To examine the relationship between homestead food production and night blindness among pre-school children in rural Bangladesh in the presence of a national vitamin A supplementation programme.

A cross-sectional study.

A population-based sample of six rural divisions of Bangladesh assessed in the Bangladesh Nutrition Surveillance Project 2001–2005.

A total of 158 898 children aged 12–59 months.

The prevalence rates of night blindness in children among those who did and did not receive vitamin A capsules in the last 6 months were 0·07 % and 0·13 %, respectively. Given the known effect of vitamin A supplementation on night blindness, the analysis was stratified by children's receipt of vitamin A capsules in the last 6 months. Among children who did not receive vitamin A capsules in the last 6 months, the lack of a home garden was associated with increased odds of night blindness (OR = 3·16, 95 % CI 1·76, 5·68; P = 0·0001). Among children who received vitamin A capsules in the last 6 months, the lack of a home garden was not associated with night blindness (OR = 1·28, 95 % CI 0·71, 2·31; P = 0·4).

Homestead food production confers a protective effect against night blindness among pre-school children who missed vitamin A supplementation in rural Bangladesh.

To assess the nutritional status and risk factors of undernutrition in post-menarcheal girls in rural northern Bangladesh.

Cross-sectional data on anthropometric measurements, dietary intake, work activity, morbidity and socio-economic status were collected from 12- to 19-year-old primigravidae (n 209) and never-pregnant adolescents (n 456) matched on age and time since menarche. Multiple regression analyses were conducted to determine predictors of stunting, thinness, upper-arm muscle (UAMZ) and fat area Z-scores (UAFZ) among the adolescent girls.

A large proportion of adolescents (49 %) were stunted (height-for-age Z-score <−2) and underweight (40 %; weight-for-age Z-score <−2), but not thin (BMI-for-age <5th percentile; ∼10 %). The mean (sd) UAMZ and UAFZ of the adolescent girls was −0·3 (0·64) and −0·9 (0·40), respectively. Lean mass increased whereas fat mass declined with age. Both stunting and thinness were positively associated with age and time since menarche (P < 0·05). Young age (12–14 years) and literacy were protective against stunting among pregnant adolescents (OR = 0·29, 95 % CI 0·09, 0·88 and OR = 0·50, 95 % CI 0·26, 0·96, respectively). Having symptoms of diarrhoea or dysentery (OR = 7·40, 95 % CI 1·43, 38·29) predicted thinness and was associated with lower UAMZ and UAFZ among never-pregnant girls (both P < 0·05). Performing light-to-moderate activities was protective against thinness among never-pregnant girls (OR = 0·43, 95 % CI 0·22, 0·82), whereas pregnant adolescents who performed high levels of strenuous activities had greater UAMZ (P < 0·05).

Undernutrition was widespread among this post-menarcheal adolescent population. Younger and literate adolescents were less likely to be stunted, whereas thinness and body composition were associated with morbidity and work activity.