| Introduction/methods. Growth retardation is widespread among children living in urban slums in Bangladesh. This study uses mixed linear regression analysis to examine factors influencing short-term and long-term child growth as specified by weight-for-height and height-for-age z-scores respectively among children in three cities, Dhaka, Khulna and Chittagong. Data were collected longitudinally between August 1991 and December 1993 on 5,275 children age 6–59 months.; Results. Marked seasonal variations in child growth as well as in indicators of economic distress, income, morbidity, and food availability were observed. Multivariate regression results show that for all age groups studied child age, maternal education, season, year, and city of residence were significant indicators of both short- and long-term growth. In addition to these factors, the short-term growth of younger children ages 6 to 23 months was also significantly affected by diarrhea, fever, birth order, and household grain consumption. Long-term growth of these younger children was influenced by paternal education, dwelling size, birth order, number of preschool aged children, and wheat consumption. Among older children 24–59 months of age, short-term growth was influenced by diarrhea, fever, respiratory infection, type of employment of the main earner, economic distress as indicated by recent food loan, food expenditure per capita, and rice consumption. Older children's long-term growth was affected by paternal education, dwelling size, and birth order. Maternal education and dwelling size are observed to not only be associated with higher levels of long-term growth, but also with improved growth velocity as measured by height-for-age z-scores improving over age among the younger children.; The magnitude of the impact of birth order and male education on child growth were found to differ by season. Hence, the association between birth order and weight-for-height scores was greater during the hot-wet, ‘lean’ season than during the cold season. Important city-wise differences in the strength of determinants in impacting child growth including child age, birth order, wealth, food expenditure, and rice consumption.; This study reports the impact of an episode of diarrhea or fever to have a negative impact on short-term child growth to the same or greater extent than reduced household food availability. Socio-demographic indicators are also found to have significant impact on the growth of young children 6–23 months of age, particularly during the hot-wet season in these urban slum populations.; Conclusion. With morbidity having greater impact on child growth than food availability and socio-demographic indicators significantly influencing child growth, particularly during the hot-wet seasons, food supplementation programs would be expected to be more effective when combined with efforts to control diarrhea, perhaps with more intensified support during the hot-wet seasons. Upscaling of efforts is needed in Chittagong and Khulna where the urban poor appear to be at significantly greater need than in Dhaka. |
