Mortalidad adulta atribuible a la demencia debido a la enfermedad de alzheimer, Brasil 2009 2013: una perspectiva Bayesiana

• Ano de defesa: 2016
• Autor(a) principal: Juan de Jesus Sandoval
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: spa
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Demência; Brasil; Demografia; Analise Bayesiana; Mortalidade; Alzheimer; Alzheimer, Doença de Mortalidade; Mortalidade Brasil 2009-2013
• Link de acesso: http://hdl.handle.net/1843/FACE-AAZQ8F

Objective: Establish a methodology to estimate with higher credibility, the measures of older adult mortality in Dementia and Alzheimer's disease (DA) on population. Methods: We used micro-data from the SIM-DATASUS at the 27 capitals of the states of Brazil, in older adult by residence's place between of years 2009 and 2013. There were corrections of subrecords and too it was adjusted at the finally mortality rates obtained by using Bayesian full methods by sex, age and years schooling. A priori probability distributions were obtained by the meta-analysis. We estimated the mortality rates ratios and the population attributable risk percentage. Results: The corrections of subrecords gave a total of 51,307 deaths for all types of dementia, of whom approximately 36,941 corresponded to deaths from Alzheimer's disease. The estimation of the median rate of mortality by the Bayesian model was 223.3 deaths per 100,000 populations for all types of dementia for men and 225.3 for women. Deaths caused by DA were 160.8 deaths per 100,000 inhabitants for men and 162.3 for women. There were not significant differences by sex, nor by significant growth mortality rates during the study period. The contribution of the Alzheimer's disease in the Brazil's mortality was 27.56 % in the elder people with zero years schooling, independent of age, sex in the study time (ICR95% 24.64 - 30.42%). Conclusions: The mortality rates for dementia and DA proved to be higher than the developed countries. The impact of low education in these diseases was therefore that one-third of the deaths on the older population were provided by the population without studies. The increase of age have been related a risk multiplicative develop dementia or DA and death posterior for this cause. In order to obtain good estimates in Public Health, the modeling Bayesian was important to correct subrecords in Alzheimer's mortality like also in other diseases of low occurrence.