[Behavior of mortality indicators in small geographic areas: relative standardized mortality ratio and standardized...

One of the most commonly used indicators in mortality studies is the Standardized Mortality Ratio (SMR) being the Relative Standardized Mortality Ratio (RSMR) an indicator of the relative magnitude of the SMR for a specific cause of death. Along with these indicators, which must be calculated by using the population at risk, there is the Standardized Proportional Mortality Ratio (SPMR), an indicator which measures the excess mortality referring to the proportional mortality established for a standard population. This paper reviews the theoretic conditions described in literature as being necessary to establish a probability interval for the RSMR centered on the SPMR as well as the correlations between both indicators and the use of the SPMR as a predictor of the RSMR. The results obtained clearly show how difficult it is for these theoretic conditions to exist. However, they do suggest that the SPMR be used as a predictor of RSMR.