Madagascar · Child Underweight, Under 5 (survey, %)

22.6%2021

Publisher note: Enquête Démographique et de Santé à Madagascar, 2021. Antananarivo, Madagascar et Rockville, Maryland, USA : INSTAT et ICF.

Unit: percent of children under age 5. Publisher measure: Underweight prevalence among children under 5 years of age (% weight-for-age <-2 SD), survey-based estimates.

Weight-for-age below -2 SD, WHO country-level survey estimates, both sexes, no additional disaggregation; retain survey comments. Reference years are irregular survey labels; fieldwork windows were not supplied by the retained projection. Matching year labels do not prove matching collection periods.

Measurement basis: survey-based estimate; reference years are irregular.

Reference-period cadence: irregular survey reference years.

Displayed values are rounded, with extra precision when needed to keep a positive estimate positive. Downloads retain the full numeric estimate and the publisher's original display text separately.

A “<” sign means strictly below the stated number. Numeric readings and uncertainty endpoints keep their own comparison signs. A number without a sign does not establish certainty or an observed outcome.

World Health Organization (Global Health Observatory); retrieved 2026-09-07. Retrieval time is separate from an official publisher release vintage.

Download this immutable source capture · Source and coverage corrections

Choose a year or compare countries

One row per country, indicator and year, with raw values, source links and reuse notices. Thresholds have an empty numeric value and separate number/operator fields; value_kind distinguishes upper bounds from absent readings.

History and uncertainty

Charts plot numeric readings. Reported upper bounds create gaps; their values, uncertainty endpoints and source notes remain available in the table and downloads.

Reported estimates and uncertainty · 9 rows

Each comparison sign applies to its own reading or uncertainty endpoint. Values retain source precision. “Not reported” identifies an absent reading or uncertainty endpoint in this source capture.

Reported estimates and independently qualified uncertainty endpoints
CountryYearReadingLower endpointUpper endpointPublisher note
Madagascar199235.233.437.1Enquête nationale démographique et sanitaire 1992. Demographic and Health Surveys. Centre National de Recherches sur l'Environnement. Antananarivo, Madagascar, 1994
Madagascar199340.9Not reportedNot reportedEnquête permanente aupres des menages - rapport principal, décembre 1995. Antananarivo, Madagascar, 1995 Converted estimate
Madagascar199530.4Not reportedNot reportedEnquête par grappes a indicateurs multiples [multiple indicators cluster survey (MICS)], Madagascar 1995 (rapport préliminaires). Antananarivo, Madagascar: Institut National de la Statistique et UNICEF, 1996 Converted estimate
Madagascar199734.532.336.9Enquête démographique et de santé, Madagascar 1997. Demographic and Health Surveys. Institut Nacional de la Statistique. Antananarivo, Madgascar, 1998 Reduced age range
Madagascar200436.734.139.5Enquête Démographique et de Santé de Madagascar 2003-2004. Demographic and Health Surveys. Calverton, Maryland, USA : INSTAT et ORC Macro, 2005
Madagascar201128.426.930Enquête Anthropométrique et Développement de l'Enfant Madagascar 2011, Round 3 Oedema data was not considered in the analysis
Madagascar201232.630.934.4Enquête Nationale Sur le Suivi des Objectifs du Millènaire pour le Dèveloppement de Madagascar, (ENSOMD) 2012-2013 Oedema data was not considered in the analysis
Madagascar201826.425.327.5Enquête par grappes à indicateurs multiples-MICS Madagascar 2018, Rapport final. Antananarivo, Madagascar : INSTAT et UNICEF 2019
Madagascar202122.621.224.1Enquête Démographique et de Santé à Madagascar, 2021. Antananarivo, Madagascar et Rockville, Maryland, USA : INSTAT et ICF.

One row per country, indicator and year, with raw values, source links and reuse notices. Thresholds have an empty numeric value and separate number/operator fields; value_kind distinguishes upper bounds from absent readings.