2.3. Case definitions

1. Lost honey bee colony is a honey bee colony that:
a) is reduced to such a small number of bees that it cannot perform the normal biological activities needed for survival (brood rearing, resource gathering) or

b) has queen problems, such as drone laying queens or drone laying  worker bees in absence of a queen, which could not be solved or

c) was missing due to burglary, or didn’t survive fire, inundation, desert storms or similar causes unrelated to health problems.

d) no longer has any living bees present.

2. Weak honey bee colony:
A honey bee colony that is not considered as lost, but in which the number of bees is less than would be expected from the colony size observed at an earlier inspection.

3. Colony Depopulation Syndrome (CDS):

This is observed if a honey bee colony shows the following conditions within a certain time-frame:

a) reduced to no, or only a few remaining, living bees in the hive and

b) no, or only a few dead bees in or in front of the hive or at the apiary while

c) food is present in the hive (Van der Zee et al., 2012).

4. Colony Collapse Disorder (CCD)
This is observed if the following conditions are present:

a) a rapid loss of adult worker bees from affected honey bee colonies, as evidenced by weak or dead colonies with excess brood populations present relative to adult bee populations (vanEngelsdorp et al., 2009);

b) a noticeable lack of dead worker bees both within and surrounding the hive (vanEngelsdorp et al., 2009);

c) the delayed invasion of hive pests (e.g., small hive beetles (Neumann et al. 2013) and wax moths (Ellis et al., 2013)) into affected colonies and kleptoparasitism of affected colonies by neighbouring colonies (Cox-Foster et al., 2007);

d) the absence of Varroa and Nosema at levels thought to cause economic damage (vanEngelsdorp et al., 2009).

5. Time frames during which honey bee colony losses occurred can be distinguished as:

a) Time frames related to seasonal characteristics:
For example winter: the period between the moment that a beekeeper finished pre-winter preparations for his/her honey bee colonies and the start of the new foraging season (usually first of “spring”).

b) Fixed time frames:
For example: observations every half year.

It is difficult to come to conclusions on losses with a fixed timeframe approach since the outcome depends on beekeeper practices such as merging, splitting, buying and selling of colonies. Recalling the numbers of colonies involved in these practices later when a questionnaire is disseminated may easily lead to errors in the data (Van der Zee et al., 2012). Another problem is that no information is collected on when these increases/reductions were made within the timeframe, with the effect that colonies bought at the start of a time frame have the same weight in the risk estimation as colonies bought later on. Not recognising these problems may severely bias the outcome.