This work was funded by Comunidad de Madrid under the SINFOTON2-CM Research Program (S2018/NMT-4326- SINFOTON2-CM), and by Universidad Rey Juan Carlos with research project “Materiales nanoensamblados para sensado y manipulación de luz en amplio rango spectral”, reference M2417, and Grupo de alto rendimiento DELFO_URJC, reference M2363, under research program “Programa de fomento y desarrollo de la investigación”. M.M., J.L, and V.T. acknowledge that part of this work has been developed within the RollFlex project, part-financed by Interreg Deutschland-Danmark with means from the European Regional Development Fund and the Southern Denmark Growth Forum. Finally, all authors acknowledge the support from the EU Framework Program Horizon 2020 for MNPS COST ACTION MP1307 StableNextSol. The performance of organic solar cells has improved significantly in recent years due to the use of non-fullerene acceptors (NFA). While processing additives are typically added to the active layer blends to enhance device performance in NFA organic solar cells, their impact on device degradation remains unclear. In this work we have compared the performance, in pristine and degraded state, between air-processed slot-die coated NFA ITOfree organic solar cells with and without the processing additive DIO, using a structure of PET/Ag/ZnO/PBDB-T: ITIC/FHC PEDOT:PSS. We observed an improvement in the power conversion efficiency of the devices when adding DIO, from 4.03% up to 4.97%. The evolution of the performance for both devices under ISOS-L1 life testing protocol reveals that the drop in efficiency is mainly due to a decay of JSC for both cells. In the short time scale the efficiency of non-DIO cells decays faster than the DIO cells, whereas in the long time scale the efficiency of non-DIO cells tends to stabilize sooner. Carrier mobilities estimated from impedance measurements decrease with time at similar rate for both degraded samples. Besides, DIO devices present a steep increase of the series resistance with time causing a decrease of the FF and thus of the efficiency. Moreover, in both degraded devices, the open-circuit voltage saturates with increasing illumination intensity. Numerical simulations reveal that a reduced anode work function of 5 eV is needed to fit experimental data.