health care workers as well as among families [15]. The belief that we cannot influence what happens becomes a destructive coping strategy, for the individual as well as for societal development. Recent development What is being done to improve the situation of the world's newborn children? While the overall development should be identified as slow, with only 34 % of countries on track to meet MDG 4, there have in the past five years been signs of an accelerating progress toward reducing the neonatal mortality rate [6]. Success, however, is unevenly distributed across the globe, with the largest decline in Latin America since 1990. Worst affected are some of the African countries, which have seen an increase (!) of neonatal mortality over the same period [6]. A strong contributing factor to this is the HIV/AIDS situation in sub-Saharan Africa, and the implications it might have on already inadequate health systems. Notwithstanding, some positive examples are showing what is possible. For example, some Latin American countries managed to reduce the number of deaths during the first month of life by over 50 %. In India, on the other hand – which accounts for the largest proportion of neonatal deaths – in absolute numbers there has been a decrease of only 11 %. This reduction is largely explained by the decline of tetanus cases. Globally since 1990, tetanus, as a cause of neonatal mortality, has been reduced most significantly by 50 % [7]. This has mainly 16 taken place through tetanus vaccinations of pregnant women. However, the challenge of reducing neonatal mortality cannot be solved through ad hoc projects. To increase the survival of the newborns and assure an even distribution of improvements, long-term strategies are needed that integrate into existing maternal and child health care programs. Possible interventions What can be done to reduce the neonatal mortality rate? Studies show that 41-72 % of neonatal deaths could be prevented if the actions we now know to be effective would be implemented on a large scale [12, 16]. Initiatives in the field are taken and implemented by governments and/or NGOs: among others, Save The Children U.S., with contributions from the Bill and Melinda Gates Foundation, has developed an action plan to improve neonatal health [17]. Marsh et al have identified five key priority areas for this work [18]: 1. use of routine maternal and newborn care and good-quality services 2. response to maternal danger signs 3. response to the non-breathing newborn 4. care for the low birth weight baby 5. response to newborn danger signs, particularly those of infection These interventions can be implemented with simple and cost-effective means. For example, in an acclaimed article in the Lancet in 2004, Manandhar and colleagues reported on a study conducted in Nepal, which, by simple means, succeeded in reducing the risk of neonatal death by 30 % (OR 0.70) over a two year period [19]. The intervention was based on participation in which women gathered in groups and themselves identified perinatal health problems that existed in their own environment. By setting up local networks and highlighting issues around childbirth and care of the newborn, a context was created in which information and knowledge could be disseminated in an effective manner, which in turn resulted in positive changes in behavior. The large decline in neonatal mortality that, in this way, was accomplished within a short time indicates that it is possible to make great efforts to improve neonatal health with relatively simple means and where community based interventions have been encouraged [20]. Recently, two studies, one from India and one from Bangladesh, have applied the same method of knowledge implementation to improve neonatal health and found diverging results [21, 22]. While the intervention was successful in the Indian states of Jhakarand and Orissa [22], it failed to show significant improvement of neonatal mortality rate (NMR) in Bangladesh [21]. 17 Contextual factors have been put forward as an explanation of the different outcome, something that remains to be explored. More research is needed If the required means should be implemented, there needs to be a solid evidence base on what works and what is cost-effective. Bhutta and colleagues, in a review of studies [10] made in the field of neonatal mortality, found that the amount of research is proportionately scarce considering that it is an area with so much influence over people's daily lives. Only 10 intervention studies in specific health systems were available for review, and there was a remarkable lack of large-scale trials evaluating the effectiveness of various interventions. Interventions to improve neonatal health are, to a great extent, about changing habits and behaviors. In order to change there is a need to identify barriers and obstacles for survival. At the same time, there is consensus on which health interventions work, thereby creating the important question of how best to implement such existing knowledge [12]. Indeed, how do we translate the existing knowledge about how neonatal survival will be improved into a sustainable working practice? [23] One way, which Manandhar and colleagues demonstrated in their study in Nepal [19], is to involve more parts of society, thereby strengthening the