Language: English Français

# Analysis of Nutritional Status

This question was posted the Assessment and Surveillance forum area and has 1 replies.

### Mark Myatt

Epidemiologist at Brixton Health

Frequent user

27 Jul 2012, 16:15

Is this all you have? If you have SAM by MUAC then you could have GAM and MAM by MUAC. If this is all you have then you can probably estimate SAM incidence in the screened population as: ``` incidence (cases per year) = number SAM cases found ``` You may want to make this a more explicit rate: ``` incidence (cases / 10,000 / per year) = (number SAM cases found / Number screened) * 10000 ``` If you can reasonably assume that there is no great difference with regard to SAM incidence between the screened and non-screened population then you can apply this incident rate to the whole population: ``` cases per year = rate * (population / 10000) ``` This is cumulative incidence. If we assume that a SAM episode lasts 7.5 months then you can estimate an average prevalence as something like: ``` SAM prevalence = [case per year * (7.5 / 12)] / population ``` This is for SAM. For GAM you might assume that there are ten MAM cases for each SAM case giving: ``` GAM prevalence = SAM prevalence * 11 ``` It is all very "back of an envelope" and very approximate. You could make this more explicit by using intervals to attempt to quantify uncertainty (e.g. episode length could vary between 6 and 9 months, the MAM:SAM ratio could vary between 8 and 12 cases of MAM for each case of SAM). Best to use data from surveys to calibrate the MAM:SAM ratio. It might be better than nothing. Note that "average prevalence" is a slippery concept. Someone should check this answer to see if any of it makes any sense. Any help?