Objectives: The objectives of this study were to investigate the difference in timing of the first antenatal visit between ethnic groups and to explore the contribution of several noneconomic risk factors.
Design: Prospective cohort study.
Setting: All independent midwifery practices in the city of Amsterdam and all six Amsterdam hospitals.
Population: Consecutive cohort of pregnant women (n = 12 381). Ethnic groups were distinguished by country of birth.
Methods: Questionnaire data showed possible risk factors for late start. A Cox-proportional hazards model was created with (1) only ethnic group and (2) the addition of all significant risk factors, both time fixed and time dependent.
Main outcome measures: Gestational age at first visit.
Results: The questionnaire was returned by 8267 pregnant women (response rate 67%). All non-Dutch ethnic groups were significantly later in starting antenatal care during the whole duration of pregnancy compared with the ethnic Dutch group (hazard ratio [95% CI]: other Western, 0.83 [0.76-0.90]; Surinamese, 0.62 [0.56-0.68]; Antillean, 0.56 [0.45-0.70]; Turkish, 0.62 [0.55-0.69]; Moroccan, 0.56 [0.52-0.62]; Ghanaians, 0.50 [0.43-0.58] and other non-Western, 0.61 [0.56-0.67]). The range at which 90% were in care varied between 16 weeks and 3 days for Dutch and 24 weeks and 4 days for Ghanaians. These differences disappeared almost totally in the non-Dutch-speaking ethnic groups when the following risk factors were added to the model: poor language proficiency, low maternal education, teenage pregnancy, multiparity and unplanned pregnancy. The differences remained in the Dutch-speaking ethnic groups.
Conclusions: We observed a disturbing delay by all ethnic groups in the timing of their first antenatal visit. In women born in non-Dutch-speaking, non-Western countries, these differences were explained by a higher prevalence of the risk factors: poor language proficiency in Dutch, lower maternal education and more teenage pregnancies. In women born in Dutch-speaking, non-Western countries, the disparities cannot be explained by higher prevalence of these risk factors, indicating that cultural factors play a role.