Neurulation is the fundamental embryonic process that leads to the development of the neural tube in two phases(Reference Kibar, Capra and Gros1). The first phase occurs between the third and fourth developmental week (primary neurulation) and leads to the formation of the brain and most of the spinal cord. Secondary neurulation occurs between the fifth and sixth developmental week and creates the lowest portion of the spinal cord including most of the sacral and all of the coccygeal regions. A neurulation defect results in the formation of neural tube defects (NTD). A multi-centre study monitoring all birth defects in Korea reported approximately 3·5 pregnancies per 10000 births to be affected by NTD(Reference Yang, Kim and Chung2). On the other hand, it has been estimated that 50–70 % of NTD could be prevented by a daily intake of at least 400 μg of folic acid (FA) in the periconceptional period(Reference Johnston3); the highest efficacy of FA is expected to occur at a daily dose level of 5 mg(Reference Wald, Law, Morris and Wald4). Unfortunately, approximately 50 % of all pregnancies in Korea are unintended(Reference Han, Nava-Ocampo and Koren5), jeopardizing our efforts to promote FA supplementation during the periconceptional period.
Because of the importance of establishing successful public programmes to promote FA supplementation among women of childbearing age, we aimed to investigate which factors are associated with a positive intake of FA in the periconceptional period among Korean women.
Methods
The study was approved by the Institutional Research Board at the Cheil General Hospital and Women’s Healthcare Center, Seoul, South Korea, and voluntary consent was obtained from all participants. Within a 6-month period, from October 2005 to March 2006, a questionnaire was given to 1313 pregnant women who were participating in a two-day training course at the maternity school of the hospital. The course was available every month and provided information for pregnant women on the prenatal, birth and postpartum periods as well as on breast-feeding benefits. The following information was obtained from the participants: maternal and gestational age, obstetric history (gravidity, parity, number of vaginal deliveries, number of previous abortions), employment status, educational level and household monthly income level. The participants also provided information on whether their current pregnancy was planned, whether they took FA supplementation in the periconceptional period, and whether they were aware of the prevention of NTD by periconceptional supplementation of FA. If participants had a positive history of FA intake, they were also asked to provide the dose according to the following three categories: unknown dose, 400 μg/d or >400 μg/d.
For purposes of the present study, the following variables were classified into two categories: age as <35 years and ≥35 years; number of previous spontaneous abortions as 0 and ≥1; employment status as employed and unemployed; educational level as high school or less and college or higher; and household monthly income level as <$US 2000 and ≥$US 2000. The latter was classified by considering that lower economic class in Korea has an average monthly income of <$US 2000 ($US 1 was equivalent to approximately 1000 Won at the time of the study). In addition, the time elapsed from 3 months before the estimated date of conception to 4 weeks postconception was considered as the periconceptional period. Based on the approximate time at which the first phase of neurulation is completed(Reference Kibar, Capra and Gros1), we limited the periconceptional period to 4 weeks postconception.
Demographic, obstetric and socio-economic data were compared between participants with positive and negative supplementation of FA in the periconceptional period by the χ 2 test. Those characteristics identified as statistically different between groups (P < 0·05) were subsequently included in multivariate logistic regression analyses in order to obtain the corresponding adjusted odds ratio and 95 % confidence interval. The statistical analyses were performed using the Statistical Package for the Social Sciences statistical software package version 10·07 (SPSS Inc., Chicago, IL, USA); all statistical tests were done with α of 0·05 by two-tailed tests.
Results
After excluding subjects with incomplete or inconsistent data, 1277 (97·3 %) women were included in the analysis. Included participants were aged 29·4 (sd 2·9) years and had a mean gestational age of 27·9 (sd 7·1) weeks. There were 131 (10·3 %) women who took FA during the periconceptional period. Of them, 84·0 % had an education level of college or higher, 74·8 % were primigravidas, 19·1 % had ≥1 previous spontaneous abortion and in 86·3 % the pregnancy was planned. The dose of FA was unknown in ninety-eight participants (74·8 %), 400 μg/d in 30 participants (22·9 %) and >400 μg/d in three participants (2·3 %).
The percentage of women with previous spontaneous abortions or who had planned their pregnancy was higher in the group who took FA in the periconceptional period than in the group who did not (Table 1). According to multiple logistic regression analyses, women who had ≥1 previous spontaneous abortions, planned their pregnancy and were aware of the protective effects of FA were more likely to take FA during the periconceptional period. The adjusted OR were 1·79 (95 % CI 1·10, 2·91; P = 0·02), 4·10 (95 % CI 2·43, 6·78; P < 0.001) and 6·63 (95 % CI 2·08, 21·12; P = 0·001), respectively (Table 2).
Table 1 Demographic, socio-economic and clinical characteristics: pregnant Korean women (n 1277), October 2005–March 2006
*Employment status at the time of periconceptional period.
†$US 1 = 1000 Won.
Table 2 Factors associated with periconceptional intake of folic acid: pregnant Korean women, October 2005–March 2006
The results were obtained by multiple logistic regression analyses and included the three variables that resulted as statistically significant in the comparisons between women who took folic acid in the periconceptional period and those who did not.
Discussion
The present study found that only a small percentage (10·3 %) of Korean women took FA supplementation during the periconceptional period. This rate is considerably lower than the 20·3 % reported previously in Qatar(Reference Bener, Al Maadid, Al-Bast and Al-Marri6), 25 % in Canada(Reference Morin, De Wals, St-Cyr-Tribble, Niyonsenga and Payette7), 32 % in the USA(8) and 45·5 % in the United Arab Emirates(Reference Abdulrazzaq, Al-Gazali, Bener, Hossein, Verghese, Dawodu and Padmanabhan9). Over the past decade the birth rate in Korea has decreased markedly, to as low as 1·08 per couple(10). However, the rate of planned pregnancies remains low. For example, only 50 % of women receiving teratogen-risk counselling services at the Korean Motherisk Program reported a planned pregnancy(Reference Han, Nava-Ocampo and Koren5), and although the percentage was higher in the present study (63·5 %), it implies that at least one third of all pregnancies are still unplanned. In the present study, the percentage of planned pregnancies was significantly higher among women who took FA in the periconceptional period than among non-takers of FA. However, the percentage of planned pregnancies was at least five times more than the overall percentage of women who took FA, suggesting that other factors may also affect the periconceptional supplementation of FA.
It is known that highly educated women know more about FA and take it more often in the periconceptional period and first trimester of pregnancy(Reference Bener, Al Maadid, Al-Bast and Al-Marri6, Reference Daltveit, Vollset, Lande and Øien11). In the present study, the percentage of women who were employed, had an education level of college or higher or had a household monthly income level ≥$US 2000 did not differ between FA takers and non-takers. On the other hand, a planned pregnancy, the adverse experience of previous abortions and awareness of the benefits of FA in pregnancy were associated with a significantly higher intake of FA during the periconceptional period. However, we were not able to clarify whether in a real-time situation these three factors would independently favour FA supplementation during the periconceptional period. In a previous study, variations in the attitude of women to take FA during the periconceptional period were secondary to the variable knowledge of its benefits among their obstetricians(Reference Power, Holzman and Schulkin12). However, the main source of information on the benefits of FA in early pregnancy appears to be the mass media(Reference Bekkers and Eskes13). This knowledge will be critical in providing public strategies and national campaigns to increase folate awareness among women of childbearing age. On the other hand, the number of unplanned pregnancies remains considerably elevated(Reference Han, Nava-Ocampo and Koren5), limiting the efficacy of public education campaigns to encourage the periconceptional use of FA supplements.
The present study was performed among pregnant women at a mean gestational age of 28 weeks; women who had miscarriages or terminated their pregnancies were not included and therefore their attitude towards intake of FA in the periconceptional period was not taken into account. In addition, we cannot rule out the recall bias at this advanced gestational age concerning the periconceptional intake of FA among participants. The dose of FA was not evaluated in detail, and therefore some women who took only a few doses during their first weeks of pregnancy were classified as subjects with positive FA supplementation. Finally, for purposes of the analyses, the study variables were categorized a priori, limiting evaluation of the effect of certain variables, e.g. age, on FA supplementation in a wider range of possibilities.
There is still an ongoing debate on the pros and cons of mandatory FA fortification(Reference Wald and Oakley14, Reference Hubner, Houlston and Muir15); among the cons, FA has been associated with potential cancer promotion as well as with a decrease in plasma homocysteine levels. In addition, FA fortification has been found to be a cause of masking vitamin B12 deficiency, particularly in the elderly population(Reference Cuskelly, Mooney and Young16). However, diet by itself does not appear to provide pregnant women with sufficient amounts of folate to prevent major birth defects. For example, in Norway, the country with the highest rate of facial clefts in Europe, the fortification of foods with FA has not been allowed(Reference Mossey and Little17). In contrast, flour and other grain products are being fortified at a minimum level of 0.14 mg FA/100 g cereal grain in Canada and the USA and a decline in NTD was observed as a result of that fortification(Reference Ray, Meier, Vermeulen, Boss, Wyatt and Cole18, Reference De Wals, Tairou and van Allen19). However, the fortification level should be enough to provide a daily intake of 0·6 to 0·8 mg FA in order to attain the highest benefits(Reference Wald20). In contrast, there is no independent gold standard or reference method for unequivocal characterization of vitamin B12 deficiency, nor an agreement on how to treat or when and how to monitor the effects of treatment available(Reference Cuskelly, Mooney and Young16). Although deficiency of vitamin B12 in an elderly subpopulation cannot be minimized, the current analysis of risks and benefits is in favour of FA fortification, especially in a population where a very low percentage of pregnant women receive FA supplementation in the periconceptional period and most of women who received FA ignore the dose.
In summary, Korean women undergoing a planned pregnancy, who had previous spontaneous abortions and who were aware of the preventive effects of FA were more likely to take FA during the periconceptional period than those who did not. However, despite its proven efficacy for preventing congenital malformations, intentional intake of FA at the periconceptional period occurred in only approximately 10 % of women.
Acknowledgements
Author contributions: J.-Y.H. conceived the idea and study design, and coordinated the study. Y.-J.C. collected the information in electronic databases. A.A.N.-O. participated in the study design. H.-K.A. and J.-O.K. performed the statistical analyses. All the authors participated in drafting the manuscript, discussing the results, and preparing, reviewing and correcting the manuscript.
Acknowledgements: The authors are indebted to pregnant women who participated in the study during the maternity course at Cheil Hospital and Healthcare Center, Kwandong University College of Medicine, Seoul, South Korea.
We have no conflicts of interest to declare.
