Infant mortality due to perinatal causes in Brazil: trends, regional patterns and possible interventions

CONTEXT: Brazilian infant and child mortality levels are not compatible with the country's economic potential. In this paper, we provide a description of levels and trends in infant mortality due to perinatal causes and malformations and assess the likely impact of changing intermediate-level determinants, many of which are amenable to direct interventions through the health or related sectors. TYPE OF STUDY: Review paper. METHODS: Two main sources of mortality data were used: indirect mortality estimates based on censuses and surveys, and rates based on registered deaths. The latter were corrected for under-registration. Combination of the two sources of data allowed the estimation of cause-specific mortality rates. Data on current coverage of preventive and curative interventions were mostly obtained from the 1996 Demographic and Health Survey. Other national household surveys and Ministry of Health Statistics were also used. A thorough review of the Brazilian literature on levels, trends and determinants of infant mortality led to the identification of a large number of papers and books. These provided the background for the analyses of risk factors and potential interventions. RESULTS: The indirect infant mortality rate estimate for 1995-97 is of 37.5 deaths per thousand live births, about six times higher than in the lowest mortality countries in the world. Perinatal causes account for 57 % of all infant deaths, and congenital malformations are responsible for 11.2 % of these deaths. Mortality levels are highest in the Northeast and North, and lowest in the South and Southeast; the Center-West falls in between. Since surveys of the North region do not cover rural areas, mortality for this region may be underestimated. CONCLUSIONS: A first priority for the further reduction in infant mortality in Brazil is to improve equality among regions, since the North and Northeast, and particularly rural areas, still show very high death rates. Further reductions in infant mortality will largely depend on decreasing deaths due to perinatal causes. Improvements in the coverage and particularly in the quality of antenatal and delivery care are urgently needed. Another intervention with a potential important impact on infant mortality is the promotion of family planning. Improving birth weight might lead to an 8 % reduction in infant mortality but the efficacy of available interventions is low.