UC Davis

ABSTRACTIn low- and middle-income countries (LMICs), the prevalence of inadequate micronutrient intake and micronutrient deficiency is high, particularly among women and young children. Sub-Saharan Africa is disproportionally affected by micronutrient deficiencies (MNDs) and the progress towards sustained reduction in MNDs has been limited. In recent years, large-scale food fortification has gained traction in LMICs, including sub-Saharan African countries, to address micronutrient deficiency. Given the long-term effort and investment that is needed to implement and sustain fortification programs, dietary intake data should be used in overall planning of fortification programs to make informed judgments about the appropriateness of the food vehicle, and the types and amounts of the specific nutrients to be added. Simulation studies provide insights about the potential contributions of food fortification to micronutrient intakes and, alongside information on program costs, can provide an efficient way of comparing hypothetical fortification scenarios that can then be translated into nutrition policies and practices. However, the usefulness of simulation modeling studies depends on the validity of dietary requirements applied and assumptions taken into consideration. Zinc Nutrient Reference Values (NRVs) set by several expert groups differ widely. The effects of these differences on the predicted impacts and on the cost-effectiveness of zinc fortification programs have not been fully explored. In the first study, we estimated the prevalence of inadequate zinc intake and the predicted impact and cost-effectiveness of zinc fortification programs using NRVs published by different authorities based on data from a nationally representative nutrition survey among children and women in Cameroon. The distribution of usual zinc intakes was estimated using the National Cancer Institute (NCI) method. Prevalence of total zinc intake below the estimated average requirement (i.e. dietary requirement) and prevalence of “absorbable zinc intake” below the physiological requirement were estimated using NRVs from 4 expert groups: World Health Organizations (WHO), Institute of Medicine (IOM), International Zinc Nutrition Consultative Group (IZiNCG) and European Food Safety Authority (EFSA). The estimated prevalence of inadequate zinc intake varied substantially ranging from 10% (IZiNCG-physiological requirement) to 81% (EFSA-physiological requirement) among children, and from 9.4% (WHO-physiological requirement) to 94% (IOM-physiological requirement) among women. These differences observed in the prevalence of inadequate intake translated to differences in the estimated benefits and cost-effectiveness of zinc fortification programs. Therefore, depending on the NRVs applied, assessments will differ regarding the need for, and the benefits and cost-effectiveness of, zinc fortification programs. Efforts are needed to harmonize NRVs for zinc. Snack products that are voluntarily fortified with micronutrients are increasingly available but are not necessarily formulated to meet known dietary nutrient gaps, so the potential impacts on population micronutrient intake adequacy are uncertain. In the second study, we predicted the impacts of hypothetical micronutrient-fortified biscuits on inadequate micronutrient intake among children and women of reproductive age (WRA) based on nationally representative data in Cameroon. We estimated usual nutrient intake distributions using the NCI method and simulated the impacts of biscuit fortification on prevalence of micronutrient intake (vitamin A (VA), folate, vitamin B-12, iron, and zinc) below the estimated average requirement, given observed biscuit consumption, in the presence and absence of large-scale food fortification (LSFF) programs. In the absence of LSFF programs, biscuits fortified with retinol (600μg/100g), folic acid (300μg/100g), and zinc (8mg/100g) were predicted to reduce the prevalence of inadequacy among children by 10.3 ± 4.4, 13.2 ± 4.2 and 12.0 ± 6.1 percentage points, respectively, in Yaoundé/Douala. However, when the impact of existing national VA-fortified oil, and folic acid and zinc-fortified wheat flour programs were considered, the additional impacts of fortified biscuits were reduced substantially. Micronutrient-fortified biscuits were predicted to have minimal impacts on dietary inadequacy among WRA, with or without LSFF programs. Given observed patterns of biscuit consumption in Cameroon, biscuit fortification is unlikely to reduce dietary inadequacy of studied micronutrients, except possibly for selected nutrients among children in urban areas, but only in the absence of LSFF programs. Zambia has been implementing mandatory sugar fortification with VA, however the contribution of VA-fortified sugar to VA intakes and status has not been directly assessed. In the third study, we predicted the potential impacts of sugar fortification with VA on prevalence of VA inadequacy, and examine its association with plasma and breast milk retinol among lactating women based on baseline data from a randomized trial from Mkushi District in rural Zambia. We simulated VA intake under various sugar fortification scenarios: 3.1 and 8.8 mg/kg (measured median fortification levels of VA in sugar from previous studies), 10 mg/kg (minimum legal requirement) and 15 mg/kg (minimum legal requirement at factory level). We applied the NCI’s bivariate model to examine associations of usual intake of sugar and dietary VA with plasma and breast milk retinol concentrations measured by high performance liquid chromatography. Our model predicted that sugar fortification with VA at 3.1 mg/kg, 8.8 mg/kg, 10 mg/kg and 15 mg/kg would reduce the prevalence of VA inadequacy by 7 (SE:6), 24 (SE:14), 30 (SE:15) and 47 (SE:18) percentage points, respectively, without increasing the risk of retinol intake above the UL. Usual sugar intake and usual VA intake were not associated with plasma retinol (β = 0.003; 95% CI: -0.015, 0.020) or with breastmilk retinol concentrations (β = 0.012; 95% CI: -0.002, 0.026). Usual VA intake from other foods sources was significantly associated with the log of breast milk retinol in µmol/ L (β = 0.001; 95% CI: 0.0002, 0.002), but not with plasma retinol (β = -0.0001; 95% CI: -0.001, 0.001). In Zambia, sugar fortification has the potential to reduce dietary VA inadequacy. However, the impacts on VA intakes and any improvements in VA status are likely to be limited if the program is not implemented as planned. Even if target fortification levels are achieved (10 mg/kg), sugar fortification alone is unlikely to eliminate dietary VA inadequacy among lactating women in Zambia. Together, the three studies from the dissertation provide evidence on the effect of using different zinc reference values on the predicted benefits and cost-effectiveness of zinc fortification programs, and the impacts of voluntary snack food fortification and mandatory sugar fortification on prevalence of inadequate intake. These findings have important implications with regard to planning and evaluation of fortification programs in LMIC contexts.