OBSERVATIONS

The attribution of the mortality demands an empiric focus. In the method here employee that differs of others in certain details but not in their fundamental characteristics, they took five empiric basic decisions at least. In the first place, of the analysis of mortality the cases were excluded in that the death cause was not well specified, and any adjustment was not attempted to compensate the subnotificación of the deaths. In second place, artificial calculations of the structure of the mortality were used for those countries with absence or great shortage of data. In third place, it was used a I average for the adjustments corresponding to the causes of death that could not be analyzed by means of the data of the OPS. In fourth place, an empiric index was developed to make adjustments for the many factors that influence in the risk that the tobacco imposes the populations. In fifth and last place, the calculation of the MAT for the United States was used as fixed value to adjust the calculations obtained with this analysis. All and each one of these decisions influenced in the final result. Also, they also exercised their influence some specific characteristics of the index and of the factors related with the attributable risk in general.

The net effect of the empiric decisions is difficult to value, but the first decision -the one of not to compensate the subnotificación neither to make prorrateos for the not well defined causes - it can be the dominant one and gives for result a too low estimate. The order of magnitude of this last one can be deduced comparing the calculation of the total mortality of Latin America obtained by means of this analysis with another calculation, obtained by means of methods of regression that was used to try to compensate the effect of the subnotificación. If one keeps in mind this difference of the global mortality that is equal approximately to 55%, the number of attributable deaths to the tabaquismo it could end up being of 155 000. It is considered more appropriate the most conservative calculation in 100 000 deaths because it keeps direct relationship with the data that in fact manage the Ministries of Health in the countries of Latin America and the Caribbean. Also, this conservative method allows to all the countries of the region to use a simple and uniform rule for its own calculations. Lastly, this focus allows to make more and more trustworthy calculations of the MAT as it goes having better data of mortality and they are tuned the estimates progressively.

The index of maturity of the tabaquismo in a population is based on the comparison of the different frequencies of lung cancer. Although it is easier to obtain exact information perhaps on this suffering that it has more than enough other dysfunctions, be not the most suitable to calculate the index. The levels of tar influence in the risk of developing this form (him cancer but, seemingly, they don't affect the risk of developing cardiovascular illnesses or obstructive lung illness. Also, the interval that lapses from the food of the consumption of tobacco and the elevation of the mortality for lung cancer cannot be representative of the interval characteristic of other illnesses. To it is added it that the election of the group of 55 to 64 years of age to calculate the index gave for result a too low estimate of the real exhibition from the population to the tabaquismo in a large part of the countries of Latin America and the Caribbean, where this habit has not still reached its maximum incidence.

As it is an empiric index, any justification methodological white doesn't exist to make the transformation in square root. Many transformations can be made, but this last one was chosen because it possesses appropriate properties to the analysis

Specifically, if it takes the square root from the inferior numbers to one a non lineal effect it is obtained: there is an increase from all the inferior numbers to one, but the effect is more marked in the numbers than they approach to zero that in those that approach to the unit. In consequence, the tendency toward increased values is proportionally adult for the countries with low frequencies that for those that show high frequencies. This election modulates, to a certain extent, the conservative nature of the index.

History      Interviews     The cigarrette     Illnesses            How to dump it        News         Links