"My concern in that area
is that there is a wide variation in treatments from 3 times 5 days apart
to up to nearly 30 days at varying intervals. That also biases the outcome."

I think we know the following about OVA:
1.  It is only effective against phoretic mites, but is very highly lethal to phoretics
2.  After application it takes about four days to kill all the phoretic mites.
3.  After application it is only effective for about four days and it is not clear at all that an application kills any mites that emerge more than a day or two after application.
4.  The % of total mites that are phoretic can range from as low as 15% to as high as 100% depending on a complex variety of factors I do not understand at all well other than it is very influenced by amount of brood.
5.  Brood tolerates multiple OAV treatments with no harm at all.

So, I spent quite a bit of time playing "what if" with Randy's mite model spread sheet.  I played with application intervals, numbers of applications  and time of year.  One of my major interests is how to control mite population growth during spring to fall when brood is present.  A lot of my interest was driven by the various suggestions on timing and number of treatments.  Now, Randy's model does not automatically accommodate any treatment interval less than twice a month.  So, in some cases I had to do some manual calculations and enter a total % kill for more than one treatment.   This is easy enough to do.  I looked at things ranging from several treatments at five day intervals, to several treatments at twice a month or even less often intervals.

What I learned from this was final mite counts in late fall when going into winter depended very little on treatment intervals.  You could treat at five day intervals for five treatments or you could treat twice a month for five treatments and the October mite numbers were very close to the same.  What was really important was the total number of independent treatments during periods where brood is present.  An independent treatment is simply a treatment far enough time wise from other treatments that the two have no over lap in kill.  Thus, based on the results from Randy's model I think getting hung up on treating every five days or every seven days or even every fourteen days is a mistake.  In any of these cases what really matters is the total number of treatments.  In my climate it looks to me like you need five independent treatments.  It also makes very little difference when you do those five treatments.  One in June, two in July and two in August are just as good at holding the Oct 1 mite counts down as two in August and three in Sept.  One caution thou.  I live in an area of low hive density and nearly zero ferals.  So, the mite immigration numbers I used were at the low end of Randy's model.  A high immigration rate would likely need another treatment or two in Oct along with the earlier five treatments althou I did not run such cases in the model.

There is one factor that may say spaced out treatments are even better.  While no one has reported data that says multiple OVA treatments of adult bees harms them it is hard to believe there are zero negative effects.  By spacing treatments out at least two weeks apart each adult bee experiences only a couple of treatments during its life on average.  As long as you do an equal job of controlling mites less treatments during any given bees life seems to be a move in the right direction.

Of course if you have any concern at all about how good a job you have done you can always do a broodless OAV in Nov or Dec and take the counts down to effectively zero.  But, as it is viruses that actually kill the hive my thinking is you want is to get the counts low in Aug so you are set up to raise healthy winter bees.  If you simply kill mites on bees that are loaded with viruses you are too late and are going to get a dead hive even if the mite count is zero after the broodless treatment.

I strongly encourage everyone to spend several hours with Randy's model and learn the population dynamics of mites in your area.  There is a huge amount of information crammed in that model.  I must have spent fifteen hours playing with it so far and am sure I am not done learning.

One caution.  I live in NE Ohio and put in brood patterns for my local area.  If you live 200 miles south of me you need to enter different data than I entered and you should expect to see different mite build ups versus month of the year.  That could easy demand one or two extra OVA treatments versus what I think is adequate for me.  I should also add at my location and with the bees I run I am more and more coming to the conclusion that anything above a 1% by an alcohol wash on August 1 is time for me to start to get concerned, particularly on big production hives.  If you are convinced your bees are ok with a 3% August 1 wash you will get by with fewer treatments than I think optimum for me.  Randy's model allows you to pick your % targets as you see fit.  It does seem clear based on largely anecdotal stories that if you live where your bees brood much of the winter you can tolerate higher mite %s than I can tolerate with my long cold winter.  If your bees are brooding you have a constant supply of new bees coming along where the bees in my hives today are pretty much the same bees that will be in my hives next March 1.  Those younger bees can compensate for older bees dying early in a warmer climate.  Where I live older bees dying early ends up with a hive too weak to make it by March due to a small cluster that can not make enough heat to recover.

I sure welcome arguments about why I am wrong and what I have missed.  Mites are a real pain.  I wish every one of them was dead.  I do not like suggesting such a large number of treatments even if it is dirt easy to do with a Provap.  I can do a hive a minute easy enough.  But, I still hate doing so many treatments.

Dick

             ***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software.  For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html