Can you please tell me what is the technique to develop cut off points for MUAC?
It depends what you want to do with MUAC.
For an application such as using MUAC in a two stage screening process in which you use MUAC in the first stage and W/H in the second stage then the common practice is to use data from one or more anthropometry surveys. The basic procedure is to apply the W/H criteria creating a "CASE BY WHZ" variable, the MUAC threshold is systematically varied and a "CASE BY MUAC" variable create, screening test indicators are calculated from a two-by-two table of "CASE BY WHZ" and "CASE BY MUAC" variables. You then select the MUAC cut-point that gives you the desired level of sensitivity, specificity, positive predictive value, and negative predictive values. The aim is usually for high sensitivity at heigh specificity. BEWARE : The two-stage screening approach can have negative consequences if MUAC is used for referral and W/H from admission. Rejected referrals can have a strong negative effect on program coverage. This is one reason that CMAM programs tend to rely on MUAC for both referral and admission.
For application such as using MUAC as an admission criteria, case-defining thresholds are bases on near-term mortality. For example, data from a considerable number of cohort studies show that mortality is elevated (i.e. above the 1 / 10,000 / day standard) in children with a MUAC below 125 mm. This finding is consistent in all studies. Reviews of this evidence-base can be found here and here on this site.
The move from 110 mm to 115 mm for SAM was made on mixed grounds. In many settled agrarian populations the MUAC < 110 mm threshold selected a very similar number of children and many of the same children as WHZ < -3 z-scores using the NCHS reference population. When the WGS was adopted it was found that MUAC < 110 mm selected fewer children than WHZ < -3 z-scores and the MUAC threshold was raised to 115 mm based on an analysis of a large number of anthropometry survey datasets. The decision was not based on mortality risk (as was the 110 mm threshold) although mortality risk does increase rapidly when MUAC drops below 115 mm.
I hope this helps.