Re-thinking "non-response" to wasting treatment: Exploratory analysis from 14 studies

Area of Science:

• Pediatric Nutrition
• Global Health
• Child Malnutrition

Background:

• Children treated for wasting are often classified as 'non-responders' if they don't meet recovery criteria within 12-16 weeks, leading to their categorization as treatment failures.
• This classification overlooks potential variations in growth trajectories among these children, potentially misidentifying those with delayed recovery as definitive treatment failures.

Purpose of the Study:

• To conduct a pooled analysis of 14 studies involving children aged 6-59 months undergoing therapeutic feeding for wasting.
• To explore the growth trajectories of 'non-responders' and critically evaluate the appropriateness of the current 'non-response' definition.
• To differentiate between subgroups of non-responders based on their anthropometric gains during treatment.

Main Methods:

• Pooled data from 14 studies including children classified as recovered or non-responders.
• Defined 'low growth non-responders' (<25th percentile anthropometric gain) and 'high growth non-responders' (≥25th percentile gain) using pooled non-responder data.
• Utilized mixed-effects generalized additive models to plot growth trajectories (MUAC, weight, height) for recovered, high growth, and low growth groups.
• Employed multivariate multinomial logistic regression to identify predictors of non-response categories.

Main Results:

• Non-responders are not a homogenous group; approximately 75% demonstrated significant growth ('high growth non-responders'), while 25% showed limited growth ('low growth non-responders').
• High growth non-responders exhibited worse initial anthropometric status than recovered children but followed a similar growth trajectory, suggesting delayed recovery.
• Low growth non-responders displayed minimal growth throughout treatment, indicating true treatment failure.

Conclusions:

• The current definition of 'non-response' in wasting treatment is overly simplistic, failing to distinguish between delayed responders and true treatment failures.
• High growth non-responders should be considered 'delayed responders' and may benefit from extended treatment duration.
• Low growth non-responders represent genuine treatment failures requiring prompt referral for further investigation and management.