Menu ENN Search
Language: English Français

SFPs - what is wrong with them and can we fix it?

This question was posted the Prevention and treatment of moderate acute malnutrition forum area and has 8 replies. You can also reply via email – be sure to leave the subject unchanged.

» Post a reply

Anonymous 432


Normal user

12 Aug 2010, 08:43

Latest after Carlos Navarro-Colorado's (et al) paper on the effectiveness of SFPs there is a tendency to turn away from them as the message seems to be misinterpreted. Do we really have to abolish them or do we have to be more intentional and more cautious about the way SFPs are desigend and implemented?

Anonymous 184

Normal user

12 Aug 2010, 10:48

May be it is important that Carlos and ENN provide a concise summary of the key messages of this study to reduce the level of misintepretation of a clearly written article.

Tamsin Walters

en-net moderator

Forum moderator

13 Aug 2010, 14:57

From Jeremy Shoham, ENN:

Dear Questionner

It is good to be able to address this question on the en-net for a. Carlos and his co-authors have on several occasions felt a little uneasy about the way the findings of the study (Measuring the effectiveness of supplementary feeding programmes in emergencies) have been represented in subsequent reports and statements. The study was based on 82 programme data sets of 16 agencies conducting emergency SFPs. Key findings were that <40% of programmes achieved SPHERE standards for key indicators and that the main reason for this was high defaulting. The study also found that the vast majority of programmes did not collect programme coverage data but that out of those that did the median coverage of moderately malnourished children was 21%. The study then went on to conclude that given weak programme performance (due largely to high defaulting) and low programme coverage impact of programmes at population level were likely to be limited. Another key finding of the study was that project monitoring was extremely weak so that data had to be re-analysed in order to interpret performance. Furthermore, it was not possible to test key potential associations or correlations with contextual factors due to weak reporting systems.

The authors then made a number of recommendations based on these findings. These were as follows;

i) The need to develop a minimum reporting package for SFPs

ii) The need to conduct a quantitative study on causes of defaulting in a number of SFPs in different contexts

iii) The need to test alternative modalities for addressing moderate malnutrition in emergencies at population level

Since publication of the study findings in 2006 and through the HPN in 2008, the ENN and SC UK have received funding to work on all three areas. We have now produced a minimum reporting package which is in the final stages of development and is being piloted in Sudan and Ethiopia. The package contains reporting guidelines and associated software.

We are also in the final stages of a defaulter study which has been conducted in three countries (Chad, Sudan and Kenya). This study has collected data over a nine month period and involves follow up of defaulters and implementation of a standardised questionnaire. Results will be out at the end of the year and it is hoped that these will throw light on the complexity of reasons behind defaulting and therefore how programmes can be better designed to minimise defaulting.

We have also just embarked on a larger study which will examine potential alternative means of addressing moderate malnutrition at population level, e.g. cash transfer, expanded general rations, decentralised SFPs, cash and food. This study is still in it's early stages but will endeavour to compare the cost-effectiveness of different approaches including outcome measures such as population impact.

The author of the question to en-net will also be aware that there are currently a large number of studies comparing the efficacy of different food products in the treatment of MAM. These studies are endeavouring to show better outcomes that those achieved with the 'old' CSB rations.

The work described above has been undertaken precisely because we do not know enough about how to address MAM in every context. The original 82 programme study attempted to be clear that the authors were not advocating for a cessation of emergency SFPs. The conclusion was far more nuanced, i.e. it may be that the traditional emergency SFP approach is not suitable for every context. There are therefore key questions regarding whether we can define the contexts where emergency SFPs are appropriate and those contexts when other approaches may be deemed to be more appropriate. Clearly there is a long way to go in answering these questions. At the moment the only tried and tested approach for addressing MAM in emergencies is through SFPs. While the findings of the 82 SFP programme study are worrying, they will hopefully in the future lead to better designed SFPs as well as other potential treatment and prevention models that can be used in a range of contexts. The message from the study is therefore most definitely not to stop employing emergency SFPs in emergencies but to explore better designed SFPs and other approaches not involving supplementary feeding which are advised and can be used in a range of emergency contexts.

I hope that this addresses some of the concerns raised by the en-net correspondent and please feel free to raise other related questions if you wish.

Jeremy Shoham

Mark Myatt

Consultant Epideomiologist

Frequent user

15 Aug 2010, 10:31

This is an interesting question. It seems that the finding that SFPs following current methods tend to be low impact programs has been translated into a view that we have nothing to offer MAM children. The proper position would, in my opinion, be that we need to work out how to fix SFPs. To be fair to Carlos and his colleagues, the report does call for testing of alternative SFP designs. I know that this has been funded and that decisions about which models of delivery will be tested will be made soon. Let us hope that they choose well.

Anonymous 184

Normal user

16 Aug 2010, 09:14

Dear Tamsin, Jeremy and Mark,

Thanks a lot for taking time to respond to my string of concerns, I fully concur with your response. As somebody had mentioned earlier, I bet a quest for a CMAM/CTC like solution has created both negative and positive anxiety amongsts actors in trying to find the magic bullet solution for MAM. Given that operational literature on SFP/MAM is either scarce or whatever is availble is limited in providing the "right quick fix" ammunition to the problem, then you will agree with me that operationally the chances of using the available information to provide "guess" guidance in the response process is almost the only immediate option.

Is there broader channel of communication that can be used to clarify the outcome of the study and highlight the exciting work in progress on defaulters, MRP etc? I don't know. I know that the en-net is one of the very useful and widely used informal communcation channels that strongly influences the nutrition world, in addition to this though what would be the other communication channel that can be used to ensure that clarity/discussions on such an important subject trickle down to decision makers at both HQ and operational level?

Thanks again


Mark Myatt

Consultant Epideomiologist

Frequent user

17 Aug 2010, 12:50


It is an interesting comment about anxiety amongst actors regarding a CTC like project for SFP since the original CTC work split agencies and others into "for" and "against" camps. Many agencies that had a significant investment in centre-based TFPs joined the "against" camp. I am not sure that many agencies have a significantly investment in SFPs for MAM and this may not be as big a problem..

I think that part of the problem is that we are not sure what we mean by "SFP". I did some work for the WHO on MAM case-definition and did a literature search on admission criteria for SFPs. What came up was that SFPs admit on a range of anthropometric (i.e. W/H, H/A, W/A, MUAC, and combinations) and various and diverse social and economic criteria. Some aim at primary prevention, some at secondary prevention (e.g. detect growth faltering in GMP and use SFP to prevent stunting), some just treat (e.g. MAM cases). Some are "blanket". Some are "targeted". I even came across a few "blanket targeted" SFPs (whatever they might be - Any ideas?). Some give RUTF. Some give RUSF. Some give CSB. Some give not very much at all. Some give money (I think that these cannot properly be called supplementary FEEDING programs since banknotes and coins have negligible nutritional value ... they are social security programs). Some treat 6-36 months. Some treat 6-59 months. and so on. When we look for guidelines and field reports we get a mass of conflicting information.

I worry that current efforts are unfocussed. For example, the project that Jeremy wrote about includes expanded GFD, cash / voucher schemes, cash and food, blanket SFP and decentralised SFP). This reflects, I think, the confusion surrounding SFP. Perhaps a broad approach will bear fruit. We do need to look carefully at all of these program modes. I'd prefer, however, a more focussed "vertical" approach.

I think the first thing that we need to do is define "SFP" is some useful and specific sense. It might even be useful to make up a new term (any suggestions?). To start the ball rolling I think we can limit what we mean by SFP to being about "finding, recruiting, and treating cases of moderate wasting in children aged between 6 months and either 3 or 5 years" we can elaborate this with "high coverage", "low defaulting", "high cure rate", "low relapse rates", &c. It's about treatment of MAM and prevention of SAM cast as a child survival program.

Getting this type of SFP right could be a lot of work. We'll need both "dump the surplus" top-down emergency SFP and more resilient programs that would better suit places like Bangladesh in which centralised programming with high logistics overheads can fail in the face of flooding and infrastructure loss.

Torben Bruhn

Health & Nutrition Expert / EC (ECHO)

Normal user

18 Aug 2010, 07:08

Dear all.
An "easy way out", a quick fix if you like, which actually might make some sense (in my head anyway), would be to stop talking about moderate and severe acute malnutrtion all together, and simply settle for acute malnutrition.
All acutely malnourished children should naturally receive treatment according to their disease/condition, vis-a-vis the CTC/CMAM/IMAM protocols, and could at OTP and/or health facility level receive what they need of support. This way we would drop SFP's altogether, introduce already recognized, tested and approved RUTF's for the treatment of ALL malnourished children.
Surely this would mean a significant increase in caseload at OTP sites and health facilities, yet the simplification of approach, i.e. one approach for all acute malnutrition, the "incorporation" of MAM children into OTP and/or health facilities, would/should assure a better outcome for these children.
And at the end of the day, wasting time, money and energy on badly functioning SFP's, with a significant number of MAM children on SFP's slipping into severe malnutrition and in need of a TFP, is something even donors can relate to.
Anyway, these just a few stray thoughts on a Wednesday morning.
All the best - Torben

Dr Basil Kransdorff - e'Pap Technologies -

CEO - e'Pap Technologies

Normal user

18 Aug 2010, 13:52

Torben - your stray thoughts have hit bulls eye in my view. If one looks at the Nutrition Industry in relation to addressing malnutrition - one needs to become a linguist with a pHd to wade through the endless actonyms that have been created for what purpose I wonder. ACT ARI ART CMAMIMCI MAM MUAC NNP OTP OVC PMTCT RUSF RUTF SAM SST TFC WFA WFH WHM they go on and on. I think the Nutrition Industry has lost the plot and that many of these acronysms are invented to hide and mask silly products and programs that cannot address the bodies needs to help make it nutrient replete. The rules need to change to creating a single objective - NUTRIENT REPLETENESS that understands - for a human being to be functional - they must be nutrient replete in all the micro and macro nutrients. All nutritional conditions from the most extreme of nutrient deficincies are in my opinion a condition of malnourishment. An obese American who is micro nutrient deficient is physiologically dysfunctional as is a malnourished African and should be a target of an effective nutrition interevntion. Part solutions or silly label claims or unfathonable acronyms that have no scientific possibility of addressing the objective of nutrient repleteness are just playing musical chairs that assign all these fancy names to create an illusion that they have value in addresing malnutrition. If they cannot make a human being nutrient replete - they have little value in making a human being physiologically functional. If one evaluates many of the products used to address malnourishment in the context of the objective NUTRIENT REPLETENESS - most if not all will be put where they belong - in the trash can. Maybe we can start to turn the tide on poverty if we change the focus to make the victims of poverty functional human beings that can participate in solutions.

Mark Myatt

Consultant Epideomiologist

Frequent user

18 Aug 2010, 15:26

Torben : I'm not sure about "easy way out" but the idea that we treat acute malnutrition in one program rather than SAM and MAM in separate programs is a very good one. The early CTC documentation (e.g. Steve Collins' HPN paper) proposed a similar model and the early CTC programs usually had well integrated OTP and SFP components. This had the advantage of having a low intensity program to discharge the previously SAM kids into and reduced interface problems between the two components so that (e.g.) we didn't keep non-responding MAM kids in SFP. Early CTCs often had OTP and SFP at the same site on the dame day. This level of integration is often missing in CMAM programs. Last year I evaluate a pretty well done CMAM program in which the major barrier to OTP coverage was that non-responders in SFP have deteriorated to SAM and this had not been detected. The OTP and SFP were run by the same agency but the interface ran in only one direction (OTP -> SFP). You've clearly seen this sort of thing too. I think you are right to raise the issue of caseload at OTP sites. This might be dealt with by good queuing systems &c. but, I guess, we will still need some division between SAM and MAM (or whatever) to keep numbers and waiting times manageable and to make best use of nurse / doctor time. A strong interface with minimal paperwork for transfer between program arms is needed as is good monitoring within the SFP arm.

Basil : There are a lot of TLAs (three letter acronyms) in any field of endeavour and emergency nutrition is no exception. Often they are convenient such as when we use MUAC rather then spell it out in full. In this case there is a clear meaning. At other times the acronyms obscure and confuse. This thread is about SFPs. The term SFP has no clear meaning (see above). When a term lacks clear meaning then we need to do something. We need to rail againt some acronyms but to rail against all acronyms strikes me as foolish. I do not think that there is any intent to exclude or mystify in the use of acronyms by the vast majority of people in our field but some terms (e.g. SFP) can be difficult to pin down. Others more competent than I can comment on your concept of "nutrient repleteness". I can give you my take as a jobbing epidemiologist ... it seems to me that "nutritional repleteness" and "physiologically dysfunctional" as properties of an individual might be difficult and expensive to measure and monitor and anything that is difficult and expensive to measure impacts negatively upon program coverage and increases costs. That's why we use crude anthropometry for case-finding, admission, monitoring, and discharge in many of our programs. As for "nutritional repleteness" as a property of the products / diets that we use ... products such as F75 and F100 (milk, paste, compressed "biscuit", &c.) have been engineered and tested for clinical efficacy. Having said that I remember an article published a few years ago that compared fortified cereal-legume blends unfavourably with, I think, dried cat food.

BTW ... I am not familiar with e'Pap and a (quick) literature search did not return any published efficacy trials.

Back to top

» Post a reply