D evaluated the effects of different assumptions on the estimated risk of inadequate zinc intake. The present analysis focuses on the Gracillin authors’ previously reported best estimates of country- and regionspecific risks of dietary zinc inadequacy, generated by comparing the estimated quantities of absorbable zinc in national food supplies with the order C.I. 19140 respective population’s theoretical physiological requirements for zinc. This analysis uses a newly created composite nutrient composition database, estimated physiological requirements for absorbed zinc as proposed by the International Zinc Nutrition Consultative Group (IZiNCG), a mathematical model (the Miller equation) to predict zinc absorption based on total dietary zinc and phytate and an assumed 25 interindividual coefficient of variation in zinc intake (Wessells et al.). FAO food balance sheets supply data on annual national food availability, and do not account for differences in dietary zinc intake among individuals and sub-groups within the population. Of particular concern, food balance sheets may be more likely to represent food intake by adults than by infants and young children, who are likely more vulnerable to zinc deficiency than others in the population [1,10,11]. Thus, food balance sheets may not provide a good estimate of inadequate zinc intake by young (preschool aged) children. On the other hand, the prevalence of low height-for-age in children under 5 years of age in a specific population reflects pre- and post-natal nutritional conditions of young children and has been recommended as an indirect indicator of a population’s risk of zinc deficiency. When the prevalence of stunting is greater than 20 , the risk of zinc deficiency may also be elevated [9]. By using 
both food balance sheet information and the prevalence of stunting, it may be 23977191 possible to estimate the risk of zinc deficiency in the whole population, including both older children and adults and preschool children.The objectives of the present study were to use the estimated country- and region-specific prevalence of dietary zinc inadequacy and country-specific rank order of estimated prevalence to: (1) examine dietary patterns associated with the estimated prevalence of inadequate zinc intake, (2) evaluate country-specific secular trends in the estimated 
prevalence of inadequate zinc intake, and (3) compare the estimated prevalence of dietary zinc inadequacy with the national prevalence of stunting in children less than five years of age and create a composite index to identify countries at the highest risk of zinc deficiency, based on both indicators. These analyses were conducted as part of the Nutrition Impact Model Study (NIMS), which was designed to synthesize information related to the health impacts of nutritional conditions and deficiencies and related interventions, in developing countries.Methods Estimation of the Adequacy of Zinc in National Food Supplies Based on National Food Balance DataThe analytic methods, and model assumptions, have been described extensively in the accompanying methodological article (Wessells et al.). In brief, the following steps were completed to estimate the national prevalence of inadequate zinc intake and calculate the country-specific rank order of estimated prevalence. Firstly, we obtained country-specific data on the average daily per capita availability of major food commodities (kcal/capita/d) from national food balance sheets. These data are provided by 188 countries.D evaluated the effects of different assumptions on the estimated risk of inadequate zinc intake. The present analysis focuses on the authors’ previously reported best estimates of country- and regionspecific risks of dietary zinc inadequacy, generated by comparing the estimated quantities of absorbable zinc in national food supplies with the respective population’s theoretical physiological requirements for zinc. This analysis uses a newly created composite nutrient composition database, estimated physiological requirements for absorbed zinc as proposed by the International Zinc Nutrition Consultative Group (IZiNCG), a mathematical model (the Miller equation) to predict zinc absorption based on total dietary zinc and phytate and an assumed 25 interindividual coefficient of variation in zinc intake (Wessells et al.). FAO food balance sheets supply data on annual national food availability, and do not account for differences in dietary zinc intake among individuals and sub-groups within the population. Of particular concern, food balance sheets may be more likely to represent food intake by adults than by infants and young children, who are likely more vulnerable to zinc deficiency than others in the population [1,10,11]. Thus, food balance sheets may not provide a good estimate of inadequate zinc intake by young (preschool aged) children. On the other hand, the prevalence of low height-for-age in children under 5 years of age in a specific population reflects pre- and post-natal nutritional conditions of young children and has been recommended as an indirect indicator of a population’s risk of zinc deficiency. When the prevalence of stunting is greater than 20 , the risk of zinc deficiency may also be elevated [9]. By using both food balance sheet information and the prevalence of stunting, it may be 23977191 possible to estimate the risk of zinc deficiency in the whole population, including both older children and adults and preschool children.The objectives of the present study were to use the estimated country- and region-specific prevalence of dietary zinc inadequacy and country-specific rank order of estimated prevalence to: (1) examine dietary patterns associated with the estimated prevalence of inadequate zinc intake, (2) evaluate country-specific secular trends in the estimated prevalence of inadequate zinc intake, and (3) compare the estimated prevalence of dietary zinc inadequacy with the national prevalence of stunting in children less than five years of age and create a composite index to identify countries at the highest risk of zinc deficiency, based on both indicators. These analyses were conducted as part of the Nutrition Impact Model Study (NIMS), which was designed to synthesize information related to the health impacts of nutritional conditions and deficiencies and related interventions, in developing countries.Methods Estimation of the Adequacy of Zinc in National Food Supplies Based on National Food Balance DataThe analytic methods, and model assumptions, have been described extensively in the accompanying methodological article (Wessells et al.). In brief, the following steps were completed to estimate the national prevalence of inadequate zinc intake and calculate the country-specific rank order of estimated prevalence. Firstly, we obtained country-specific data on the average daily per capita availability of major food commodities (kcal/capita/d) from national food balance sheets. These data are provided by 188 countries.