Considerable evidence suggests that malnutrition affects human performance, health and survival, including physical growth, morbidity, mortality, cognitive development, reproduction, physical work capacity and risks for several adult-onset chronic diseases. In recent decades, development agencies and governments have emphasized selective interventions to improve health and nutritional status, such as immunizations, oral rehydration, antibiotics and micronutrients, with child survival as a major motivation and justification. Although the efficacy of these approaches for improving child survival has been amenable to study, providing some of the rationale for using these approaches, it has not been possible to test directly the effects of improvements in general malnutrition. The present study quantified the effects of changes in general malnutrition, as measured by child weight-for-age (WA), on changes in child survival in 59 developing countries, using aggregate, longitudinal data at national and subnational levels from 1966 to 1996. Mixed model analysis (in SAS) was used, to take advantage of the multilevel and longitudinal nature of these data sets. Changes in WA have a statistically significant effect on changes in child mortality, independent of socioeconomic and policy changes represented by the secular trend. The secular trend in mortality began earlier and leveled off at higher mortality rates in populations with a higher prevalence of malnutrition. Gaps in coverage of selective interventions are more likely and more serious in the more malnourished populations. Continued reduction in mortality will require improved targeting of selective interventions and general nutritional improvement to the most marginal populations.