Hi Alexandra It would be interesting to hear from the other commentators, but from what I know.... 1. Only 2-3 nutrition programmes have documented evidence of having reached >85% coverage. 2. None of these were in camp settings. The SPHERE standards are great insofar as they serve as a common benchmark, and to a certain extent, they reflect what community-based nutrition programmes can achieve. In my experience, however, the standards make two assumptions for which there is little documented evidence. The first is that urban programmes should succeed in reaching a higher proportion of cases than rural programmes (70% in the former, 50% in the latter). We know from experience that urban programmes face significant barriers (and high defaulting) which makes this assumption problematic. The second one, which you picked up on, is that camp programmes should succeed in reaching an even higher proportion of cases (90%). From experience, we know that camp settings are not static, people move in and out, caretakers often prioritize IGAs, etc. so to expect coverage to reach 90% as a MINIMUM standard is unrealistic in my opinion. When you look at all of these assumptions, it does seem like the thinking at the time when SPHERE standards were developed, was that physical proximity would be the single key element influencing the coverage of a nutrition programme. We know have ample evidence to suggest that this is not the case, and coverage is influenced by a range of other factors. My thoughts....

I would be curious to hear from colleagues from UNHCR. From my understanding, UNHCR has been doing coverage surveys (CSAS and/or SQUEAC) routinely for some years now. As Saul has said, we are working on a limited set of surveys done in camp settings (IDP and/or refugee) so UNHCR might be able to provide us additional insight and experience on this issue.

Just confirming some of what Saul has said and adding some observations ... (1) I have seen CMAM coverage (by CSAS and SQUEAC) reach 70%, 80%, and even 89%. This has been in a small handful of programs. (2) These were mainly in rural settings. The 89% was from a mixed setting in Bangladesh (i.e. urban, peri-urban, and rural areas covered) with highly local delivery. You can see the SQUEAC assessment report [url=https://dl.dropbox.com/u/73741635/reportSQUEAC.pdf]here[/url]. Regarding SPHERE ... I have no objection to benchmarks. I do, however, have a few concerns about SPHERE and CMAM: (a) The standards should reflect the use of MUAC as the primary admission criteria. The use of W/H is a known "coverage killer". (b) The standards may be most appropriate for emergencies. See (c) below. (c) The standards should have a temporal dimension. This may not be so important in emergencies as we want and expect to have coverage moderately high as quickly as possible. We usually have plenty of resources coupled with vertical logistics to help us achieve this. I am finding that I am now often looking at CMAM programs in different settings (i.e. post-emergency, development) with much lower resource availability and high levels of integration with local health systems. These non-emergency settings present a different set of challenges and opportunities to emergency settings. I think that in these settings we have to accept slow start and incremental scale-up. This means that SPHERE standards are too high in the short to medium term. In these settings our programs should have a long term (sustainability) focus. This means that we can use (e.g.) SQUEAC over several years to improve coverage until it exceeds SPHERE standards. I have (e.g.) just returned from a Coverage Monitoring Network training of SQUEACers looking at an iCMAM program in Northern Kenya that has been running for a few years with low-level NGO support (some logistics support but mostly training). Coverage was about 45% and patchy. It is not fanciful to imagine that this program with better integration with EPI, SFP, GFD, food security, MCH, CHW program, clinical services, &c.; better recruitment and retention of CBVs; better definition of CBV roles; removal of the W/H confusion at clinic sites; and a few outreach sites could (over the next three years) achieve 70% or 80% coverage (against a SPHERE minimum standard of 50%). Here we see SPHERE standards as being too low in the longer term. See (d) below. (d) I worry that the SPHERE standards can cap ambition. It is often necessary to point out that reaching a minimum does not mean you are doing a great job. All it means is that you are not doing a bad job. (e) Reflecting Saul's comments ... Achieving good coverage in urban programming is proving more difficult than in rural settings. The SPHERE standards appear to be badly informed here. (f) Also reflecting Saul's comments ... the term "camp" is a catch-all term. IDP camps are usually very different from refugee camps. Some camps are well run and some camps are so badly run that one is tempted to think that "no administration might be better than this administration". Camps can also be very fluid and subject to large population changes over short time-scales. Having said all that ... it should be possible to achieve high coverage in a well-run camp with CMAM delivered by a capable and committed implementing partner with good integration with other services, frequent screening, and screening at reception. I do doubt, however, that coverage much above 90% is achievable. It should be noted that SPHERE lacks internal consistency here. A defaulter is, by definition, someone who we know needs treatment but is not receiving treatment (if this were untrue then they would be "discharged cured"). This means that defaulting is a direct coverage failure (it is also an indirect coverage failure). With 15% defaulting we cannot have coverage > 85% by simple arithmetic. I think that the SPHERE standards relating to CMAM coverage need overhauling. I hope this helps.