Statistical method to rule out cross-contamination?

I hope there is someone who can help me find the right method for my needs.

I have conducted a study in which I captured bees at several different field sites. I dissected bees and examined them for the presence of a pathogen. Before doing the dissections, I assigned each bee a number. I cleaned my tools carefully between each dissection, but there is always the possibility of cross-contamination between specimens. I would like to use a statistical method to determine the probability of this happening.

If there was contamination between samples, then it should be more likely that I would get a positive finding immediately after dissecting a positive bee, than immediately after dissecting a negative bee.

Here's the rub:
1) the bees were stored in individual vials in a freezer, but grouped with others captured at the same site. I pulled them out of the freezer and assigned them numbers as I grabbed each one. I did not mix them all up, so field site effects are a concern. It did turn out that some field sites yielded more infected bees than others, as one would expect.

Any ideas of what test I should use?

I found seven instances in which I dissected bees from different field sites consecutively. In 2 of those instances, an infected bee came immediately after another infected bee. In 5 cases the bee dissected immediately after an infected bee was not infected. To me this suggests there was not between sample contamination.