Menu ENN Search
Language: English Français

Plausable MUAC range for BMI calculation of Women in Reproductive Age Group [15 - 49]

This question was posted the Assessment forum area and has 2 replies. You can also reply via email – be sure to leave the subject unchanged.

» Post a reply

Anonymous 377

Program Manager

Normal user

29 Aug 2010, 21:00

I have recently collected data on weight, height, age, muac and pregnancy status of women in RH age group [15 -49]. Besides the effort made to make correction during data collection, still some of the height recorded on the questionnaire was unlikely for a person with that age range, which we found during double entry.

My question is, is there any plausible range of MUAC for women in this age category to be excluded from the analysis? For example < ---- cm is unlikely for a15 years woman etc.

Thank you!

Tamsin Walters

en-net moderator

Forum moderator

31 Aug 2010, 11:03

Please see below an amendment to the question above. The questionner would like advice on 2 points:

I have recently collected data on weight, height, age, muac and pregnancy status of women in RH age group [15 -49]. Besides the effort made to make correction during data collection, still some of the height recorded on the questionnaire was unlikely for a person with that age range, which we found during double entry.

My questions are

1. Is there any plausible range of MUAC for women in this age category to be excluded from the analysis? For example < ---- cm is unlikely for a15 years woman etc.

2. Is there any plausable range of height for women in this age category to be excluded from the analysis?

Many thanks.

Mark Myatt

Consultant Epideomiologist

Frequent user

1 Sep 2010, 13:04

Something similar has been discussed elsewhere on this forum:

http://www.en-net.org.uk/question/233.aspx

You might want to censor using thresholds derived from the survey data. This approach uses the survey data. Using the survey data, calculate the mean and SD and use these to create your thresholds. For example, if you have a mean MUAC of 24.3 cm with and SD of 1.9 mm then your thresholds might be 23.4 - (5 * 1.9) = 13.9 cm and your upper threshold might be 23.4 + (5 * 1.9) = 32.9.

You could also use the reference approach ...

For height, you can look to the WHO 2007 growth reference which is described at:

http://www.who.int/growthref/en/

The H/A reference is at:

http://www.who.int/growthref/who2007_height_for_age/en/index.html

Looking the row for 180 month (15 years) at the table:

http://www.who.int/growthref/hfa_girls_z_WHO2007_exp.txt

We have a median of 161.6692 with an SD of 6.879

You can choose a threshold based on z-scores. This may be -5 z-scores or -6 z-scores. In this table the SD is equivalent to one z-scores so -5 z-scores is:

161.6692 - (5 * 6.879) = 127.2742

and -6 z-scores is:

161.6692 - (6 * 6.879) = 120.3952

You should think carefully about the thresholds since you may have a great deal of stunting in early years and during adolescence.

Also you need an upper limit. This might be +4 z-scores. So ... with a range of -6 to +4 z-scores you have 120.4 cm to 180.2 cm.

MUAC : As far as I know, there is no internationally agreed reference for older children and adults. This has been discussed elsewhere on this forum:

http://www.en-net.org.uk/question/130.aspx

Local references are sometimes available. Data that can be used as a reference may be available in the form of national health surveys. For example, the report for the "Health Survey for England 1998" presents tables for MUAC/A, associated standard errors, and selected percentiles. The median MUAC for a 15 year old girl is reported as 25.6 and the 15% percentile is reported as 22.3 cm. The difference between these (i.e. 25.6 - 22.3 = 3.3) should be a little over one SD. Rounding this down to 3 gives (e.g.) for -3 z-scores:

25.6 - (3 * 3) = 16.6 cm

We might choose -4 z-scores as an extreme value. This would give a lower threshold of:

25.6 - (4 * 3) = 13.6 cm

Which seems extreme to me.

I hope this is of some use.

Back to top

» Post a reply