Abstract Introduction Pesticide exposure increases the risk of chronic disease among farmers. Understanding exposure is necessary for epidemiological and regulatory purposes. Since 2014, worker exposure has been assessed in the registration process by EFSA, using the OPEX model. Data specific to fruit-growing workers is limited to five European studies conducted in the 1980s by pesticide companies, among others. We compared exposure predicted in the regulations with that measured in field studies. Methods In 2016-2017, dermal exposure to captan and dithianon was measured in French farmers during 121 days of re-entry (net folding and deployment, thinning) and harvest, using patches and cotton gloves. Exposure was calculated using several parameters (task, personal protective equipment (PPE), treatment schedules). Exposure was recalculated from dislodgeable foliar residues (RDF) measured 2 to 312 days after application in 20 observations. Relationships between measured and calculated exposures were studied by linear regression. Results Exposure depended on PPE worn and tasks performed (thinning, net folding > harvest, net deployment) due to differences in pesticide accumulation on plants over the season. Most exposures calculated using default settings were 100 times higher than measured exposure. The model underestimated exposure recalculated with measured DFR in all observations for dithianon and almost all for captan. Discussion In the regulatory process, re-entry exposure is only calculated when it occurs immediately after application. Exposure measured up to 300 days after application was never zero. When re-entry was not immediate after application, the model underestimated exposure. Conclusion This demonstrates the importance of using field studies in the registration process to ensure a truly conservative approach.