Big and important questions! One of the problems with W/H based indices (including BMI) is that they are strongly influenced by body-shape as measured by the sitting to standing height ratio (SSR) or similar measures such as the ratio of trunk to limb length. Pastoralists in warm climates and at low altitudes are known to have low SSR (i.e. short trunks and long limbs). Low SSR leads to lower means W/H and higher prevalence of GAM in pastoralist populations. Since SSR varies widely between ethnic groups and locations the relationship between W/H and mortality will also vary widely between ethnic groups and locations (this has been observed and reported in the scientific literature). The relationship between MUAC and mortality does not vary much between ethnic groups and locations (this has been observed and reported in the scientific literature). Incidentally, at the other end of the distribution we have mountain dwellers who tend to have short limbs and broad chest - W/H makes these populations appear "overweight" or "obese". I used "lower" and "higher" above because to use "underestimate" and "overestimate" requires a "gold standard" of nutritional status. It is usual to consider W/H as the "gold standard" but there is no evidence to support this practise - It is just a nasty habit we seemed to have picked up some time in the 1980s. The terms "nutritional status" and "anthropometric status" are often used interchangeably. Nutritional status refers to the internal state of an individual as it relates to the availability and utilisation of nutrients at the cellular level. This state cannot be observed directly so observable indicators are used instead. There are a range of observable indicators (biochemical, clinical, and anthropometric) of nutritional status, none of which taken alone or in combination are capable of providing a full picture of an individual's nutritional status. There is, therefore, no single "gold-standard" indicator of nutritional status. Nutritional status can be usefully defined at the individual, as opposed to the cellular, level as the ratio of nutrient reserves (muscle and fat) to the nutrient requirements of organs (brain, liver, heart, kidneys, lungs, &c.). It is generally recognised that muscle plays a special role as a nutrient reserve during infection and that infection is a major etiological factor in acute undernutrition. W/H expresses the relationship between weight and height. In children, about 4% of weight is nutrient reserves in muscle. About 96% of weight is, therefore, unrelated to nutrient reserves. Height is almost completely unrelated to the nutrient requirements of organs. W/H measures, therefore, the ratio of something that does not strongly reflect nutrient reserves to something that is almost completely unrelated to nutrient requirements. MUAC, however, is directly related to muscle mass and is, therefore, a direct measure of nutrient reserves. The limited evidence that is currently available suggests that an index known as the lean-mass ratio (LMR), the ratio of the estimated mass of the limbs to the estimated mass of the trunk, is the best anthropometric indicator of nutritional status. The available evidence suggests that MUAC uncorrected for age or height is a better indicator of nutritional status than all other practical indicators and that W/H is not associated with LMR and is the worst practical indicator of nutritional status. An alternative to examining the association between an anthropometric indicator and nutritional status is to examine the prognostic or predictive value (i.e. of predicting death) of various indicators. This approach makes a great deal of sense if you consider that we are running child survival programs. When this has been done, W/H has been consistently shown to be least effective predictor of mortality and that, at high specificities, MUAC is superior to both height-for-age and weight-for-age. In terms of indicators that are practical to collect in developmental and emergency settings, MUAC has the best claim to be a practicable "gold standard" of nutritional status. If you accept these arguments then the "gold standard" to use is MUAC. This means that in all populations (not just in pastoralist populations) MUAC should be the preferred indicator. Now we have decided on a "gold standard" we can see that W/H overestimates prevalence in pastoralist populations. NOTE : W/H will underestimate prevalence in (e.g.) mountain dwellers and cold climate populations. Now for something controversial ... The question of thresholds is interesting. My view is that the thresholds are about GAM and were established for GAM. With MUAC we measure GAM but with W/H we mistakenly measure something else or something confused / confounded. With MUAC we have a more useful indicator for GAM and, I think, we can use the existing thresholds. To answer the big question in the title of this thread "Why prevalence using MUAC is not useful as trigger level for humanitarian response?" is difficult. I think that MUAC is the most useful indicator. I think the real questions should "How on Earth did we ever decide that W/H should be used for this purpose?" and "How can we stop W/H being used?".

Dear Mark, Thank you very much for your detailed reply. Let me ask follow up question for clarification regarding what i mean agreed threshold for MUAC. For example using WHO has set cut-off points to define the magnitude of the situation. These cut-off points are GAM of WFH <5% acceptable, 5-10% poor, 10-14 serious, >15% critical. How about for MUAC? What does it mean when MUAC prevalence is 5% or 8% or 11%? When do we say it is acceptable, poor, serious or critical? Thanks

These (i.e. 5%, 10%, 15%) are nice round numbers. I do not think it was ever intended that they be linked to specific indicators (e.g. they have not been changed by the ongoing shift from NCHS to WGS references) and meant to refer prevalence of a morbid condition known as "acute malnutrition". I am away from my office at the moment and can't shuffle through my old library boxes of articles but I think you might find that the same thresholds were used when we used W/A, MUAC, and MUAC/H case-definitions in the pre-W/H era. I will be in the field for another month or so ...so ... can anyone check this out? I think that the same thresholds can be used for GAM prevalence by MUAC using the MUAC < 125 mm case-definition. I would be very interested to hear what others think about this issue.

Dear Anonymous 81 Comment to your last message Nobody knows how the WHO thresholds based on WFH you quote in your mail were obtained. They are indeed quoted in the WHO 2000 document on nutrition and emergency. See: http://whqlibdoc.who.int/publications/2000/9241545208.pdf p 75 Fig 12. But no clear rationale is given in the document (nor in other WHO documents) for these thresholds. In the same way as nobody will give you a clear rationale for these WFH cut offs, I am afraid no one will give you clear indications on how to take decisions based on MUAC. Although someone may have used some criteria based on MUAC somewhere with some success, as happened with WFH indicators. In any case, it would be a mistake to assume these thresholds are written on stone. And it would be inappropriate to take a decision on what intervention to set up (ie blanket vs targeted) based only on anthropometric criteria. The WHO table mentions "aggravating factors", without giving much detail. Indeed, factors related to the general food situation (eg, time to harvest, quality of the rations, access to food) may lead to very different outcomes for the same levels of malnutrition assessed by anthropometry. Taking the broader picture is more important for decision making than focusing on anthropometry (by whichever criteria). In my opinion, it would help to have these aggravating factors described in more detail in future versions of this table. Another limitation of this decision table is to assess food security in terms of energy availability only. There may be a need for a blanket supplementation if a key nutrient (eg niacin or thiamine or zinc or whatever) is missing in the diet, even if the ration is adequate in terms of energy and if you are below critical levels in terms of WFH (or MUAC). Comment to your first message You ask if there is ongoing research on respective outcomes of low MUAC and WFH. This has been extensively looked at in the 80's, and these old studies led to the conclusion that MUAC was superior to WFH to assess the risk of dying. You have a good review of the issue in: Myatt M, Khara T, Collins S. A review of methods to detect cases of severely malnourished children in the community for their admission into community based therapeutic care programs Food Nutr Bull. 2006 Sep;27(3 Suppl):S7-23. available at: http://www.who.int/nutrition/publications/severemalnutrition/FNB_0379-5721_A_review.pdf It would be impossible nowadays to repeat these studies, as you need to leave malnourished children untreated to get relevant information, which is impossible to do now that we have effective community based treatments for malnourished children. You can only follow up children who have been treated, which will distort the risk assessment. I am afraid this is not really relevant and you have to rely on old studies.

Dear Andre Briend, Thank you for your detail explanation. Yes it is true to consider the "aggravating factors" for further understanding of the situation and also for appropriate response. The probelm is everyone is addicted to the WHO tresholds. most of the decision makings are influenced by these magic treshold figures. Thanks

I am not convinced that "everyone is addicted to the WHO thresholds. Most of the decision makings are influenced by these magic threshold figures". I am writing this from Bangladesh. Reputable surveys (e.g. DHS) from here have the prevalence of wasting (defined as WHZ < -2) as being consistently above 20% but we don't talk every year about the famine in Bangladesh. In Somali Region of Ethiopia (e.g.) prevalence (again by WHZ < -2) is regularly reported as being above 20% but we don't have DEC appeals and start airlifting in food every time that happens. I think that these thresholds are very often ignored. At the other end of the thresholds ... In the southern African famine of 2002 / 2003 (this is the one where Zambia, a supposedly famine-afflicted nation, refused GM maize) we had a full-blown regional intervention on the grounds of a handful of dubious quality household economy assessments and vulnerability assessments from a small number of settings. I was working with VALID doing CTC development work at this time in Malawi during a "national nutritional emergency" and we saw wasting prevalence (again by WHZ < -2) rocket to an astonishing 4.5% in one district. I was also working for ITI developing a survey method for trachoma in Malawi in 2002 and an INGO working in the district found wasting to be as high as 6%. I had my teams screen dozens of villages in the same district using MUAC and we found a handful of cases. There must be reasons for this. There are "aggravating factors" and there are "political factors". I think there is also the problem that as long as we continue to use W/H we have a prevalence estimator that has no clear or consistent meaning.