Urban site
Late ANC initiation by unemployed women may be attributed to inability to meet the costs associated with the ANC visit such as transport and food, despite the actual health service being provided free of charge. Transport has been identified as the biggest cost of ANC in the South African context [37]. Employed women may also have access to health insurance and seek private health care services and better quality of care compared to unemployed women who rely on the public health facilities. In the public health sector in general, increasing the number of ANC clinics that are physically detached from the main health facility and introducing mobile ANC could increase utilization of these services and motivate early ANC initiation and adequate attendance. These measures could address women’s lack of access to antenatal clinics and eliminate travel costs associated with visits to the clinics.
Pregnancy desire is an important factor in determining timing of ANC initiation and our results confirm findings in previous studies [6, 28, 38]. It may therefore be useful to focus on increasing contraceptive uptake and family planning services in these populations [39]. Health education programmes should be tailored to the different social contexts and age categories of women, to encourage free interaction and be aimed at informing and educating all women on the different family planning options available to prevent unintended pregnancy and make these options easily accessible to all women.
Interestingly, fewer psychosocial variables examined were significantly associated with ANC attendance than would be expected. For example social support, social capital, self-esteem and mental health status were not significantly associated with outcome variables as has been found in the literature [16, 24, 30]. These findings can be attributed to reduced power of the study arising from how the variables were originally measured, the adaptations (or lack thereof) made to the instruments used and the questionnaire design. Because the original study from which the data was derived was based on sample size calculations for alcohol exposed pregnancies [17] the smaller sample size arising from the strict inclusion criteria for this study may have reduced the power of the study. Increasing the sample size would address these limitations and ensure that the study is sufficiently powered. Also, non-response (26 % and 17 % for urban and rural sites respectively) could have led to failure to capture those women with poor mental health and low self-esteem etc., resulting in the current findings. Other psychosocial factors associated with late initiation of antenatal care and infrequent ANC visits which were not considered in the current study include abortion contemplation and delayed diagnosis of the pregnancy [6].
When all factors are considered together, the results suggest that a rural coloured woman who initiated ANC early and attended ANC adequately tended to be a married woman with no previous miscarriage, who was not highly religious and had desired to fall pregnant when she conceived (wanted pregnancy) with a partner with higher education (completed at least Grade 9) and who was the father of the child she was carrying during that pregnancy. A black woman from the urban site who initiated ANC early and adequately attends ANC visits was employed and had desired to fall pregnant when she conceived (wanted pregnancy). The fact that the patterns of associations differed in these two populations is consistent with our a priori expectation that the social and cultural context of these two populations are very different [17]. Because the data used in this study came from a cross-sectional survey, our findings cannot confirm causal relationships because of a lack of temporality, but suggest associations between the psychosocial factors and outcomes which could be usefully explored in longitudinal studies.