Context Understanding tuna diet and its drivers can provide valuable insights into the dynamics of pelagic ecosystems and their capacity to adapt to environmental and human-induced stressors. Aims To reconstruct tropical tuna dietary histories by using different metrics of stomach fullness and to assess their association with fisheries-related, environmental and biological covariates. Methods We examined stomachs from 8229 skipjack (Katsuwonus pelamis), yellowfin (Thunnus albacares) and bigeye tuna (T. obesus) captured in the western and central Pacific Ocean between 2001 and 2021, classified each on the basis of categorical and continuous fullness metrics, then built statistical models to gauge covariate effects. Key results Only models for the binary fullness metric (i.e. empty or non-empty) performed well. Tuna with empty stomachs were more likely to be caught on active gear (e.g. purse seine) and earlier in the day. Skipjack and bigeye tuna with empty stomachs were more likely to be associated with floating objects (e.g. fish-aggregating devices). Conclusions Our results add to the evidence that tropical tunas forage more effectively during the day and more actively when not associated with floating objects. At the individual level, tuna stomach fullness is highly variable. Implications Further research is needed to understand the factors governing this promising indicator of ecosystem change.