Population Decrease the Coalition line is 2 per 100 bees, so 0.3 is under it. Open this example.

What a drop of that size means

Vita Bee Health, another treatment maker, explains in a post on efficacy rates that an efficacy of 96.8% "means that 96.8% of the mites that were in the colony during the treatment period were killed." Its example: 40,000 bees with an infestation rate of 2.5% is about 1,000 mites. Killing 968 of them leaves 32, a post-treatment rate of 0.08%. Run through this page, 2.5 before and 0.08 after is a 96.8 percent drop, the same figure. Vita's post states that a wash is how the infestation rate is measured before and after.

Reading the after level

The thresholds are the ones the varroa mite calculator uses, from Table 1 of the Honey Bee Health Coalition's Tools for Varroa Management (9th edition): 1 mite per 100 bees for Dormant and Population Increase, 2 for Peak Population and Population Decrease. At or above the number counts as above. The guide also says to sample again after any treatment to check it worked.

A big percent drop and a level above the line can both be true: 10 falling to 3 is a 70 percent drop and still over 2. The reverse also happens, and NOD notes that a post-treatment level that looks concerning may have come down from an extreme one. Read the drop and the after level together.

What a second wash can miss

My suggestion: use the same method both times (alcohol wash, same bee count), since a sugar shake and a wash can read differently. The alcohol wash steps list what to repeat.
My suggestion: a wash shows only the mites on the bees you sampled. Mites under capped brood are not in the jar, so a low count is a lower bound on what is in the colony.
My suggestion: mites can arrive from other colonies after the treatment ends. A rising count three weeks later may be new mites rather than a failed treatment, and one pair of counts cannot tell the two apart.
My suggestion: in 21 days the colony size changes too. Counting the bees in each sample, as the Coalition suggests, keeps the per 100 figure honest.