For question 1: There are many examples of programs applying 'modified Hearths' in the context of food aid to varying degrees of success. In a non-emergency setting that is food insecure with significant food aid, a modification to the model would be necessary. The program would not likely be able to reinforce the concept of cooking together with locally available foods, but can still be a useful model for identifying positive deviant behaviors around feeding practices (using the food aid), hygiene and care-giving that could be then applied to Hearth sessions in which mothers and children come together, prepare a meal together and reinforce the PD behaviors through various BCC techniques. In an emergency setting, this would be much harder as mothers may not have time or be able to gather together for significant periods of time over several days. Question 2: There are quite a few examples of nutrition programs that have combined PD/Hearth with CMAM. Most of these applications have only been in the last couple of years, but some programs already have data showing a good synergy between the two. Because PD/Hearth is typically attached to GMP or other community monitoring system, moderate and severely malnourished children are identified with WAZ and referred into the program. If the health workers also use MUAC as is increasingly the case, children with SAM can first be referred to CMAM and once they have recovered, can then move into the Hearth. This is addressed to some degree in the addendum to the PD/Hearth manual, and this and many other PD/Hearth resources can be found at http://www.coregroup.org/working_groups/pd_hearth.cfm. For more information on specifics of country applications and data, you can email a list of nutritionists at nutrition@coregroup.org, who typically quite enjoy a good debate that this sort of question is sure to inspire.

Hi there, I found this discussion while I am searching for ressources on field experiences about intervention combining PD/Hearth and CMAM in non conflict setting like where I am working now in Niger. Could any one advise me where I can find guidance on how to calculate sample size of a baseline survey for a combined PD/Hearth and MAM intervention? I think about using a KPC sampling as a venue for household survey and get antropometric measurements of all HH under 5.is this appropriate?

The Positivie Deviance website is http://www.positivedeviance.org/about_pdi/index.html and might have something useful. Randa Wilkinson is one of the Tufts faculty working on the initiative and would know where to direct you. I think her address is randa.wilkinson@tufts.edu but if that doesn't work, you can try the contact on the web site