Menu ENN Search
Language: English Français

screening of SAM cases

This question was posted the Prevention and treatment of severe acute malnutrition forum area and has 7 replies. You can also reply via email – be sure to leave the subject unchanged.

» Post a reply

Normal user

30 Sep 2013, 06:28

Hi,

I can't reach the target of SAM cases in the area.......... the SAM rate in the locality is high based on last survey done, active case finding is there, in the catchment areas of 2OTPs in locality and well planned but still high numbers of screening done monthly and we found high number of normal cases and MAM, but SAM cases is really limited
how can i justify such situation

Thank you all

Mark Myatt

Consultant Epideomiologist

Frequent user

30 Sep 2013, 18:55

Sometimes you miss a target because you have poor spatial and temporal coverage of screening and of recruitment. Sometimes you miss a target because the target is wrong. You have to decide which is the case in your setting. You can do some SQUEAC activities to check this out.

Note that such "from recent survey" targets are often inaccurate because the typical survey sample size will result in a SAM estimate of poor relative precision, population estimates can be inaccurate (a particular problem in emergencies and when you have a nomadic population), and the prevalence to incidence conversion factor is often just an informed guess (and the assumptions behind it is questionable when incidence and / or episode length changes thought the year). The targets are better than nothing but, in some cases, will not be much better than nothing.

Rogers Wanyama

Emergency Nutrition Specialist

Normal user

30 Sep 2013, 19:21

Hi

You could also consider assessment - intervention lag time i.e the lag time between the "last survey findings" and when the program for SAM management was set up.

Rogers

Hamid Hussien

Nutritionist Tearfund UK

Normal user

30 Sep 2013, 19:41

Hi
you need to review all previous surveys indicators done in catch up area to compare with your survey result if possible
some times due to mistake on survey planning by SMART ( inaccurate design effect or precision)
or you can do some coverage survey to identify real figure of current nutrition status in Target area
Salam

Mark Myatt

Consultant Epideomiologist

Frequent user

1 Oct 2013, 15:54

I hope that this (i.e. insistence that WHZ is the only true measure of SAM - I still find it odd that we elevate a (normative) sign to the status of a disease) is not still a problem. All SMART surveys should collect MUAC data for needs analysis. I believe that the SMART software now supports this. There are, unfortunately, still a few irrational people kicking against including MUAC in SMART surveys.

What may be still be an issue is that the wrong prevalence estimate (i.e. using the WHZ / oedema case-definition) has been used. For MUAC-based programs the MUAC / oedema case-definition should be used. For mixed MUAC / WHZ programs it is tempting to use a combined MUAC / oedema / WHZ case-definition but this will tend to overestimate case-load as screening using WHZ with good spatial and temporal coverage is very difficult to achieve and this will keep coverage down. Some sort of correction (e.g. multiply the MUAC / oedema caseload target by 1.1 to allow for 10% WHZ-only admissions) might make for a more realistic target.

This is a roundabout way of saying "Yes. The case-definition used might be wrong and you should check and recalibrate the target accordingly".

Anonymous 2365

nutritionist

Normal user

1 Oct 2013, 18:24

Thank you all

Abu Ahammad abdullah

Normal user

17 Oct 2013, 03:42

Hi Mark,

Is there any basis or establish guideline for the above mentioned correction: Multiply the MUAC/Oedema caseload target by 1.1%. Or, this is also another informed guess.

In addition, In my three years hands on experience with SAM cases, I found around 80-90% children identified with SAM by MUAC is about 6-30 month of age. 1-2 cases we found with some other chronic medical complications e.g. hernia problem, some has mental abnormalities etc. Is this MUAC is valid for elder children? although we know MUAC is golden standard for identifying acute malnutrition. One programme research in Cambodia by Unicef also validate my suspicion.

It would kind enough, if you help me in this regards.

Mark Myatt

Consultant Epideomiologist

Frequent user

19 Oct 2013, 11:42

A few things ...

(1) MUAC is not a gold standard for MAM / SAM. It is just better (in terms of predicting near term mortality that can be avoided with nutritional rehabilitation) than other indicators. It is the best performing indicator from a pretty weak bunch.

(2) It is often said that MUAC "preferentially selects" younger children. This simplifies the situation. The use of a fixed MUAC threshold in the 6-59 month age group tends to select children at a fixed level of mortality risk. For example, a 36 month old child with a MUAC = 110 mm has similar risk to an 18 month old child with a MUAC = 110 mm. The case-defining thresholds are based on risk. Mortality risk decreases rapidly with age so an indicator that picks up more younger than older children is behaving as we want it to. This is in strong contrast to (e.g.) W/H that selects children using a threshold taken from a restricted "ideal" population. In this case the indicator selects on comparative thinness and, in some populations, comparative thinness is associated with good outcomes. Low WFH is associated with increased mortality but this is probably due to the loss of muscle mass reflected in loss of weight reflected in low W/H in some children. In this case low W/H selects low MUAC kids in a complicated and expensive way. If there is an issue about MUAC excluding children then the best approach is to increase the MUAC threshold rather than adopt another indicator. In short ... MUAC will tend to select (on average) younger children than W/H but is still valid in older children.

(3) The "multiply by 1.1" rule-of-thumb is based on experience and applies to programs that use MUAC as the primary indicator but will admit on W/H for children that attend a clinic and have W/H assessed. The "1.1" will vary from setting to setting. If you have W/H assessed for all children attending all clinics then it may need to be larger than that. If you have a "MUAC only" program which may accept the occasional W/H referral as a discretionary admission than the "1.1" might be close to "1". Better, I think, to have all children at all clinics assessed using MUAC (this will improve coverage) than to have a few large clinics screen using W/H at a few clinic sessions. I very rarely work with programs that do anything with W/H other than passive case-detection at district hospitals (most PHCs do not have height boards and so cannot use W/H) and this informs the "1.1" figure.

(4) Almost everything in science is an informed guess. The qualifier "informed" is what make it different from "just guessing". We can replace "informed" with "evidence-based" if you'd like. Informed guesses are always better than uninformed guesses.

I hope this helps.

Back to top

» Post a reply