Neonicotinoid insecticide use has been suggested as a cause of honey bee colony decline; however, detection rates and concentrations of neonicotinoid insecticide residues in field-collected honey bees have been low. We collected honey bee and beebread samples from apiaries in agricultural, developed, and undeveloped areas during 2 years in Virginia to assess whether landscape type or county pesticide use was predictive of honey bee colony exposure to neonicotinoid insecticides. Trace concentrations of the neonicotinoid imidacloprid were detected in honey bees (3 of 84 samples, 2.02-3.97 ng/g), whereas higher concentrations were detected in beebread (5 of 84 samples, 4.68-11.5 ng/g) and pollen (three of five pollen trap samples, 7.86-12.6 ng/g). Imidacloprid was only detected in samples collected during July and August and was not detected in honey bees from hives where neonicotinoids were detected in pollen or beebread. The number of hives sampled at a site, county pesticide use, and landscape characteristics were not predictive of neonicotinoid detections in honey bees or beebread (all P > 0.05). Field surveys may underestimate honey bee exposure to field-realistic levels of pesticides or the risk of exposure in different landscapes because of low detection rates. Undetectably low levels of exposure or high levels of exposure that go undetected raise questions with regard to potential threats to honey bees and other pollinators.