By favoring early diagnosis, mammography screening decreases breast cancer mortality and treatment costs. However, participation in public screening programs is low in many countries. We ran a randomized field experiment to assess whether costless manipulations of the informational content (restricted or enhanced information) and the framing (gain or loss framing) of the invitation letter to the breast cancer screening program in Messina (Italy) affects participation. We show that giving enhanced loss-framed information about the risks of not having a mammography increases the take-up. This manipulation is most effective among subgroups with lower baseline take-ups, thereby reducing inequalities in screening. Finally, subjects exposed to this manipulation are much less likely to postpone the screening conditional on participation, revealing enhanced awareness about the risks related with delayed participation.