The 14-Day Discharge Paradox in Facility-Based Management of Severe Acute Malnutrition: A Multicentre Analysis of Length of Stay, Weight Gain, and Treatment Outcomes From Nutrition Rehabilitation Centres in Gwalior District, Madhya Pradesh, India.

BACKGROUND

A minimum inpatient stay of 14 days is recommended for children with severe acute malnutrition (SAM) admitted to nutrition rehabilitation centres (NRCs) in India, yet facility-level data consistently report mean stays well below this benchmark, and the consequences for weight gain and treatment outcomes have rarely been quantified.

OBJECTIVE

This study aimed to quantify the relationship between length of NRC stay and (i) average daily weight gain, (ii) attainment of the ≥ 8 g/kg/day target, and (iii) discharge outcomes (recovery, defaulter, non-responder) in a multicentre cohort of SAM children in Gwalior district, Madhya Pradesh, India.

METHODS

A prospective cohort of 332 SAM children admitted to the five NRCs of Gwalior district during 2021 was stratified by length of stay into three groups (≤ 7, 8-13, and ≥ 14 days). Stratum-wise mean weight gain, target attainment, and discharge outcomes were compared using analysis of variance (ANOVA) and χ² test. Adjusted associations were estimated by multivariable linear regression (weight gain) and multinomial logistic regression (outcome), controlling for age, sex, admission weight, oedematous SAM and comorbidity burden.

RESULTS

Mean length of stay was 9.88 ± 3.19 days; only 37 children (11.1%) completed ≥ 14 days. Mean weight gain rose monotonically across strata: 5.32 g/kg/day (≤ 7 days) → 7.18 g/kg/day (8-13 days) → 11.47 g/kg/day (≥ 14 days) (F = 14.83, p < 0.001). The proportion achieving the ≥ 8 g/kg/day target rose from 16.7% (14/84) to 31.8% (67/211) to 67.6% (25/37) across strata (χ² = 38.91, p < 0.001). Discharge recovery rose from 58.3% (49/84) to 73.5% (155/211) to 86.5% (32/37) (χ² = 12.74, p = 0.002), and defaulter rate fell from 30.9% (26/84) to 18.0% (38/211) to 5.4% (2/37) (χ² = 14.92, p < 0.001). In adjusted analysis, each additional inpatient day was associated with a 0.71 g/kg/day increase in weight gain (β = 0.71; 95% CI 0.45-0.97) and a 12% reduction in odds of default (adjusted odds ratio (aOR) 0.88 per day; 95% CI 0.83-0.93).

CONCLUSION

Nine out of every 10 NRC stays in Gwalior ended before the 14-day benchmark, and the associated outcome deficit was measurable. Re-aligning discharge with anthropometric criteria (≥ 15% weight gain or ≥ 5 g/kg/day for three consecutive days) rather than fixed-day rules, combined with wage-equivalent caregiver reimbursement and structured community follow-up, may translate into better outcomes within existing infrastructure, a hypothesis that warrants confirmation in interventional or implementation research.