Many pollination studies with honey bees have examined the effect of colony density on crop yield and yet overlook the effect of variation in the population size of these colonies. High colony density in northern highbush blueberry has been met with concerns from beekeepers who feel higher densities will intensify outbreaks of European foulbrood (EFB, Melissococcus plutonius, Truper and dé Clari), a honey bee brood disease. The purpose of this study was to confirm the prevalence of EFB in colonies pollinating blueberries and to determine whether field-level variation in the population of adult workers in colonies explained variation in blueberry fruit set and/or yield. We addressed these objectives over the course of two production seasons at 13 commercial blueberry fields in Oregon, USA, stocked with identical densities of 10 colonies/ha. We confirmed that all colonies had negligible symptoms of EFB at the start of blueberry pollination, but 53% of colonies in 2019 and 41% in 2020 had symptoms immediately following the pollination season. We also validated a method for rapidly assessing adult honey bee colony populations, namely by counting the rate of foragers returning to colonies, and it was found to be strongly correlated to true internal adult bee population independent of year and ambient temperature at the time of evaluation. Using returning forager counts, we determined there was considerable variation in the average population of colonies at each field, ranging from an estimated 10,300 to 30,700 adult worker bees per colony. While average colony strength did not predict variation in fruit set, it was related to variation in yield, independent of year. Our linear model of flight count (as a proxy for colony strength) predicts estimated yield increases of up to 25,000 kg/ha of blueberries could be achieved by colonies stronger than the recommended six frame minimum, suggesting that higher pollination benefits could be achieved without increasing hive density if stronger colonies are promoted.