Research Scientist, UCLA Mindfulness Awareness Research Center
Dissertation Title: "US Folic Acid Fortification Policy: Methods for Evaluating Folate Intake and Projecting Health Outcomes"
Increased prenatal intake of folate or folic acid early in pregnancy can significantly reduce the risk of neural tube defects (NTDs) in newborns. In an effort to increase folate intake among women of childbearing age, the U.S. Food and Drug Administration in 1998 mandated the fortification of all enriched grain products with folic acid. Such fortification may also reduce the risks of heart disease and colon cancer, while increasing that of B-12 deficiency masking. This research quantifies folate consumption before and after policy implementation, evaluates the cost-effectiveness of different fortification strategies, and explores the potential bias associated with categorizing continuous variables--such as folate--in decision-analytic models.
In chapter one, we estimate population-level changes in folate intake since policy implementation. Our analysis shows a substantial increase in folate consumption after fortification, with considerable variations by age, gender, and race/ethnicity. The proportion of women of childbearing age consuming greater than 400 mcg/day of folate did increase but did not reach the FDA's 50% target, with blacks and Mexican-Americans lagging behind whites. The percent of older persons consuming over the tolerable upper intake level increased only among some populations and remained under 5% for all groups.
In chapter two, we quantify the projected health, quality-of-life, and economic outcomes for NTDs, myocardial infarctions (MIs), colon cancer, and B-12 masking associated with four fortification scenarios. We predicted that for all post-fortification strategies, NTD incidence would decrease by up to 40% and that of MIs and colon cancer would decline by up to 20%, with significant variations by fortification level, gender, age, and race/ethnicity. Fortifying with 700 mcg/100 g enriched grain strongly dominated all other scenarios at both the population and subgroup levels, with predicted population benefits of 278,849 QALYs gained and $4 billion saved.
Finally, in chapter three we assess the tradeoffs between model bias and complexity when categorizing continuous risk factors in decision-analytic models. We found that categorization tended to overestimate life expectancy gains, although the bias did not exceed 4% and was strongest when the risk factor effect was highest. When at least three categories were used, the bias remained under 1.5%.