Hi Joseph,

There are no sphere standards for relapse after cure from acute malnutrition. As a rough guide, in a research programme that collected data from an outpatient programme in one district in Malawi we saw a 1.9% relapse rate within 3 months. That is a relapse from cure from SAM to becoming SAM again. We had some children that became MAM within the 3 month period post-discharge but most spontaneously recovered without intervention within the 3 months. Children were followed up monthly. 

This relapse rate was an average across the district where some locations had access to SFP programmes post discharge and some did not. We didn't see any association between the likelihood of relapse and thee presence of an SFP, however we didnt have a large enough sample size to power that statistical analysis. 

I hope that helps,

Paul

Paul is right. There are no SPHERE standards for relapse after cure. This reflects the fact that data on relapse are rarely collected or reported. We looked hard for standards for the Malawi trial and settled on less than 10% of children experiencing a negative outcome of relapse to program admission criteria or non-accidental and non-violent death during the three month follow-up period following a child being discharge as cured (MUAC > 125 mm, no oedema, clinically well). This standard was taken from:

Ashworth A, Efficacy and effectiveness of community-based treatment of severe malnutrition, Food and Nutrition Bulletin, 2006;27(S):S24–S48

Collins S, Sadler K, Dent N, Khara T, Guerrero S, Myatt M, Saboya M, Walsh A, Key issues in the success of community-based management of severe malnutrition, Food and Nutrition Bulletin, 2006;27(3):S49-S79

which mention a 'three fold criteria of success" fro CMAM programs of case fatality rates below 5%, mean weight gains at or above 5g / kg / day, and relapse / readmission rates below 10%.

This 10% standard for mortality or relapse was accepted by ethical review bodies in Malawi and the USA.

There has been some more work done on relapse since the Malawi trial. Work done in Pakistan by Save the Children and the Centre for Global Child Health (Toronto) found  7.4% of children (discharged with MUAC > 115 mm, > 15% proportional weight gain, no oedema, clinically well) relapsed within six months of discharge:

Dale, N. M., Salim, L., Lenters, L., Sadruddin, S., Myatt, M., & Zlotkin, S. H. (2018). Recovery and relapse from severe acute malnutrition after treatment: a prospective, observational cohort trial in Pakistan. Public Health Nutrition, 16, 1–7

A study in Burkina Faso:

Somassè, Y. E., Dramaix, M., Bahwere, P., & Donnen, P. (2015). Relapses from acute malnutrition and related factors in a community-based management programme in Burkina Faso. Maternal & Child Nutrition, 12(4), 908–917

found 10.5 relapse to SAM (WHZ < -3) after one year (discharge was at WHZ >= -2). This study also investigated factors associated with relapse.

Looking that these studies it seems clear that estimates of relapse are influenced by duration of follow-up so we can expect more relapses to occur with longer follow-up periods, and that higher discharge criteria are associated with lower levels of relapse. It strikes me that relapse < 10% might be a useful standard for followup periods of six months. I think that a follow-up period of longer than six months risks classifying new episodes as relapse. It also seems likely that we will see increased relapse with increased frequency of follow-up.

It seems that we need some standardisation of methods to estimate relapse rates and that may only come if we have a SPHERE standard for relapse. I think that I would plump for something like six months of followup with monthly frequency analysed using Kaplan–Meier methods (i.e. we look at the "survival" of cure amking best use of the data we can get) with a < 10% relapse standard. Others on this forum may have better / more informed ideas.

I hope this is of some use.
 

Here is "Guidance To Improve The Collecting And Reporting Of Data On Relapse In Children Following Treatment In Wasting Programmes. A statement from CORTASAM" that is curremtly being distributed. This may be useful.