Sublethal effects of plant protection products on bees: learnings from an abandoned meta-analysis

Chemical plant protection products are designed to manage pests, weeds, and diseases in a wide range of crop plants, ensuring stable yields. Although effective in limiting pest populations, they can come in contact with nontarget organisms, including bees. Such unintended exposure can result in lethal and sublethal effects, such as altering movement, learning, and feeding behavior. Although the existence of such effects is well documented in bees, there have only been a handful of attempts to quantitatively assess their occurrence across products and species. In this study, we attempted to fill this gap by performing a formal meta-analysis on the occurrence, strength, and potential dose dependency of sublethal effects in adult bees moving beyond a simple analysis of their existence. We found an extensive bias in the existing data in terms of product-species pairs tested, with the majority of studies focusing on insecticides targeting the nicotinic acetylcholine receptor (e g., imidacloprid and clothianidin) in the honeybee. In addition, we found that a large proportion of the studies do not adequately report necessary methodological details and results to reliably extract the information needed to conduct a formal meta-analysis across chemical plant protection products and the different types of sublethal effects. We discuss the causes and consequences of these findings for studying sublethal effects in bees beyond demonstrating their existence and emphasize well-established recommendations to improve the reporting standards of such studies.