To determine whether a possession score or a poverty index best predicts undernutrition and anaemia in women of reproductive age (15–49 years; WRA) and children aged 6–59 months living in Côte d'Ivoire.

Anthropometric measurements were converted to Z-scores to assess stunting, wasting and underweight in children, and converted to BMI in WRA. A venous blood sample was drawn, and Hb concentration and Plasmodium spp. infection were determined. A possession score was generated with categories of zero to four possessions. A five-point (quintile) poverty index using household assets was created using principal component analysis. These socio-economic measures were compared for their ability to predict anaemia and malnutrition.

Data were from a nationally representative survey conducted in Côte d'Ivoire in 2007.

A sample of 768 WRA and 717 children aged 6–59 months was analysed.

Overall, 74·9 % of children and 50·2 % of WRA were anaemic; 39·5 % of the children were stunted, 28·1 % underweight and 12·8 % wasted, while 7·4 % of WRA had BMI < 18·5 kg/m2. In general, there were more stunted and underweight children and thin WRA in rural areas. The poverty index showed a stronger relationship with nutritional status than the possession score; mean Hb difference between the poorest and wealthiest quintiles in children and WRA was 8·2 g/l and 6·5 g/l, respectively (13·9 % and 19·8 % difference in anaemia, respectively; P < 0·001), and Z-scores and BMI were significantly better in the wealthiest quintile (P < 0·001).

The poverty index was generally a better predictor of undernutrition in WRA and pre-school children than the possession score.

To determine how much of the variation in nutritional status of Bangladeshi children under 5 years old can be attributed to the socio-economic status of the family.

Nutritional status used reference Z-scores of weight-for-age (WAZ), height-for-age (HAZ) and weight-for-height (WHZ). A ‘possession score’ was generated based on ownership of a radio, television, bicycle, motorcycle and telephone, and the availability of electricity, with categories of 0 to 4+ possessions. A five-point (quintile) ‘poverty index’ was created using principal component analysis.

The Bangladesh Demographic and Health Survey 2004 was the source of data.

A sample of 4891 children aged <5 years was obtained.

Some 57·8 % of the sample was either stunted, wasted or underweight (7·7 % were stunted, wasted and underweight). Of those stunted (48·4 %), 25·7 % were also underweight. Underweight and wasting prevalences were 40·7 % and 14·3 %, respectively. Mean WAZ, HAZ and WHZ did not differ by sex. Children of mothers with no education or no possessions were, on average, about 1 sd more underweight and stunted than those with higher educated mothers or with 4+ possessions. The possession score provided much greater discrimination of undernutrition than the poverty index. Nearly 50 % of children from households with no possessions were stunted, wasted or underweight (only 27 % in the poorest quintile), compared with only 3–6 % of children from households with 4+ possessions (over 13 % in the richest quintile).

Maternal education and possession score were the main predictors of a child’s nutritional status. Possession score was a much better indicator of undernutrition than the poverty index.