Abstract

CORREA, Paulo César Rodrigues Pinto; BARRETO, Sandhi Maria  and  PASSOS, Valéria Maria de Azeredo. Methods for estimating smoking attributable Mortality: a review. Epidemiol. Serv. Saúde [online]. 2008, vol.17, n.1, pp.43-57. ISSN 1679-4974.  http://dx.doi.org/10.5123/S1679-49742008000100005.

Reliable estimates of smoking attributable mortality (SAM) are essential for planning, funding and implementing anti-smoking programs successfully. This study aims to describe methods for estimating SAM, searching Medline, Lilacs and Bireme/PAHO/WHO databases for papers on SAM published up to October 20, 2005, and analyzing the potential use of such methods in Brazil. Were found and analyzed seven methodological approaches for the estimation of SAM. The population attributable fraction method (PAF) was first discussed by Levin in 1953. The second method used was the excess mortality, and later an 'indirect method' that infers the prevalence of smoking by observing the excess rate of lung cancer mortality in the target population, as compared with an unexposed reference population. Estimates of SAM can also be done using physician reports on death certificates. Three different methods estimate SAM considering the effect of potential confounding factors. There is no systematic registering of tobacco use in death reports in Brazil and correction of possible confounding factors does not cause important differences in estimates obtained by the PAF. For this reason, methods which could be used in order to estimate smoking attributable mortality in this country are the PAF, and the indirect method to infer smoking prevalence from the excess mortality due to lung cancer.

Keywords : smoking and mortality; mortality; attributable risk; attributable mortality.

        · abstract in Portuguese     · text in Portuguese     · Portuguese ( pdf )