Objective: To investigate gender inequalities in child growth and nutritional status in relation to socio-economic status in Bangladesh.
Design: A 16-month longitudinal study of child growth measuring anthropometric and socio-economic status.
Setting: A rural area of Jamalpur district, northern Bangladesh.
Subjects: 1366 children from 2 to 6 years of age.
Methods: Child height and weight were measured monthly. Morbidity, food intake and health-seeking behaviours were assessed fortnightly. Multivariable analyses were performed on the growth and nutritional status of male and female children in relation to socio-economic factors including father's occupation, parental education, birth order and family size.
Results: There was no evidence of gender bias in farming and trading/employee households but landless female children had significantly poorer height-for-age (P < 0.001) and weight-for-age (P < 0.001) than their male counterparts. During a period of natural disaster, a statistically significant interaction was observed between father's occupation and sex (P < 0.05) such that the combination of being female and being landless was more detrimental to nutritional status than either variable alone. Over the following 16-months, catch-up-growth was apparent in landless female children who grew significantly more in height-for-age (P < 0.001) and weight-for-age (P < 0.001) than their male counterparts.
Conclusions: Gender inequalities in health in Bangladesh varied significantly according to occupational status, such that the effect of sex was dependent upon occupation. These effects were statistically significant during the period of natural disaster but became insignificant as local conditions improved. This demonstrates both temporal and socio-economic variation in gender inequalities in health.
PIP: During February 1989-June 1990 in Bangladesh, local field assistants collected data on 1366 children 2-6 years old, attending maternal and child health clinics operated by a nongovernmental organization, and living in 13 villages in Jamalpur District situated on the banks of the Jamuna River. The field assistants made home visits to record child morbidity every 2 weeks and to measure child height and weight once a month. During January-April 1989, this area suffered from extensive food shortages due to a prolonged drought and one of the worst floods recorded in Bangladesh. Gender bias was not apparent in farming and trading/employee households. In landless households (i.e., fathers' occupation was laborer), girls were significantly shorter and less heavy than boys (p 0.001), however. During a natural disaster, fathers' occupation significantly interacted with sex (p 0.05). Specifically, children who were both female and living in a landless household were more likely to have poor nutritional status than children who were female and living in a farming or trading/employee household and children who were male and living in a landless household. This interaction was not apparent as local conditions improved. Over the 16 months following the natural disaster, landless girls grew significantly more in height-for-age and weight-for-age than landless boys (p 0.001). In other words, these girls experienced more catch-up growth than the boys. At the end of the study, nutritional status varied only according to socioeconomic status but not according to gender. These findings suggest that gender bias within this population depends on changes in food availability and the rural economy. Thus, child nutrition programs should target landless girls, who are at highest risk of gender discrimination and malnutrition during economic adversity.