Incidence correction factors for moderate and severe acute child malnutrition from two longitudinal cohorts in Mali and Burkina Faso

View this article as a pdf

Research snapshot¹

Accurate estimates of the burden of acute malnutrition (AM) are essential to support policy makers and nutrition programmers in decision making including predicting AM caseloads and preparing sufficient resources for treatment. As AM can be a transient state, incidence estimates, as opposed to prevalence rates, are needed to determine the true annual burden. In the absence of longitudinal data, obtaining incidence estimates is challenging and, as such, AM burden is typically approximated by converting prevalence estimates from cross-sectional surveys to a cumulative AM incidence using an ‘incidence correction factor’, ‘K’.

This study aimed to add to the body of evidence related to incidence correction factors. It estimated ‘K’ factors for severe acute malnutrition (SAM) and moderate acute malnutrition (MAM) based on longitudinal data (monthly data collected for 18 months) from two cluster-randomised control trials conducted between 2014 and 2017 in Burkina Faso and Mali (Innovative Approaches for the Prevention of Childhood Malnutrition-PROMIS study). SAM cases were included in the incidence calculation if they were preceded by one or more SAM-free months. Children who met the MAM criteria while recovering from SAM were not included in the MAM incidence calculation. Data was compared using complete (weight-for-length z-scores, mid-upper arm circumference (MUAC) and oedema) and partial (MUAC, oedema) definitions of SAM and MAM.

‘K’ factors for SAM were 9.4 and 5.7 in Burkina Faso and in Mali, respectively. The ‘K’ factors for MAM were 4.7 in Burkina Faso and 5.1 in Mali. MUAC and oedema based definitions of AM did not lead to different ‘K’ estimates and thus results suggest that ‘K’ can be reliably estimated when only MUAC and oedema based data are available.

When comparing incidence correction estimates to available prevalence data in the two countries, prevalence was found to underestimate the annual burden of SAM by a factor of 7 to 10 and that of MAM by a factor of 6. This study highlights the need for more contextualised incidence data to accurately measure AM burden.

Subscribe freely to receive Field Exchange content to your mailbox or front door.