INICSA   23916
INSTITUTO DE INVESTIGACIONES EN CIENCIAS DE LA SALUD
Unidad Ejecutora - UE
congresos y reuniones científicas
Título:
MULTILEVEL AND FLEXIBLE MODELS TO STUDY THE LIFETIME OF CHRONIC DISEASES: COMPARATIVE ANALYSIS OF COHORTS FROM BRAZIL, ARGENTINA AND ITALY.
Autor/es:
VAZ DE ARRUDA SILVEIRA L; DIAZ MP; DE LA QUINTANA AG; OSELLA AR
Lugar:
San Pablo
Reunión:
Congreso; 27th Conference of the International Society for Environmental Epidemiology - Addressing Environmental Health Inequalities; 2015
Institución organizadora:
Universidad de San Pablo y ABRASCO
Resumen:
Study of the elderly in this century is of primary interest as population ageing brings up a new health profile for this group. Chronic diseases such as diabetes mellitus, cardiovascular disease, obesity and cancer are prevalent diseases worldwide and exhibit a strong association with ageing. Follow-up studies in Brazil (n1=400), Argentina (n2=1200) and Italy (n3=5000) have been conducted in order to identify risk factors in chronic diseases survival. A comparative study of elderly populations with chronic diseases (cancer, hypertension, cardiovascular diseases, among others) was built up to compare time to several end-points. Cox proportional hazards regression was the starting modelling point and then relaxing the proportional hazards assumption by adapting cubic splines (flexible parametric models). A hierarchical structure was then applied in order to account for variability coming from each study (Brazil, Argentina and Italy). Preliminary results indicate that there are different covariates patterns conditioning chronic diseases survival on different selected end-points (metastasis, recurrence, etc.) between countries. In Italian cohort, moderate physical activity was associated with a cancer hazard rate reduced of about 20% compared with sedentary subjects. Considering metastasis occurrence (end-point) in women with breast cancer (Argentina), the presence of other chronic diseases (hypertension, diabetes, cardiovascular diseases) implied a higher survival rate conditioned by the nutritional and socio-economic status of subjects. As regard to Brazil study, similar covariates significantly reduced survival of 69 years old or older subjects.