THE LANCET Introduction Folic acid has been shown to prevent neural-tube defects (NTD) if taken during the periconceptional period. 1,2 Government initiatives designed to implement these findings have suggested an increase in intake of 400 g/day. 3,4 Compliance with these recommendations has been poor. 5 Consequently, many experts believe that primary prevention will be possible only through food fortification; 6 however, this issue is hotly debated. In the USA, 140 g folic acid will be added to every 100 g grain from Jan 1, 1998. This action should deliver an average increase of 100 g per day, based on dietary modelling and on the assumption of equal bioavailability. 7 This dose will almost certainly not mask pernicious anaemia, 8 but it could be too low to prevent NTD. A decision as to whether food should be fortified with folic acid, and if so at what level, is awaited in the UK. The difficulty with any fortification programme is that people who have low intakes of the fortified food must consume sufficient folic acid to benefit, whereas those who have high intakes must not receive a potentially harmful dose. Delivery of a dose of 400 g/day to women of reproductive age would result in a substantial proportion of the population, including elderly people, being exposed to levels of folic acid in the milligram range—amounts that may mask pernicious anaemia. 9 One critical piece of information is missing. What is the minimum dose of folic acid that will prevent NTD? The answer to this question cannot be obtained easily. Further trials with lower doses of folic acid and the occurrence of an NTD as the primary endpoint would be unethical, because women given lower doses might have affected children. Moreover, a food fortification trial would require a very large population, and a suitable control group could be difficult to identify. An alternative approach is to use an endpoint that can be related to the occurrence of NTD— namely, red-cell folate. A relation between folate status as determined by this measurement and the risk of NTD has been established 10 and is accepted. 11. Previous work has shown an inverse relation between the risk of an NTD and maternal red-cell folate. 10 At values below 150 g/L, the risk of an NTD was 6·6 per 1000 births, whereas when red-cell folate was greater than 400 g/L the risk of an NTD was only 0·8 per 1000 births. The overall population risk was 1·9 per 1000 births. Thus, if the average value of red-cell folate in the population could be increased such that all pregnant women had values above 400 g/L, the risk of NTD could be reduced by almost 60%. The risk of NTD can, therefore, be estimated by measurement of maternal red-cell folate. The effect of various levels of fortification on red-cell-folate status can be assessed, which will allow a systematic appraisal of the effect of such doses on NTD risk. We undertook a randomised trial of several doses of folic acid to find out how much folic acid a food fortification Summary Background Although a daily supplement of 400 g folic acid has been shown to prevent neural-tube defects (NTD), most women do not take the recommended supplement. Thus, food fortification is to be introduced in the USA and is being considered in the UK. Because of safety concerns, the USA has chosen a level of fortification that will increase the average woman’s intake by only 100 g. Such an increase, although safe, may be ineffective; but a trial to assess its efficacy would be unethical. Because women with red-cell folate concentrations above 400 g/L have a very low risk of NTD, we undertook a randomised trial of several folic acid doses to find out how much is needed to reach this protective concentration. Methods We screened 323 women. 172 with red-cell folate between 150 g/L and 400 g/L were invited to take part in the trial. 121 women were randomly assigned placebo or 100 g, 200 g, or 400 g daily of additional folic acid. Compliance was monitored by having the women sign a dated sheet when taking the tablet. 95 women completed the 6-month study. Findings There were significant increases in red-cell folate in all folic acid groups. The placebo group showed no significant change. The median incremental changes and median post-treatment concentrations were 67 g/L (95% CI 43–120) and 375 g/L (354–444) in the 100 g/day group, 130 g/L (108–184) and 475 g/L (432–503) in the 200 g/day group, and 200 g/L (125–312) and 571 g/L (481–654) in the 400 g/day group. Interpretation A fortification programme that delivered 400 g folic acid daily to women would protect against NTD, but at the expense of unnecessarily high exposure for many people. Delivery of 200 g daily is also effective against NTD and safer for the general population. Based on projections from the positive folate balance in the group that received 100 g daily, this dose taken continually, as it will be in fortified food, will also produce an important decrease in NTD. Lancet 1997; 350: 1666–69 See Commentary page ???? 1666 Vol 350 • December 6, 1997 Minimum effective dose of folic acid for food fortification to prevent neural-tube defects Sean Daly, James L Mills, Anne M Molloy, Mary Conley, Young J Lee, Peadar N Kirke, Donald G Weir, John M Scott Early reports RCSI Department, Coombe Women’s Hospital, Dublin, Ireland (S Daly MRCOG); Branches of Epidemiology and Statistics, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD, USA (J L Mills MD, M Conley MA, Y J Lee PhD); Departments of Clinical Medicine (A M Molloy PhD, Prof D G Weir FRCP) and Biochemistry (Prof J M Scott ScD), Trinity College, Dublin; and Health Research Board, Dublin, Ireland (P N Kirke FFPHMI) Correspondence to: Dr Sean Daly, Division of Maternal Fetal Medicine, Thomas Jefferson University, 834 Chestnut Street, Suite 400, Philadelphia, PA 19107, USA