Seek and you shall find: An assessment of the influence of the analytical methodologies on pesticide occurrences in honey bee-collected pollen with a systematic review.

Honey bee mortality and colony losses have been reported worldwide. Although this phenomenon is caused by a combination of factors, agrochemicals have received special attention due to their potential effects on bees. In agricultural and urban environments bees are exposed to several compounds that may interact in unexpected ways, but information on the extent of pesticide exposure remains unclear. Several monitoring studies have been conducted to evaluate the field-realistic exposure of bees to pesticides after their release on the market. However, their outputs are difficult to compare and harmonize due to differences in the analytical methodologies and the sampling protocols (e.g. number of screened compounds and analysed samples, and detection limits (LODs)). Here, we hypothesize that the analytical methodologies used in the monitoring studies may strongly affect the pesticide occurrences in pollen underestimating the real pesticide exposure. By mean of a systematic literature review, we have collected relevant information on pesticide contaminations in the honey bee-collected pollen. Our findings showed that the pesticide occurrences were associated with the analytical methodologies and the real pesticide exposure has likely been underestimated in some monitoring studies. For four highly toxic compounds, the LOD used in these monitoring studies exceeded the doses that cause toxic effects on honey bees. We recommend that, especially for the highly toxic compounds, the LODs used in the monitoring studies should be low enough to exclude lethal or sublethal effects on bees and avoid "false negative" samples.