If there are three measurers of anything there will be three different results. This applies to MUAC, weight, height, WHZ, distance between pupils, leg length, the position of a star, GPS co-ordinates, &c. I am not sure which guideline you are referring to but 5 mm is a large inter-oberver variation. I have been a round the block a few times and have seldom seen variations this large even when comparing mothers' measurements after a few minutes training to those of experienced health workers. Training and motivation are important with all measurements. I suspect that a strong expectation effect may operate. It is often asserted that, in terms of precision and accuracy of measurement, MUAC compares unfavourably with W/H. Evidence supporting such assertions is, however, elusive. There is a lot of assertion, a lot of anecdote, and a lot of hand-waving in front of poorly collected data but no properly conducted experiments (that I know off). Younger children tend to become agitated during weight and height measurement under field conditions. This has a negative impact on the precision and accuracy of height and weight measurements. Weight is strongly influenced by hydration. Giving a child (e.g.) a drink of water will increase their weight. Voiding the bowel or bladder will decrease their weight. Heavy parasitism with Ascaris lumbricoides may bias weight measurements upwards. Oedema biases weight upwards and is a poorly recognised sign. Weight has considerable natural variability in addition to measurement error. Height is also influenced by hydration (dehydrated children have reduced heights) and there is diurnal variation (height decreases during the day). Height has considerable natural variability in addition to measurement error. Few studies have looked at the performance of health workers with MUAC and WHZ and the results are that MUAC has better accuracy and reliability compared with WHZ. WHZ is found as the least accurate and the least reliable measure. Measurement, then, is not so easy. It seems that MUAC is the better measure for all sorts of reasons related to error. So far we have looked at error. We have to consider what is being measured or why we measure it. I think CMAM is a child survival program. MUAC is a measure of mortality risk. It is the best prognostic indicator for mortality / survival than the other common measures (again W/H is the worst). If we think of CMAM as a "nutrition program" treating a peculiar type of thinness (i.e. a statistically defined thinness strongly influenced by body shape and body surface area) then WHZ would be the ideal indicator (if it were not for accuracy and reliability issues). Criticising MUAC for error without looking at the issue of errors in alternative measures (and at what is being measured and why) is not, I think, sensible. The alternatives from MUAC are very far from being a "gold standard". If MUAC is "bad" then it the the "best of a bad lot".

Dear Mark Myatt, thanks for comments on my questions. Pls refer me the guidelines for MUAC monitoring. I really need your opinion about how much percentage of SAM cases can develop complications in OTP?

Dear Suhail, I am not sure which guideline you are referring to. Much of the work on MUAC monitoring (as opposed to discharge) is currently in press. I can post links here when it becomes available. At present I suggest you use MUAC for admission and discharge (e.g. MUAC > 125 mm for two visits) and use weight for monitoring with MUAC alongside so you can gain experience with using MUAC for this purpose. The proportion of SAM cases that develop complications in OTP will depend on severity of SAM at admission. If (e.g.) you can get most cases in with MUAC above 110 mm then you should have very few complications, short lengths of stay, high cure rates, low default rates, and few complications. This will also depend on your within-clinic programming. If you deliver the full protocol including the systemic antimicrobial and, if indicated, and antimalarial then complications should be uncommon. I would be concerned if complications requiring an OTP to inpatient transfer were more than a few percent after the first few months of program operation. There are people here with a lot more day-to-day experience with OTPs than I have and they should be able to advice. I hope this helps.

Okay, this is really great to read your opinion about percentage of complications developed by SAM cases.