Progress Report as of March 2002


Climate Variability and Human Health Impacts in Mexico

The team obtained several retrospective databases in Mexico with a focus on the period 1980-2000, which contains at least four El Niņo episodes for analysis.

  • Morbidity and mortality of malaria and dengue by state for varying periods. 
  • Climatological data from nearly 5000 stations on a national scale from 1930 to 1990 and a select number of 322 stations for 1973-1998. Climatological norms and vertical profiles of weather parameters are also available. 
  • Marginality indices at the national and state scales. These measure social vulnerability by taking into account seven socioeconomic indicators from the 1990 national census and a 1995 national survey.
  • Entomological data based on malaria and dengue field indices in the health jurisdictions of state governments. They need to be organized into electronic databases.
  • Trends in deforestation and 1970-2000 annual number of forest fires. 

For each disease, the team selected three Mexican states for detailed study based on the epidemiological importance of that disease in each state in the period 1980-2000.

  • Malaria -- Oaxaca, Chiapas, Sinaloa
  • Dengue -- Tamaulipas, Veracruz, Colima

Preliminary results compare calendar years with El Niņo and calendar years without El Niņo. For these purposes, El Niņo years from 1941 to 2001 are designated as follows: 1941, 1972-1973, 1982-1983, 1986-1987, 1991-1992 and 1997-1998. An analysis of mean rainfall in Mexico from 1941-2001 shows that the months with the highest rainfall in years without El Niņo (June - September) tend to experience reduced rainfall in El Niņo years. Rainfall in autumn and winter tends to be higher in El Niņo years.

An analysis of total malaria cases in Mexico for 1986 - 2000 (excluding years 1990, 1991 and 1992, for which information was missing) shows that malaria tends to be higher in El Niņo years (1986-1987 and 1997-1998). Significant differences emerge from January to May and in July, September and October. 

An analysis of total classic dengue cases in Mexico for 1986 - 1999 shows that the greatest incidence occurs in the second half of the calendar year. In this analysis, the mean incidence for classic dengue in El Niņo years is higher in each month of the year, but these differences are not statistically significant. An analysis of dengue hemorrhagic fever in Mexico for 1994 - 1999 shows that its incidence also tends to be highest in the second half of the calendar year. It is difficult to make any inferences about the effect of El Niņo years. 

An analysis of the total number of forest fires and hectares affected in Mexico for the period 1970-2000 shows that forest fires of varying scope take place annually.