Revisiting crude and standardized mortality ratios : which ones to assess health systems performance?

Abstract

Background:

Standardized Mortality Ratios-SMR are often used to assess performance of health systems. Age adjusted SMR are rates one would expect to observe in the study population if it had the same age distribution as the standard year itself. SRM are thus artificial data, appropriate to compare groups (countries, for example) or time trends, isolating this analysis from age effects and its variability over time. Each main health data source (WHO, OECD, Eurostat) adopts a different year as population standard.

But are mortality SMR better than crude ratios– CMR, to monitor the evolution of diseases, to infer on the performance of health systems or to compare countries?