In this month's Bulletin
Child deaths still falling, but more slowly
About 10.5 million children under five died last year throughout the world - about 36% in WHO's Africa Region, 33% in South-East Asia, 14% in the Eastern Mediterranean, 10% in the Western Pacific, 4% in the Americas and 2% in Europe. The risk of a child not reaching his or her fifth birthday is nearly eight times greater in Africa than in Europe. The good news is that over the past decade the number of under-five-year-olds dying, worldwide, has fallen by 2.2 million, or 17.5%. And over the past half-century the risk of a child dying before five has fallen from 25% to 7%. These estimates come from an analysis of mortality trends by Ahmad et al. (pp. 1175-1191). They show, however, that the decline in child mortality has slowed down in all WHO Regions over the past two decades. Only seven countries, though - Botswana, Namibia, Niger, Zambia and Zimbabwe, in Africa; the Democratic People's Republic of Korea, in South-East Asia; and Papua New Guinea, in the Western Pacific - have experienced increases in child mortality. The slowing pace of falling mortality is particularly disturbing in places, such as Africa, where life expectancy is low and prospects for economic development generally slim. The authors end on a note of cautious optimism, suggesting that globally today's child mortality figures could be halved over the next three decades if current trends continue and - an even bigger "if" - if major threats to health, such as HIV/AIDS and malaria, do not increase dramatically.
Not one but many magic bullets spur child survival
Now in its 16th year, the Demographic and Health Surveys (DHS) programme, which is supported by the United States Agency for International Development (USAID), has conducted more than 100 national household surveys in more than 60 developing countries to glean data on the population and on the health and nutrition of women and children in these countries. In an attempt to find out which health or other factors might be responsible for the declining child mortality rates that have been observed in most countries over the past 50 years, Rutstein (pp. 1256-1270) studied data from 89 DHS surveys conducted in 56 countries. The countries were chosen because they had hosted at least two DHS surveys at different periods, thereby allowing identification of health-related factors that changed over time and might therefore account for the trend in child mortality. The factors most strongly linked to declining child mortality turned out to be improved nutritional status and environmental conditions (better water supply, sanitation and housing). Next came greater availability of medical care during pregnancy, at birth and for children with diarrhoea, followed by the availability of electricity and education of mothers. The author, however, offers a caveat: since the study showed that not one but rather a configuration of several factors is responsible for declining child mortality, placing too much emphasis on one or two possible factors (such as immunization or oral rehydration salts) and neglecting others (medical care for children with respiratory illness, for example, or low immunization coverage, especially with measles vaccine) could slow and even reverse declining child mortality. A multifaceted strategy is needed, especially in the face of the escalating threat from HIV/AIDS and resurgent malaria in many parts of the world.
Fewer stunted children but not everywhere
Growth retardation, or stunting, is a direct consequence of poor feeding and is associated with increased child mortality. The proportion of stunted children is a measure of the extent to which a population is meeting its basic needs for food, housing and health care. But getting a handle on how well a country or a region or the world is doing in terms of child nutrition has generally stumbled on the lack of comparable data over time. A new study by de Onis et al. (pp.1222-1233) uses the WHO Global Database on Child Growth and Malnutrition, set up in 1986 in response to this lack, to look at trends in the prevalence of stunting in developing countries. They found that stunting in developing countries has dropped in prevalence from 47% in 1980 to 33% today and if these trends continue will fall to about 29% by 2005. That means that there are some 182 million stunted children in developing countries today, or 40 million fewer than two decades ago. Of the 182 million, 70% live in Asia, 26% in Africa and 4% in Latin America and the Caribbean. A downward trend is seen for all regions, although Africa shows the smallest decline, down from 41% to only 35%. In Eastern Africa, stunting has been on an upcurve since 1980 and today affects 48% of pre-school children, or 22 million, vs 12.9 million 20 years ago. The arrows are pointing in the right direction for most parts of the world, but there is no room for complacency and in some places active concern should be the order of the day.
Oral rehydration therapy, a likely lifesaver
With global child mortality figures tumbling, the search is on to find the reason. Better nutrition seems to be playing a leading role (see the second and third paragraphs, above). Another plausible contender, according to data presented by Victora et al. (pp. 1246-1255), could be the increasingly widespread use of oral rehydration therapy (ORT) to manage diarrhoeal disease. Of course, many factors may be acting concurrently both in the prevention and the causation of death, but two trends over the past decade certainly add up to a strong argument. For one thing, the estimated number of under-five-year-olds dying from diarrhoea fell precipitously - from 3.3 million to 1.5 million, a saving of 1.8 million young lives - while in the same period, deaths from all causes in this age-group fell by 2.2 million. For another, the proportion of diarrhoea episodes managed with ORT in developing countries soared, from 40% to 69%. Case studies from four developing countries show similar trends. 
Fresh Perspectives
The Bulletin welcomes for its Perspectives section views, hypotheses, points for discussion, or commentaries on issues of public health interest. Contributions should be a maximum of 850 words and should not contain reference lists; they will be edited and may be shortened. Please submit texts electronically if possible, by email (bulletin@who.int) or through our web site (www.who.int/bulletin).