[A review of original papers that analyze the effects of air pollution on mortality, 1994-1998]

The different studies published and indexed in Medline and in the Spanish Medical Index (IME) from 1984 up to 1998 (IME: 1971-1996) are reviewed. To start with, an assessment was made of the number of publications put out which have to do with the impact of pollution on human health as well as the percentage thereof which analyze the mortality as a health indicator. Afterward, the original works published within the January 1994-June 1998 period were reviewed in detail. Most of the original articles are combined time series studies which analyze, on a single-day basis, the relationship between the levels of pollution and the mortality for non-accidental causes. Other original articles were also summarized based on an approach other than the time series one, highlighting different cohort studies which relate the levels of pollution to the mortality on an individual level and which confirm the findings of prior ecological studies. The pollutants analyzed most often are the different types of particles, although studies are also common regarding the impact of sulfur dioxide, nitrogen dioxide, ozone and carbon monoxide. The method of analysis most used is the Poisson regression, into which different confusion variables bearing an impact thereon such as seasonality and tendency, temperature and relative humidity and day of the year are basically added. In the studies reviewed, the positive significant findings prevail, being consistent especially for particles, which is the pollutant most analyzed. The articles studying other pollutants, although fewer in number, also indicate a significant relationship between pollution levels and the mortality.