À Deus por me amparar nos momentos difíceis, me dar força interior para superar as dificuldades, mostrar os caminho nas horas incertas e me suprir em todas as minhas necessidades. Muito especialmente, desejo agradecer ao meu orientador Prof. Doutor Nuno Lunet, pela disponibilidade, atenção dispensada, paciência, dedicação e profissionalismo. À minha co-orientadora, Filipa Fontes, por tanto cuidado, atenção, incentivo e pela disponibilidade em tempo integral. Aos colegas do Mestrado em Saúde Pública do ISPUP pelo companheirismo e pelos momentos de entusiasmo partilhados em conjunto. À minha irmã, pelos momentos de suporte e carinho. Ao meu marido, Tarcisio, pelo incentivo, compreensão e encorajamento, durante todo este período. Ao meu pai, por ter me dado o dom precioso da vida, o carinho e pelo apoio em minhas escolhas. À minha mãe, pelo exemplo de força e superação, pelo exemplo de vida e por ter me ensinado que o maior bem adquirido em nossas vidas é a educação e o conhecimento pois estes nada nos pode tirar. 12 Background: Cognitive decline associated with breast cancer diagnosis and treatment has been increasingly described. We aimed to quantify the association between cancer treatments and cognitive decline in early stage breast cancer patients during the first year of follow-up, taking into account the potential effect modification of anxiety before treatment. Patients and Methods: We conducted a prospective cohort study that included 506 women with newly diagnosed breast cancer, followed for one year. Patients were evaluated with the Hospital Anxiety and Depression Scale (HADS) and the Montreal Cognitive Assessment (MoCA) before starting treatment and at the 1-year follow-up. We described standardized scores of the mean variation in each MoCA subdomain between the two evaluations, and used Poisson regression to compute adjusted relative risks (RR) and corresponding 95%CI for the relation between different patients' and treatments' characteristics and cognitive decline, according to the patients' anxiety status at baseline. Results: During the first year after diagnosis, 34 patients presented cognitive decline and 38.0% were classified as having anxiety prior to treatment. Within each subdomain of the MoCA, the participants with cognitive decline had a significantly higher variation in the attention and memory subdomains, among those without anxiety, and for attention, visuospatial/executive function and orientation, for those anxious. The presence of anxiety at baseline was not significantly associated with cognitive decline (RR=1.43 95%CI: 0.73, 2.81), but modified the effect of chemotherapy (P for interaction=0.030), with the RR for chemotherapy being higher among patients with no anxiety prior to treatment (adjusted RR=4.35, 95%CI: 1.26, 15.02, for schemes including doxorubicin + cyclophosphamide). Conclusions: Anxiety modified the association between chemotherapy and cognitive decline; among women with no anxiety at baseline the risk of cognitive decline was four-fold higher among patients undergoing c...