In 1993, a project was established to develop more sustainable arable farming systems. Fields were divided into two plots to compare the environmental and economic performance of conventional and integrated farming systems (IFS). As environmental impact assessment (EIA) techniques were, at that time, in their infancy, differences in crop protection strategies could only be evaluated using quantity of pesticide applied. EIA techniques are now more sophisticated and using data from the original project, potential impacts have been re-assessed using the p-EMA model. The results show that using the quantity of pesticide applied as a surrogate indicator of risk is unsound, and that where the site has few sensitive habitats there may be little difference in predicted environmental impact between the different strategies. However, if the site is habitat-rich, IFS strategies can significantly lower the predicted risk. The results also demonstrate the need for substantial site and pesticide information to improve IFS strategies.