That is almost a direct quote from page 23 of the April 2006 SMART Methodology document (this appears to be the current version according to the SMART website). I am not sure that the SMART manual deserves to be treated as holy writ. It is, perhaps, a little insulting to instruct us to "refer [to] the 2008 SMART Manual" when it is clear that many (most?) of us have referred to SMART documents, attended SMART training, applied the SMART methodology, used the SMART software &c. We have done all that and find that we need something more than is provided at present. Briefly addressing the criticisms of MUAC : Age adjustment : This has been tried (along with height adjustment) and has been shown to reduce the predictive power for near-term mortality of the indicator (see below for a brief discussion of why this is important). MUAC without adjustment actually works better than MUAC with adjustment. Universality and cut-points : The exact weight cut-point for (e.g.) WHZ = -2 for a 85 cm tall boy will vary depending on the reference / standard used. There are many references / standards that could be used. Most familiar to us are NCHS-1977 and WGS. There are many more. The USA, for example, uses their own reference as do most European countries. W/H depends somewhat on body-shape and body-shape is not universal. For example, W/H using either NCHS-1977 or WGS will select fewer children from high altitude settings (large chests and short limbs) than from Sahel pastoralists (long legs and short bodies) even if the true nutritional status of both populations are identical. So, WHZ = -2 does not mean the same thing in different populations. The situation with W/H is that cut-points are not universally agreed and, even when the cut-points are the same (i.e. we all decide to use the same reference) they mean different things depending on where you use them. If this is what you mean by "universally accepted" then I fail to see the virtue of it. I think that we need to consider the purpose of our interventions. Are they child survival programs? Are they special-kind-of- thinness treating programs? I am of the opinion that they are child survival programs. In this case we need an indicator that selects children at risk of near-term mortality who can be treated with (e.g.) the CTC protocol so that their mortality risk is greatly reduced. When we look at indicators and case-defintions in this way we find that MUAC (uncorrected by age or height) outperforms all other practical indicators and W/H performs worse than all other practical indicators. This is a universal finding. It does not matter who you are or where in the world you are ... your MUAC predicts mortality consistently. This is not true of W/H. The predictive power of W/H differs from place to place and, even when it is pretty good, it is still not as good as MUAC. Now to cut-points ... there is, I believe, universal agreement that MUAC < 110 mm is the minimum SAM case-definition and (less clear-cut) that 110 mm <= MUAC < 120 mm is the minimum MAM definition. If prevalence is low or resources available we can raise these so that (e.g) SAM is MUAC < 115 mm. We are not justified, however in lowering them. It is a sort of universal agreement. Use as a screening tool :Advise now is not to use two-stage screening (MUAC then W/H) in CTC and CMAM programs since this has been shown to create problems of rejected referrals and community disengagement and has been shown many times in many contexts to have a devastating effect on program coverage. The relative merits of MUAC and WFH in case-definitions of acute malnutrition has been the subject of long and acrimonious debate. The SMART view is not the only view nor is it the view best supported by the available scientific evidence. Academic reviews of this topic favour MUAC above W/H. It is difficult to debate this issue since the idea that W/H is a useful indicator for GAM, MAM, or SAM is (following Jeremy Freese) "perhaps more vampirical than empirical - unable to be killed by mere evidence - the hypothesis seem so logically compelling that it becomes easy to presume that it must be true, and to presume that the natural science literature on the hypothesis is an unproblematic avalanche of supporting findings". The scientific literature favouring W/H over MUAC is sparse and consists, to a large part, of assertion untainted by evidence. MUAC is used as an admission criteria (i.e. not just a screening / referral criteria) for SAM in CTC and CMAM programs. This is useful because many primary health centres in developing countries do not have height boards or the staff needed to make an accurate and reliable height measurement (most manuals, including SMART - Figure 5 and Pages 70-71, state that two staff are required. Some, such as guides from FANTA and MSF state that three staff are required), the IMCI syllabus does not cover height measurement or W/H calculations, and height boards are not part of essential clinic equipment packs. It may seem a strange point to have to make but it is, surely, not rational to promote an indicator of acute malnutrition such as W/H that cannot be used in most settings where acute malnutrition is a problem. The CTC model of intervention integrates inpatient therapeutic feeding, outpatient therapeutic feeding, and targeted supplementary feeding. Some CTC programs now use MUAC as the sole admission criteria in all of these program components. The need is for surveys that can be used to estimate need so programs can plan properly. When you have programs that admit on MUAC then you need to include MUAC in surveys. I think that your definition of "decisions" is too limited. It may be that prevalence by W/H informs decisions such as whether to intervene in an area but many other decisions about the size and scope of programming can only be made using MUAC. Also, national programs such as the Ethiopian EOS are MUAC-only programs and need MUAC in surveys for all program decisions. I think that the SMART / ENA software should strive to produce the tools needed to inform programming and avoid trying to present a consensus (i.e. W/H is entirely wholesome and MUAC is only useful for limited applications) when no such consensus exists. Many (most?) of us are using MUAC and would like SMART to help us with this rather than imply that we are weak-minded.
|