Objective: To investigate the prevalence of vitamin A deficiency (VAD) among pregnant women in rural Bangladesh, and examine the relationship between various factors and vitamin A status. Setting: Community Nutrition Promoter (CNP) centres in Kapasia sub-district of Gazipur district, Bangladesh. Design: A cross-sectional study. Subjects: and methods Two hundred women, aged 18-39 years, in their second or third trimester of pregnancy were selected from seventeen CNP centres in four unions of Kapasia sub-district where they usually visit for antenatal care. Various socio-economic, personal and pregnancy-related information, dietary intake of vitamin A and mid-upper arm circumference (MUAC) data were collected. Serum retinol (vitamin A) concentration was determined. Results: More than half (51 %) of the pregnant women had low vitamin A status (serum retinol <1.05 孯l/l) with 18.5 % having VAD (serum retinol <0.70 孯l/l). Fifty-three per cent of the women's vitamin A intake was less than the recommended dietary allowance. By multiple regression analysis, MUAC, per-capita expenditure on food and wealth index were found to have significant independent positive relationship with serum retinol concentration, while gestational age of the pregnant women had a negative relationship. The overall F-ratio (10.3) was highly significant (P = 0.0001), the adjusted R 2 was 0.18 (multiple R = 0.45). Conclusion: VAD is highly prevalent among rural pregnant women in Bangladesh. Gestational age, nutritional status, per-capita expenditure on food and wealth index appear to be important in influencing the vitamin A status of these women. An appropriate intervention is warranted in order to improve the vitamin A status. 頔he Authors 2008.