Volume 1312, Issue 1 p. 91-104
Original Article
Free Access

Legislative frameworks for corn flour and maize meal fortification

Phillip Makhumula

Corresponding Author

Phillip Makhumula

Lilongwe, Malawi

Address for correspondence: Phillip Makhumula, A44_31, Lilongwe, Malawi. [email protected]Search for more papers by this author
Omar Dary

Omar Dary

International Economic Growth Division, Abt Associates, Bethesda, Maryland

currently at the U.S. Agency for International Development's Bureau for Global Health, Washington, D.C.

Search for more papers by this author
Monica Guamuch

Monica Guamuch

Guatemala City, Guatemala

Search for more papers by this author
Carol Tom

Carol Tom

Nairobi, Kenya

Search for more papers by this author
Ronald Afidra

Ronald Afidra

Flour Fortification Initiative, Kampala, Uganda

Search for more papers by this author
Zo Rambeloson

Zo Rambeloson

FHI 360, Monitoring and Evaluation, Durham, North Carolina

Search for more papers by this author
First published: 12 February 2014
Citations: 14

Abstract

Corn flour and maize meal fortification can benefit the consumer when the added nutrient contents are in amounts appropriate to address nutrient gaps. Legislative instruments (standards and regulations) are needed to provide guidance to the producers and food control authorities. We reviewed a number of national standards and regulations of fortified corn flour and maize meal and identified constraints; contrary to current belief, the practice of using minimum contents or ranges of nutrients has caused confusion, misinterpretation, and conflict, and should therefore be abandoned. On the basis of the findings, a model of fortification legislation is proposed, in which the additional content and the expected average nutrient content in a final product are recommended as the main parameters for quality control and enforcement. For labeling, the average content, or one adjusted to the expected content of the product at the market, can be applied. Variation in micronutrient contents should still be checked to ensure homogeneity but with adherence to clear procedures of sampling and testing, which should be part of the standards and regulations.

Introduction

Fortification of corn flour and maize meal is one potential nutrition intervention for managing micronutrient inadequacies in countries where maize is a staple food produced industrially.1 These conditions are fulfilled in many sub-Saharan African and South American countries, where the fortification of this staple has either already started or is being considered.

A program of food fortification should be guided by technical specifications that respond to epidemiological need2 and to industry context of each country. Those specifications are gathered in standards and regulations, which are the most common legislative instruments applied to fortified foods.3 A few countries have also enacted laws for food fortification, although this practice is increasingly abandoned because of its rigidity in making timely modifications in response to changes in nutrition epidemiology or industry technological advances.

The objective of this paper is to gather existent legislative instruments applicable to the fortification of corn flour and maize meal, to compare them, identify common limitations and constraints, and to propose a generic model that could improve the preparation of these technical and legal instruments. For the latter, this paper uses the estimations done for the case of Kampala, Uganda, in Guamuch et al.,4 in which details about how to use food and nutrient intakes to define nutrient contents in fortified foods are described. We explain in this paper how to calculate the values of several parameters that should appear in the legislative frameworks of fortified foods. These parameters are: the additional micronutrient contents that should be incorporated into the foods during the fortification process; the final average content and the allowable variation of contents at production centers; and the micronutrient contents to be claimed in the food label. We also discuss the importance of including the description of a sampling system as part of the standards and regulations. Details about physical and microbiological requirements are mentioned but not described in detail, since they are also part of the food standards, irrespective of whether the foods are fortified. Thus, this paper focuses on the fortification specifications for fortified corn flour and maize meal, and presents a model of how to calculate them.

Methodology

Standards and regulations of corn flour and maize meal fortification from countries in Africa and the Americas were collected and reviewed. Emphasis was placed on aspects related to the type and addition of micronutrients, and various ways to express them. Types of legislative instruments and their modes of application (voluntary or mandatory) were noted. In a few countries where data were available, a special effort was made to critically analyze the information on food enforcement activities to assess the degree of compliance of the corresponding standards. Constraints and limitations in how the micronutrient contents are specified in the legislative instruments were identified.

We present a generic model for overcoming the identified limitations of current legislative frameworks; we base our calculations on the nutrient contents that were estimated for maize flour for the population of Kampala, Uganda, and that were based on 24-h dietary recall and the simultaneous use of several fortification vehicles aiming to fill the nutrient gaps. Details of the data and procedures for estimating nutrient gaps and the fortification contents are discussed in Guamuch et al.4

Types of legislative instruments for corn flour and maize meal fortification

Countries use different instruments to legislate food fortification. In a few countries, fortification is directly introduced as part of a law. This is the case in Kenya, where Legal Notice No. 62 of the Food, Drugs, and Chemical Substances Act is used. Most other countries prefer to enact standards or regulations (Table S1).

A law is produced by the legislative branch of a government, and its preparation usually takes a long time. Standards are theoretically under the responsibility of the bureau of standards of the countries, although in the case of foods, they are sometimes produced by government ministries, such as ministries of health, trade, agriculture or others, depending on who is responsible for food control in the country. Regulations are by decree of the executive office of the government through a specific ministry.

Standards are the collection of technical specifications that a product should fulfill, including the generic name of the product; physical, chemical, and microbiological properties; use of acceptable additives; packaging and labeling; and, often, sampling procedures and analytical assays. In principle, standards are not mandatory, but if a producer would like to claim any of the properties mentioned in a given standard, for example, the condition of fortification, the application of the standard becomes obligatory for that product. However, a country could decide if the standard is going to be mandatory, either by stating this in the same standard or issuing a regulation that makes obligatory the application of the standard.

Preparation of a standard requires widespread consultation with all interested parties, for example, producers, public institutions, academia and research organizations, consumer protection groups, and others. The procedure is initiated with the identification of the need to have a standard. Once the decision has been reached to create a standard, a draft is prepared and circulated to members of a relevant technical committee that prepares a document to be circulated for comments to key stakeholders and public scrutiny. The draft standard proposals are disclosed for public opinion for a specific period of time before they are adopted and made official. Most of the countries that have introduced corn and maize flour fortification are using standards (Table S1).

Regulations are legislative instruments that specific ministries can enact on the basis of their responsibility to guarantee public health. A regulation could be as simple as making compulsory a specific standard or as complex as including all the technical requirements of a standard, plus details about roles and responsibilities of several institutions for enforcing it. A regulation could also include penalties and fines for noncompliance. Brazil, Costa Rica, and South Africa followed this mechanism for implementing corn fortification. In summary, regulations describe in detail how to apply laws and standards that are classified as mandatory.

If large and formal factories within a country produce most of the flour or meal, making a standard or regulation mandatory is relatively easy, and producers prefer this because it creates a level playing field for the staple whose branding and specific additional values may not be the deciding factor for the consumer to purchase it. If a large proportion of the flour or meal comes from small and informal enterprises, then mandatory compliance is more difficult to recommend and may be limited to formal and centralized operations or those production centers that fit certain criteria as defined by the government. However, a standard or regulation could still be applicable as mandatory for anyone who would like to claim the condition of fortification regardless of the size or nature of the production center. If fortification is not claimed, then the producer is exempt from using the legislative instrument sanctioned to guide manufacturing and trade of a fortified food.

According to the agreements of the World Trade Organization (WTO), a standard or a regulation could be made mandatory in a country, but before that, the country needs to send to the WTO the public health justification and evidence for requesting such a condition. Otherwise, exporting countries may contest the application of the standard or regulation as an unjustified technical barrier to trade.5 Although laws (acts) are more difficult to produce and adjust, in certain instances they have been valuable. For example, the Philippine Food Fortification Act of 2000 (Ref. 6) spelled out foods whose fortification was to be mandatory on enactment of the Act in 2000, and this included sugar. However, the Governing Board of the National Nutrition Council issued in 2013 a resolution excluding sugar from the list of mandatory foods for fortification in defiance of what is in the Act (Manila Bulletin, April 1, 2013). It took the intervention of the Philippine Minister of Health to have the resolution declared null and void because allowing the council the discretion to remove, exclude, or limit application of mandatory fortification identified by law is tantamount to an amendment of the law by administrative legislation. A similar situation happened in Guatemala at the beginning of the 2000s, when the government tried to eliminate the requirement for fortification in order to allow for the importation of nonfortified product, purely for economic gains and contrary to the public health interest.7 However, the fortification program continued because the law mandated it. In summary, under certain circumstances that place health programs at risk because of weaknesses of government structures, a law can be helpful. On the other hand, the challenge with establishing a food fortification program by enacting a law derives from the difficulty of fully incorporating all of the technical adjustments required to satisfy the law. A strategy to overcome this situation has been the creation of generic laws governing only the type of fortified foods that a country will adopt and reserving the technical specifications for other regulations.

However, it is important to point out here that despite the mandatory nature of any legislative instruments, enforcement has been very weak in most developing countries, to the point that availability of information for confirming compliance is nonexistent or very scarce. This means that without reliable and efficient food control systems, the mandatory nature of a legislative instrument does not achieve its purpose and, therefore, becomes just a collection of technical specifications that a product must conform to only in theory.

Many countries have specific organizations for preparing legislative instruments and for enforcing them. In Ghana, the Ghana Standards Authority develops standards, while the Ghana Food and Drugs Authority implements enforcement of standards for food. Similarly, the Tanzania Food and Drug Authority (TFDA) monitors compliance with standards developed by the Tanzania Bureau of Standards (TBS). Having a different monitoring agency from the standard development agency ensures independent monitoring of compliance, and therefore a more objective way to check not only conformity with the standards but also their validity.

Types of micronutrients added to corn flour and maize meal

Table 1 summarizes the different micronutrients that are added to corn flour and maize meal in the countries whose legislative instruments are listed in Table S1. Regardless of the type of product (precooked flour, flour, meal, or nixtamalized/masa flour), all countries add iron; and most countries, with the exception of Brazil and Tanzania, also add thiamin (B1), riboflavin (B2), and niacin (B3). Most countries also incorporate folic acid, except Venezuela, Nigeria, and the United States. Vitamins A and B6 appear preferentially in African countries, although Venezuela also adds vitamin A to precooked corn flour. The addition of vitamin B12 and zinc is new both in Africa as well as in the Americas, and it reflects the increasing attention to these micronutrients worldwide. In summary, corn flour or maize flour are versatile vehicles for nutrient fortification for most micronutrients. Notable exceptions are calcium (because of the large amounts required) and vitamin C (because it is lost during storage and food preparation). Even vitamin D in dry forms can be added to flour, which highlights the flexibility and potential of flour products as suitable fortification vehicles. Here, it should be noted that the formulations listed in Table 1 reflect the interest in a specific micronutrient when the formulation was developed. For example, vitamin A was included in the Venezuela formulation when that vitamin was of particular interest in the 1990s; similarly, Brazil focused on folic acid and iron during the period that advocates urged that flour should have at least these two micronutrients; and more recently, Central America and the East, Central, and Southern African (ECSA) countries have considered the inclusion of zinc and vitamin B12 now that accumulated evidence shows that the deficiency of these two micronutrients is prevalent in developing countries.

Table 1. Types of micronutrients that are being added or planned to be added to corn flour and maize meal
Countries Thiamine (B1) Riboflavin (B2) Niacin (B3) Vitamin B6 Folate as folic acid (B9) Vitamin B12 Vitamin A Iron Zinc
Precooked corn flour
 Venezuela X X X X X
Corn flour or maize meal
 Brazil X X
 U.S.A. X X X X X
 Kenya X X X X X X X X X
 Malawi X X X X X X X X
 Nigeria X X X X X
 South Africa X X X X X X X X
 Tanzania X X X X X X X X
 Uganda X X X X X X X X
 ECSA X X X X X X X X X
Nixtamalized corn flour
 Central Americaa X X X X X X X
 Costa Rica X X X X X
 El Salvador X X X X X
 Guatemala X X X X X
 Mexico X X X X X
 U.S.A. X X X X
  • a For this paper, Central America includes Costa Rica, El Salvador, Guatemala, Honduras, and Nicaragua.
  • ECSA: eastern, central, and southern Africa region.

Table S2 lists the types of iron sources specified in the legislative instruments for corn flour fortification. Ideally, the iron compound with the better bioavailability (i.e., intestinal absorption) should be preferentially used.8 However, the iron sources with this characteristic (sodium iron ethylenediaminetetraacetate (NaFeEDTA) and ferrous bisglycinate) are not only expensive but also may cause negative changes in the sensorial properties of the flour; therefore, they are used at low amounts, which may be inadequate for populations with low maize flour intake. Ferrous sulfate has a much lower cost than the above two compounds, but its bioavailability is lower and it has stronger negative interactions with the food matrix; therefore it is reserved only when the flour intake is very large and the fat content is low. Ferrous fumarate is intermediate between ferrous sulfate and compounds with best bioavailability. However, even ferrous fumarate was found to discolor precooked flour of Venezuela, and it became necessary to replace part of it with reduced iron.9 Here, it should be pointed out that while the official standard of Brazil states that any of the iron sources listed in Table 1 may be used in fortification, in practice only reduced iron is used by producers because it does not affect the sensorial properties of flour and because it is the cheapest source.10 This example illustrates that a standard should not be “too open”, and in the case of iron, it is important to specify the iron sources that should be used. In any case, it is of prime importance that a given food standards authority must confirm compatibility-recommended iron sources with the food matrix at specified contents. This is the reason why the standards of the ECSA countries left open the possibility of using ferrous fumarate either in combination with or instead of NaFeEDTA.

Use of reduced iron should be discouraged because of its very low bioavailability, which would require very large fortification amounts in order to be effective and yet are not feasible. However, reduced iron is still used in Brazil, the United States, and probably Nigeria, and, partially together with ferrous fumarate, in Venezuela. Reduced iron is also the common source for many other commercial products that claim to be fortified with iron, such as breakfast cereals. Theoretically, electrolytic iron has a bioavailability slightly lower than ferrous fumarate or ferrous sulfate, and has better compatibility with the food matrix.11 However, it is very difficult to differentiate electrolytic iron from reduced iron once it is in fortified food; this is the main reason for the uncertainty in Nigeria, where electrolytic iron is specified in its standard; yet reduced iron has a lower cost and it is chemically indistinguishable from electrolytic iron, which is why the use of electrolytic iron is not considered an attractive option for staple fortification.12

In summary, fortification of corn flour with iron is still an area of research and technological development; it is not wise to recommend or introduce a specific iron compound at a specific content without confirming first that those conditions are compatible with the type of products produced and consumed in each country. Organoleptic properties are crucial in determining how much of a given nutrient can be added, especially in the case of some nutrients such as iron. Bioavailability of the nutrient compounds also plays a role in determining how much should be added in order to address a specified nutrient need. Developments in the science of fortification will continue to influence the type and amount of nutrients that can be added to address various nutrient needs.

Micronutrient contents specified in standards and regulations for corn flour and maize meal fortification

Table S3 summarizes different parameters associated with micronutrient contents as presented in existing legislative instruments. It is clear that apart from the many different content values there are various ways of expressing the micronutrient contents, a condition which may not only be confusing but also could promote erroneous practices. For example, in Venezuela it is not clear if the minimum, mean, and maximum contents were calculated taking into consideration the intrinsic content of the micronutrient in the unfortified flour or only the added amounts. As this is not specified, it is assumed that the values consider the intrinsic content, and therefore the amount of micronutrient to add is going to be less than the values presented in the standard. The amount added would be just enough to comply with the requested content, which includes the intrinsic amounts of the micronutrients present in the unfortified flour. As a minimum value is specified, it is then sufficient that the food complies with the minimum content, which further reduces the amounts of micronutrients to be added.

In the case of Brazil, the amounts requested to be added are equivalent to the value of the minimum content. When this amount is added to the intrinsic content, the final value is likely to comply with the minimum value, especially if the variation of the added micronutrient is not wide. In such a case, the average content would represent the added amount plus the intrinsic content of the micronutrient in the unfortified food. However, this is not explicit in the regulation, and therefore the producers could interpret the minimum values as averages and then subtract the amount of micronutrients already present in the unfortified food, contrary to the intentions of the standard developers.

The fortification values of maize meal in the United States suggest that they have included the intrinsic contents; otherwise they will not be as high as they are in the standard.

In South Africa, the regulation only specifies means, and the values suggest that they represent only the added amounts and not the final values if the intrinsic micronutrient contents were taken into consideration. Otherwise the niacin content in the standard should not have been lower than the intrinsic content. In this case, there is no need to add any niacin to the flour. Likewise, if using this interpretation, the additional amounts of other micronutrients will be lower too: for example, only 1.5 mg/kg of folic acid will be needed instead of 2.0 mg/kg of folic acid to reach the mean value of 2.0 mg/kg of the standard, because the unfortified flour already has 0.5 mg/kg of folate. Similarly, only 26 mg/kg iron will be needed instead of 35 mg/kg, because the unfortified flour already contains 9 mg/kg of iron.

A similar situation as that in South Africa occurs in Nigeria. In Nigerian standards, the mean contents of niacin and iron are below or very near to the intrinsic contents of these two micronutrients in the unfortified flour. As a consequence, the producers could eliminate or add lower amounts of those two micronutrients and they would still comply with the standard. The requested content of vitamin A is very high in Nigeria, which increases the cost of fortification to a level that could make fortification unattractive and unfeasible.

The ECSA countries have more detailed standards in which not only the added amounts are specified but also the minimum, mean, and maximum contents. In these standards, the intrinsic contents of the micronutrients in the unfortified flour were considered for establishing the other parameters in the standards. This is supported by the fact that the mean values approach the combined values of the added and intrinsic contents in the unfortified food, except of niacin. Perhaps, in this case, the assumed intrinsic content of niacin was lower than the value presented in the table. Here, it is also important to point out that because of the fact that iron from NaFeEDTA can be specifically assessed in fortified flour,13 the ECSA's standard distinguishes between added iron and total iron when NaFeEDTA is used as the source of this mineral.

The proposed standard of Central America has the peculiar characteristic that the minimum values coincide with the added amounts plus the intrinsic content of the micronutrient in unfortified flour. It is also important to point out here that the intrinsic contents are average values and not minimum values, and therefore a large portion of samples may be found with contents lower than the specified values despite the fact that the correct added amounts of the micronutrients are used. The fortification contents of the individual countries of Central America are similar to the proposed standard for the region, but they mainly focus on the minimum content. This is a risky decision because the producers could interpret the specification of contents as mean values instead of minimum contents. In the standard of Mexico, like that of Brazil, the added amounts appear as the minimum contents.

Despite the fact that the United States does not have a fortification standard for nixtamalized (masa) corn flour, the Food Composition Table14 from the U.S. Department of Agriculture (USDA) shows the nutrient contents of both unfortified and fortified flour. The mean contents of the fortified masa flour suggests that the micronutrient amounts that are added are higher than in any other country, except for folic acid, which apparently is not being incorporated into this flour yet.

Current inspection practices for enforcement of fortification standards

With the collection of legislative instruments, we tried to obtain results of micronutrient contents coming from inspection activities in various countries in order to assess the degree of conformity with the standards. This information is nonexistent or unavailable in most cases, and only accessible for a few micronutrients, principally iron. Tables 2-4 show the results for Venezuela, Costa Rica, and El Salvador, respectively.

Table 2. Micronutrient contents (mg/kg) measured in fortified precooked corn flour in Venezuela, 1995
Parameter Thiamine (B1) Riboflavin (B2) Niacin Total iron
Content in the standard (range) 2.0–5.0 1.6–4.0 33–82 30–80
 Measured content (P10–P90)a 1.5–4.1 0.8–3.6 13–71 31–63
 Average content ± SD 2.8 ± 1.0 2.2 ± 1.1 42 ± 23 46 ± 13
 Intrinsic iron content in unfortified productb 0.7 0.6 27 10
Estimated added amountc 2.1 1.6 15 36
 Percent samples below the minimum 31% 29% 35% 11%
 Coefficient of variation 36% 50% 55% 28%
 Number of samplesd 15 15 15 15
  • Source: Instituto Nacional de Nutrición, Venezuela.9
  • a Estimated range based on the mean and standard deviation, and calculating percentile 10 and 90 as the mean minus and plus 1.28 SD, respectively.
  • b Except the content of iron that comes from the Venezuelan Food Composition Table, the other values are for unfortified corn flour obtained from the U.S. Department of Agriculture Food Composition Database.14
  • c Subtracting the intrinsic content in the unfortified flour to the average measured content in the fortified product.
  • d These samples are composite samples made by mixing 10 kg each of 10 single samples.
Table 3. Micronutrient contents (mg/kg) measured in fortified nixtamalized corn flour in Costa Rica, 2011–2012
Folic acid Total iron
Parameter 2011 2012 2011 2012
Content in the standard (minimum) 1.3 1.3 22 22
 Measured content (P10–P90)a 1.1–2.1 0.6–1.6 28–52 21–56
 Average content ± SD 1.6 ± 0.4 1.1 ± 0.4 40 ± 9 38 ± 14
 Intrinsic iron content in unfortified productb 0.3 0.3 15 15
Estimated added amountb 1.3 0.8 25 23
 Percent samples below the minimum 25% 69% 2% 12%
 Coefficient of variation 25% 36% 23% 37%
 Number of samples 30 23 30 23
  • Source: Jennifer Lee, Standardization and Control United, Regulation of Products with Interest for Health, Ministry of Health, Costa Rica.
  • a Estimated range based on the mean and the standard deviation, and calculating percentile 10 and 90 as the mean minus and plus 1.28 SD, respectively.
  • b Subtracting the intrinsic content in the unfortified flour to the average measured content in the fortified product.
Table 4. Micronutrient contents (mg/kg) measured in fortified nixtamalized corn flour in El Salvador, 2003–2012
Thiamine—2003 Total iron
Parameter Brands A, C, D Brand B 2010 2011 2012
Content in the standard of 2003—thiamine and 2008 iron (minimum) 3.3 3.3 40 40 40
 Measured content (P10–P90)a 1.9–2.6 4.6–6.7 34–68 37–65 40–65
 Average content ± SD 2.2 ± 0.3 5.6 ± 0.8 51 ± 13 51 ± 11 52 ± 10
 Intrinsic iron content in unfortified productb 2.0 2.2 15 15 15
Estimated added amountb 0.0 3.4 36 36 37
 Percent samples below the minimum 100% 0% 20% 16% 12%
 Coefficient of variation 14% 14% 20% 22% 19%
 Number of samples 12 4 ? ? ?
  • Source: Ana Lila de Urbina and Celia de Hidalgo, Environmental Health Unit; Mayra de Vela, Max Bloch Central Laboratory, Ministry of Health, El Salvador.
  • a Estimated range based on the mean and the standard deviation, and calculating percentile 10 and 90 as the mean minus and plus 1.28 SD, respectively.
  • b Subtracting the intrinsic content in the unfortified flour to the average measured content in the fortified product.

By examining the data in Tables 2-4, it is clear that the food industry is aiming to comply with the minimum of the standards, which for them represents the average content to add. In Venezuela, half of the requested minimum content of niacin was added, probably assuming that the unfortified flour has already a good intrinsic amount of this micronutrient. In El Salvador, samples A, C, and D (Table 4) were not fortified at all with thiamine in 2003, and this was revealed when the intrinsic content of thiamine was subtracted from the measured average content. Furthermore, sample B showed that exactly the amount specified as the minimum content in the standard was added. Perhaps for Costa Rica and El Salvador, the amount decided to be the added contents were transferred to the standard as the minimum contents, and therefore the fortification programs in those countries are satisfying the biological goal. However, as these contents were expressed as minimum values in the standards, 25–69% of samples for folic acid in Costa Rica and 12–20% of samples for iron in El Salvador could be found as not conforming to the corresponding standard. This situation can cause difficulties between the food control authorities and the flour producers, because despite the fact that the producers are complying with the fortification process, the analytical results are going to suggest partial incompliance with the standard.

In Costa Rica, a small percentage of samples were found to have below the minimum required content of iron (Table 3), because the intrinsic content of iron in the unfortified flour is high. However, if only the added iron is measured, nearly 50% of samples are found to be noncompliant. For the same reason, in Venezuela between 11% and 35% of samples were found to contain under the required minimum contents of iron, thiamine, riboflavin, and niacin (Table 2). The case of Venezuela is interesting because the nutritionists who designed the program were probably thinking that the industry was going to add the average of the range they specified, but the industry selected the minimum value as its average of addition. We are aware that this situation is happening with other fortified foods in other countries in the world; it demonstrates that the use of minimum values is inappropriate for food fortification standards.

Tables 2-4 also show that a large variation (coefficients of variation between 15% and 55%) of the analytical results for the micronutrient contents is normal, even when using composite samples. It is, therefore, logical to consider a wide range of content variation in the standards. This has been a matter of concern from nutritionists who do not realize this fact and sometimes are unhappy that the standards allow for large variations. This misunderstanding about the large natural variation in the micronutrient contents when one powder (the fortificant premix) is added to another powder (the flour) has resulted in the adoption of a minimum content, under the assumption that this is the amount everyone is going to comply with. In nutrition epidemiology, the essential parameter is the average additional amounts of micronutrients that are being supplied, not that single samples of the fortified food reach a specified minimum content. The variation of contents around the mean (an allowable range) is a parameter for food control, as it is going to be explained when we present a model to design standards for fortification, but it is useless for estimating the contribution of fortified foods in human nutrition. Here, it is also important to point out that when food composition databases are used, it is the average value that is used for computing micronutrient intakes, not the variation of the estimations from where those averages come from.

Because of the various factors that affect the variation in nutrient levels at different production facilities, the average concentration becomes the more important parameter for determining compliance. However, efforts must be made to reduce the variation in the contents and to maintain most results close to the target average concentration. This is achieved by having effective and efficient standard operating procedures for maintaining quality and ensuring adequate documentation of various quality assurance and quality control procedures. Training of personnel is critical to ensure that processors have the capacity to internally monitor themselves adequately so that any deviations from the set standards can be corrected immediately, maintaining fortified food that is consistent in quality at all times.

Other specifications included in the standards and regulations

A standard for fortified corn flour or maize meal will contain many other parameters in addition to the details of micronutrient contents.15 These other aspects are either presented in complementary standards, as is the case in the United States, Central America, El Salvador, and Venezuela, or they appear in the same and comprehensive standard, as is the case in Mexico and the ECSA countries. In the first case, the complementary stipulations are referred to in the standards that are specific for the fortification characteristic. Elements that are commonly included in food fortification standards or their linked documents are:
  1. Generic name of the product.
  2. Description about how the product is processed and manufactured.
  3. Purity parameters.
  4. Physical characteristics: specified particle size, maximum humidity.
  5. Maximum allowable amounts of filth and insect remains.
  6. Maximum allowable amounts of chemicals (heavy metals), natural toxins, pesticide residues, and microbiological contaminants.
  7. Sources of the micronutrients (specific fortificants, for example, stating the iron source from ferrous fumarate).
  8. Allowed additives.
  9. Packaging and labeling, including sometimes a micronutrient panel (amounts of micronutrients per specific weight of the product, accompanied by the corresponding percentage of the recommended nutrient intake, using as reference the values of the adult male as recommended by the Codex Alimentarius, or using the values for a 2000 kcal diet, as a rough average for the family).
  10. Sampling procedures, including number of samples, amount of each sample, and preparation of composite samples (combination of single samples to reduce number of analysis as well as to decrease heterogeneity in the contents of certain substances, such as the micronutrients; a common practice has been mixing 5–10 single samples per composite sample).
  11. Reference analytical assays (physical, chemical, and microbiological).

In the case of the regulations, they also specify the institutions that are responsible for the enforcement, as well as the type of penalties to be applied if noncompliance is found.

A generic model for estimating the fortification parameters of food fortification legislative instruments

The development of a standard for fortification considers nutrient requirements that are relevant to micronutrient needs of the population. The amount of micronutrients to be added is based on the gap in nutrient intake established among a population. This gap takes into account the amount of nutrients from what is considered as the main local diet and any other interventions used to increase nutrient intake, including supplementation and intake from other fortified foods that provide the same nutrients. This existing intake is then compared against the estimated dietary recommended intakes to determine the nutrient gap that should be corrected through consumption of fortified foods, and among them the estimated specific contribution by fortified corn meal or maize flour. Guamuch et al.4 describes in detail the calculations and procedures in order to estimate potential efficacious added contents, which at the same time are safe for the target population. We are using the calculations done for Kampala, Uganda.

Once the type and amount of micronutrients to be added is known, the formulation of the standard should follow a systematic procedure. Thus, first the amounts of micronutrients to be added should consider expected losses from the factory to the consumer and, where necessary, an extra amount (overage) is added to compensate for such losses. Second, the intrinsic content of nutrients should be considered when defining the total average content of micronutrients in the final product and this average is taken as a reference for internal quality control at the production centers. The following step is to estimate an allowable range of variation around the mean, and this is going to depend on the number of single samples incorporated into each composite sample, the nature of the food matrix, the type and amount of the micronutrients, the capacity of the mixing system, and the precision of the analytical assays. For flours, the objective is to define a sampling procedure whose coefficient of variation is lower than 30% under acceptable manufacturing practices. The number of single samples in each composite sample should be just enough to reduce variation under acceptable conditions of production. The acceptable variation of contents around the mean is an important parameter for accepting individual results, both for quality control as well as for inspection by government officials. The aim is to ensure that the analytical average is very near the specified average and not near the lower limit of the allowable range of variation. Good processing systems will have low variations, and those that are just acceptable might have variations that produce results within the allowable range. However, irrespective of the variations, the average values should be similar among all production centers.

For inspection purposes, the sampling framework should replicate the one used in the production center; this means preparing composite samples, handling, and analyzing them in a manner similar to what the factory is doing. Ideally, the standards should specify the procedures and methods. A common mistake has been that government inspection uses results of single food samples, whereas quality control in factories are based on several composite samples, and, as a consequence, the variation of contents in the first case could be too large and therefore the inspection unfairly judges the performance of the factories’ work. It should also be kept in mind when interpreting results from samples collected at retail stores, that the results may be more variable than those collected at factories, owing to the decay and segregation of the nutrients added to the flour during storage and transportation. In other words, the standards should also take into consideration variation up to the market place. Sampling procedures should also be tailored for this level.

Table 5 illustrates the calculations to determine the amounts of each micronutrient that should be added to the fortification vehicle using the recommendations for the nutrient contents as calculated in Guamuch et al.4 The added nutrient content is based on the recommended nutrient content plus an amount to compensate for losses owing from the time it takes to go from factories to the consumer's table; this additional amount is known as overage. Table 5 also shows the estimated costs of weighting of each micronutrient in the overall cost of fortificant premix. The added amount of each micronutrient very rarely appears in standards and regulations, although this is the most important parameter for guiding the food fortification process. The quality control and inspection procedures, as well as the formulation of the micronutrient premix, depend on the quantities of micronutrients that should be added to the fortification vehicle. Therefore, the added amounts should be clearly defined and expressed in the legislative instruments of food fortification.

Table 5. Micronutrient contents to add to maize meal for fulfilling nutrition recommendations—example of Kampala, Ugandaa
Micronutrients Nutritional recommendation (mg/kg) Percent losses from factories to consumers Overage (%)b Amount to add (mg/kg)c Estimated cost (US$/MT)d
Vitamin A 2 20 25 2.5 $1.67
Thiamine (vitamin B1) 8 33 49 11.9 $0.57
Riboflavin (vitamin B2) 5 15 18 5.9 $0.43
Niacin (vitamin B3) 40 15 18 47 $0.57
Pyridoxine (vitamin B6) 0 15 18 0.0 $0.00
Folic acid (vitamin B9) 1.5 20 25 1.9 $0.32
Vitamin B12 0.029 15 18 0.034 $1.38
Iron (from NaFeEDTA)e 30 0 0 30 $1.46
or
Iron (from ferrous fumarate) 60 0 0 60 $0.95
Zinc (from zinc oxide) 60 0 0 60 $0.44
  • a Using recommended micronutrient contents as justified and described in Guamuch et al.4
  • b Overage = (1/1–loss in proportion) – 1.
  • c Amount to add = nutritional recommendation + (nutritional recommendation × overage in proportion).
  • d Using the Food Fortification Formulator and international CIF prices in April 2013.
  • e If the specified amount of iron from NaFeEDTA was found incompatible with the food matrix, then it could be replaced by twice the content of iron using ferrous fumarate for an approximated similar bioavailability.

Table 6 shows the calculation of the quality control parameters of food fortification. The average expected contents of each micronutrient is estimated by summing the added micronutrient contents plus the intrinsic average content in the unfortified food. This value becomes the target content that the producers should aim for and the average value that the analytical results should approach. As the quality control procedures have variability, it is always appropriate to estimate the allowable variation range, as shown in Table 6. The lower and upper limits of that range are only reference points for interpreting the meaning of single results. The added amounts of micronutrients should conform to the average, not to the extreme values of the allowable variation range. The legislative instruments should also describe a sampling framework that is going to be used to check for compliance; for example, the number and amounts of single samples that should make up a composite sample, the quantity of the composite sample that is going to be subjected to analytical examination, and the type of assays that are going to be used. All these details are frequently missed, but they are indispensable for enforcing the technical specifications.

Table 6. Micronutrient contents for food quality control of fortified maize meal in Kampala, Ugandaa
Allowable variation range (mg/kg)b
Micronutrients Amount to add (mg/kg) Intrinsic content in unfortified meal (mg/kg) Average content at factories (mg/kg) Lower limit Upper limit
Vitamin A 2.5 0.0 2.5 1.7 3.3
Thiamine (vitamin B1) 11.9 0.7 12.6 8.6 16.7
Riboflavin (vitamin B2) 5.9 0.6 6.5 4.4 8.6
Niacin (vitamin B3) 47 27 74.1 50 98
Pyridoxine (vitamin B6) 0.0 1.0 1.0 0.7 1.3
Folic acid (vitamin B9) 1.9 0.5 2.4 1.6 3.1
Vitamin B12 0.034 0.000 0.034 0.023 0.045
Iron (from NaFeEDTA) totalc 30 9 39 26 52
Iron (only from NaFeEDTA)d 30 0 30 20 40
Zinc 60 4 64 43 85
  • a Using amounts to add of each micronutrient as estimated in Table 5.
  • b A coefficient of variation of 25% is assumed for all micronutrients, although values could be different depending on the micronutrient, the number of single samples in each composite sample, and the precision of the analytical assays. The lower and upper limits coincide with percentiles 10 and 90, respectively, and they were calculated by subtracting or adding to the average the same average × CV/100 × 1.28.
  • c If NaFeEDTA is found incompatible with the food matrix, it could be replaced by twice the amount of iron from ferrous fumarate for an approximated similar bioavailability.
  • d The amount of iron included here is only from NaFeEDTA, because there are analytical assays that could determine only the iron coming from this source.

Finally, the nutrient contents to be presented in the label of the fortified product should be estimated. The average nutrient contents at factories could be used for this purpose, or, if a more refined figure is desirable, the estimated average content at retail stores could be used after taking into account losses between the factories and the market. For the latter, Table 7 shows the calculations where the average content at retail stores is defined as the amount resulting from a subtraction of half of the loss anticipated between production and retail from the average content. For example, in the case of riboflavin, the average content is 6.5 mg/kg whereas the loss between production and retail is 10%. By subtracting 0.3 mg/kg (5% of 6.5 mg/kg) from the average content, the label content becomes 6.2 mg/kg.

Table 7. Micronutrient contents for labeling of fortified maize meal in Kampala, Ugandaa
Micronutrients Average content at factories (mg/kg) Percent losses from factories to retail stores WHO–RNI recommended nutrient intakes for adult malesb Average content at retail stores (mg/kg)c Percent RNI per 100 g product
Vitamin A 2.5 15 600 μg 2.3 38
Thiamine (vitamin B1) 12.6 15 1.2 mg 11.7 97
Riboflavin (vitamin B2) 6.5 10 1.3 mg 6.2 48
Niacin (vitamin B3) 74.1 10 16 mg 70.4 44
Pyridoxine (vitamin B6) 1.0 10 1.3 mg 0.9 7
Folic acid (vitamin B9) 2.4 15 400 μgd 2.2 94d
Vitamin B12 0.034 10 2.4 μg 0.032 133
(Intrinsic iron) (9) (0) (14.0e) (9) (6)
(Iron only from NaFeEDTAf) (30) (0) (7.0g) (30) (46)
Total iron 39 0 39 49h
Zinc 64 0 27.4e 84.0 31
  • a Using amounts to add to each micronutrient as estimated in Table 5.
  • b Source: World Health Organization (WHO)/Food and Agriculture Organization of the United Nations (FAO), Vitamin and mineral requirements in human nutrition, 2nd edition, 2004. The recommended nutritional intake (RNI) values of adult males are used as the reference for labeling following the usual practice in the Codex Alimentarius.
  • c It is assumed that at retail stores the probable average will be intermediate between the average content at factories and that average less than half of the percentage of losses from factories to retail stores. These values could be used for labeling purposes in the nutrient panel, together with the percentage of the recommended nutrient intake that 100 g of the product is supplying.
  • d Folate is described in terms of dietary folate equivalents. As folic acid is more bioavailable than dietary folate, the supply as folic acid is multiplied by 1.7 before estimating the nutritional equivalent of this nutrient.
  • e For diets that have low bioavailability for minerals.
  • f The amount of iron only from NaFeEDTA is included here, because there are analytical assays that could determine only the iron coming from this source.
  • g As iron from NaFeEDTA has twice the bioavailability of intrinsic iron, the RNI has been divided by 2.0.
  • h This amount includes a 43% supply of iron from NaFeEDTA and 6% from the intrinsic iron in the unfortified flour. If iron comes from ferrous fumarate, then the total amount, including the intrinsic iron, will be combined using the same RNI value—under this condition 14 mg Fe per day for adult males.

Conclusions

The benefits of fortification can only be realized when increases in nutrient intake result from consuming adequately fortified corn flour and maize meal. To sustain the provision of adequately fortified flour to the population, it is important to ensure consistency in the addition of the nutrients by all producers. This consistency comes about by applying standards of fortification that are appropriate to address a nutritional need and are presented in such a way as to facilitate adequate addition and provide guidance for effective monitoring. Nevertheless, a food standard does not guarantee adequate addition unless enforcement is being conducted regularly to trigger corrective action when necessary.

We have shown that the current use of minimum values or even a range for guiding micronutrient contents in fortified foods is not working. This practice has not only introduced confusion and conflicts between producers and food control authorities but, even worse, it has also failed in many instances to supply the desirable amounts of micronutrients to the targeted population. Two basic parameters for successful food fortification are the amount of micronutrient to add and the expected average contents in the fortified food. For the latter, the intrinsic content of micronutrients in the unfortified food plus the average amounts of micronutrients that are being added should be calculated. For labeling and enforcing purposes, the expected average content at factories could be used, or, if a consideration for losses during distribution on the market is done, an adjusted average content at retail stores could be adopted.

There is variation in the determination of micronutrient contents owing to many factors, such as the nature of the food matrix, the amount and nature of the fortificants, performance of the mixing equipment, the number of single samples incorporated into composite samples, precision of the analytical assays, and others. As such, legislative instruments should clearly describe the sampling framework and procedures that are going to be used for checking compliance. These details could be incorporated into the text of the relevant standard or placed at the end as appendices. We emphasize here that the most important value for checking fortification compliance is the average: neither a minimum nor the lower and upper limits of a range are valid parameters for enforcement especially when using a few results. Monitoring variation is important as a quality assurance and control practice, and for interpreting the distribution of the results coming from single samples, but this should not have prominence over the estimation of the average content.

The model for estimating fortification parameters that we have described may appear complex at the beginning but in reality these are the steps that should be systematically followed in order to prepare clear, objective, and fair fortification standards. Clear standards facilitate good working relationships between producers and food control authorities, and support the achievement of goals of nutritional programs that strive to use food fortification as a strategy in public health nutrition.

Standards and regulations are important as they are guiding documents designed to ensure meaningful, increased nutrient intake. If they are well designed, they facilitate effective monitoring and enforcement of food fortification for adequacy and implementation of corrective action when necessary. Standards provide specification for the final product but need legal backing from regulations that are based on laws of a given country or state. Regulations could make standards mandatory and should provide adequate information on roles and responsibilities, over and above providing clarity of added micronutrients.

Acknowledgments

We thank the following persons for providing us with unpublished results and valuable information: Jennifer Lee, Unidad de Normalizacion y Control (Standardization and Control Unit), Direccion de Regulacion de Productos de Interes Sanitario (Regulation of Health Interest Products), Ministry of Health, Costa Rica; Ana Lila de Urbina and Celia de Hidalgo, Unidad de Salud Ambiental (Environmental Health Unit), Ministry of Health, El Salvador; Mayra de Vela, Laboratorio Central “Max Bloch” (Central Laboratory “Max Bloch”), Ministry of Health, El Salvador; Aldo Heladio Verver y Vargas Duarte, Mary Carmen Fernandez Dosal, and Jose Antonio Jimenez, Comision Federal para la Proteccion contra Riesgos Sanitarios (COFEPRIS), Federal Comission for the Protection against Health Risks, Ministry of Health; Dr. José Félix Chávez from Venezuela; Dr. Rubén Grajeda from the Pan American World Health Organization (PAHO)/World Health Organization (WHO).

This manuscript was presented at the WHO consultation “Technical Considerations for Maize Flour and Corn Meal Fortification in Public Health” in collaboration with the Sackler Institute for Nutrition Science at the New York Academy of Sciences and the Flour Fortification Initiative (FFI), convened on April 8 and 9, 2013 at the New York Academy of Sciences in New York City. This article is being published individually, but will be consolidated with other manuscripts as a special issue of Annals of the New York Academy of Sciences, the coordinators of which were Drs. Maria Nieves Garcia-Casal, Mireille McLean, Helena Pachon, and Juan Pablo Peña-Rosas. The special issue is the responsibility of the editorial staff of Annals of the New York Academy of Sciences, who delegated to the coordinators preliminary supervision of both technical conformity to the publishing requirements of Annals of the New York Academy of Sciences and general oversight of the scientific merit of each article. The workshop was supported by the Sackler Institute for Nutrition Science at the New York Academy of Sciences and the Flour Fortification Initiative (FFI). The authors alone are responsible for the views expressed in this article; they do not necessarily represent the views, decisions, or policies of the institutions with which they are affiliated or the decisions, policies, or views of the World Health Organization. The opinions expressed in this publication are those of the authors and are not attributable to the sponsors, publisher, or editorial staff of Annals of the New York Academy of Sciences.

    Conflicts of interest

    The authors declare no conflicts of interest.