The seasonality of infant deaths due to diarrheal and respiratory diseases in southern Brazil, 1974-1978.

Death certificate information was used to examine seasonal variations in infant deaths attributable to diarrheal and respiratory diseases during a 5-year period (1974-78) in the Brazilian state of Rio Grande do Sol. The state has a subtropical climate with 4 marked seasons in which the temperature ranges from 14 degrees Centigrade to 24 degrees Centigrade. Rainfall is fairly evenly distributed througout the year but varies from year to year. Death registration is relatively high in the area, and deaths are coded according to the 8th revision of the International Classification of Diseases. In this study all diseases classified as 000-009 were considered to be diarrheal deaths, and all those classified as 460-519 were considered to be respiratory deaths. A total of 40,219 infants deaths were recorded during the 5-year period. Monthly mortality rates were computed. The denominator for all monthly rates was the number of infants alive at the midpoint of the 5-year period as estimated from census materials. The same denominator could be used since there was little change in the birth rate during the 5-year period and little seasonal variation in birth rates. 16.2% of the infant deaths occurring during 1974-78 were attributable to diarrheal diseases, and 19.8% were attributable to respiratory diseases. About 1/2 of all the respiratory and diarrheal deaths occurred during the 1st 4 months of life. For example, the diarrheal death rate at months 0-3 was 120/100,000 infants, and by the 12th month the rate declined to 7.5/100,000 infants. 30.5% of the deaths attributed to respiratory diseases occurred during the 1st month of life, and then the rate declined sharply. There were marked and statistically significant seasonal variations in deaths due to both causes. The incidence of diarrheal deaths was almost 4 times higher during the summer months (January-February) than during the winter months (July-October). This correlation was significant at the 0.0008 level. The pattern was reversed for respiratory deaths. The respiratory death rate was twice as high during the winter months June-July as during the summer months of December-Februray. This correlation was significant at the 0.0002 level. There was no significant relationship between rainfall and deaths due to respiratory and diarrheal diseases. Multiple linear regression analysis revealed that temperature explained 68.9% of the variation in diarrheal deaths and 85.9% of the variation in respiratory deaths. Infants born in October through December (the months just preceding the summer months) had a higher risk of diarrheal death than infants born during other months. In contrast, infants born in March through July had a higher risk of death from respiratory disease than infants born during other months. The overall rate of infant deaths due to respiratory disease was 860.4/100,000 infants. The overall rate of infant deaths due to diaarheal diseases was 70.4/100,000 infants. Some caution should be used in interpreting the results of the study. The overall rate for diarrheal deaths appears to be relatively low for the region and may indicate an underreporting of deaths. Furthermore, the cause of death noted on the death certificate was accepted without further checking. Data on the major causes of all infant deaths in also provided.