Forecasting Mortality by Causes of Deaths Coherent to the All-Cause Mortality Forecasts Using the Lee-Carter Model

Julia Calazans , Centro de Desenvolvimento e Planejamento Regional (CEDEPLAR)
Bernardo L. Queiroz, Universidade Federal de Minas Gerais

Mortality forecasts by cause of deaths are a powerful source of information on the future burden of disease. Despite its relevance, mortality forecasting by cause of death is still a major methodological challenge for demographers. Several studies have been developed to project mortality by a specific cause of death. Although these studies are of great relevance, they usually do not incorporate the demographic trends for total mortality in the analysis. The objective of this study is to propose an alternative method to forecast mortality by cause that is coherent to the total mortality forecasts, based on the Lee-Carter model. The model was tested using Brazilian mortality database for four causes of death: cardiovascular disease, neoplasms, external causes, and other diseases. Despite some limitations of the data, this alternative proved to be satisfactory for the mortality forecast by cause of deaths in the Brazilian context.

See extended abstract

 Presented in Session 31. Mortality Predictions