| Household Daily Intake Summary | ||||
| Median and interquartile range (IQR) by food group | ||||
| Metric | Total Foods | Maize Only | Plant-based1 | Animal-based1 |
|---|---|---|---|---|
| Households (N) | 10741 | 10583 | 10575 | 10259 |
| Grams (median) | 3073.5 | 782.1 | 2484.2 | 474.7 |
| Grams (IQR) | 1725.4 - 5479.2 | 265.2 - 2207.8 | 1364.1 - 4521.1 | 231.9 - 856.8 |
| Energy - kcal (median) | 7359 | 2855 | — | — |
| Energy - kcal (IQR) | 3904 - 13578 | 968 - 8058 | — | — |
| Protein - g (median) | 215.1 | 73.7 | — | — |
| Protein - g (IQR) | 114.2 - 396.5 | 25 - 208 | — | — |
| Iron - mg (median) | 63.79 | 21.2 | — | — |
| Iron - mg (IQR) | 33.75 - 118.17 | 7.19 - 59.83 | — | — |
| Zinc - mg (median) | 37.56 | 17.29 | — | — |
| Zinc - mg (IQR) | 18.63 - 74.55 | 5.86 - 48.79 | — | — |
| 1 Plant-based and animal-based datasets only track weight (grams), not full nutritional composition. | ||||
Intake Calculation
Module 1: Nutrient Intakes
Overview
Household consumption surveys like ENCOVI report food acquisition at the household level, while nutrient intake analysis requires individual-level estimates. This document transforms household food consumption into individual nutrient intake through three key steps:
- Standardize maize products — Convert tortillas and flour to grain equivalent weight to isolate the fraction affected by biofortification.
- Separate food sources — Calculate intake from maize and non-maize sources independently, enabling scenario modeling where only maize nutrients change.
- Distribute to individuals — Apply Adult Male Equivalent (AME) factors to allocate household consumption to each member proportionally to their energy requirements.
The separation of maize from other food sources is fundamental to biofortification impact modeling. Because biofortification only affects specific nutrient levels in maize, we need to know precisely how much of each person’s iron, zinc, and protein intake comes from maize versus other foods.
Household Food Processing
Maize Standardization
In Guatemala, maize is rarely consumed as raw grain. Most households eat tortillas — the staple food prepared through nixtamalization — or use commercially milled maize flour. To assess the impact of biofortification, we must convert these processed products back to their maize grain equivalent.
Following INCAP guidelines for dietary intake assessment in Guatemala, weight conversion factors are applied:
| Product | Factor | Rationale |
|---|---|---|
| Tortillas | ×0.52 | Nixtamalization adds water and calcium hydroxide (cal); ~52% of final weight is maize grain |
| Maize flour | ×0.98 | Dry milling retains ~98% of grain weight |
| Raw grain | ×1.00 | No conversion needed |
After weight conversion, nutrient values are reassigned to match raw white maize composition from the INCAP Food Composition Table (per 100g: 365 kcal, 9.42g protein, 2.71mg iron, 2.21mg zinc). Nutrient calculations then reflect the maize fraction that biofortification modifies.
TipRaw Maize Composition After Weight Conversion
Tortilla composition carries nutrients contributed by the nixtamalization process (calcium from cal) and water content that dilutes nutrient density. Applying raw maize values to the converted grain-equivalent weight isolates the component of a tortilla that biofortification modifies. Total dietary energy intake is unchanged by this step: the conversion identifies the maize-derived nutrient fraction within the household diet.
Source: Conversion factors derived from INCAP dietary assessment protocols.
Separating Maize from Non-Maize Sources
Biofortification increases the micronutrient content of food crops through conventional plant breeding. When farmers adopt biofortified maize varieties, only the maize-derived portion of dietary intake changes — nutrients from beans, eggs, vegetables, and other foods remain constant.
To model intervention scenarios accurately, we calculate intake separately for:
- Maize sources: Grain, tortillas, and flour (codine 7, 17, 21).
- Non-maize sources: All other foods (calculated by subtraction).
Plant-Based and Animal-Based Foods
The proportion of plant-based foods in the diet is a descriptor of dietary composition at the household level. It is computed here and used in downstream analyses to characterise the dietary patterns of the target population.
Plant-based foods (vegetal_cereal == "si") include cereals (maize, rice, wheat), legumes (black beans, lentils), vegetables, fruits, tubers, and oils.
Animal-based foods (vegetal_cereal == "no") include meat (beef, pork, chicken), dairy products (milk, cheese), eggs, and fish.
These datasets track weight only (grams per day), not full nutrient profiles: they are used to derive the plant-based percentage of each household’s diet, which is reported as a descriptor of dietary composition.
Household Daily Intake
NoteData Processing Details
- Unit conversions: All food quantities converted to grams using expert-provided equivalence table.
- Recall period: Daily intake calculated by dividing 15-day recall period by 15.
- Edible portion: All calculations include adjustment for edible portion (
pct_aprov) to account for the inedible portion of food (peels, bones, etc.). - Food groupings: Total (all foods), Maize (grain, tortillas, flour), Plant-based (cereals, grains, vegetables, fruits), Animal-based (meat, dairy, eggs, fish).
Individual Intake Distribution
ENCOVI reports what the household acquired, not what each person consumed. Nutrient intake, requirements, and subsequent health outcomes, however, are individual-level phenomena. Aggregate household consumption data masks individual-level variation — a 3-year-old and an adult man have very different nutrient requirements that simple per-capita division would not capture.
Adult Male Equivalent (AME)
Individual nutrient intake distribution requires assigning each household member a fraction of household intake proportional to their energy requirements. This is done via Adult Male Equivalent (AME) factors, calculated on the synthetic edad_meses variable produced by Estimate Age in Months. Month-level resolution is what allows AME to be assigned within the 6-11 month range, the complementary feeding window.
The Adult Male Equivalent (AME) method assigns each household member a factor proportional to their energy requirements relative to an adult male aged 18-30 years (AME = 1.0) (Weisell & Dop, 2012). Household food intake is then distributed according to each member’s share of total household AME.
The AME factors used here follow INCAP’s dietary recommendations for Guatemala (Recomendaciones dietéticas diarias del INCAP, 2012):
| Age Group | Males | Females |
|---|---|---|
| 0 – 5 months | 0 | 0 |
| 6 months – 1 year | 0.21 | 0.21 |
| 1 – 2 years | 0.27 | 0.27 |
| 2 – 4 years | 0.37 | 0.37 |
| 4 – 7 years | 0.44 | 0.44 |
| 7 – 10 years | 0.56 | 0.51 |
| 10 – 14 years | 0.73 | 0.65 |
| 14 – 18 years | 0.96 | 0.73 |
| 18 – 30 years | 1.00 | 0.74 |
| 30 – 60 years | 0.95 | 0.74 |
| 60+ years | 0.76 | 0.65 |
For example, in a household with a 25-year-old man (AME = 1.0), a 25-year-old woman (AME = 0.74), and a 3-year-old child (AME = 0.37), the total household AME is 2.11. The child’s share is 0.37/2.11 = 17.5% of household food intake.
NoteInfants Under 6 Months
Infants under 6 months receive ame_porcentaje = 0, reflecting the WHO recommendation for exclusive breastfeeding during this period. These infants are not allocated any household food consumption in subsequent calculations, as their nutrient requirements are assumed to be met through breastmilk. The synthetic edad_meses variable enables this precise distinction between infants under 6 months (excluded) and children aged 6-11 months (included with AME = 0.21).
Maize vs Non-Maize Separation
Individual intake is calculated separately for maize and non-maize sources.
ImportantWhy This Separation Matters
This separation is essential for biofortification modeling — only maize-derived nutrients will change in intervention scenarios, while non-maize intake remains constant.
Survey Design
ENCOVI uses a complex sampling design with Primary Sampling Units (UPM) and expansion weights that allow inference to the full Guatemalan population (see Table 2 for the weighted population estimate). All population-level statistics require survey-weighted estimation, with standard errors accounting for clustering within UPMs.
Summary
Sample Characteristics
| Sample Characteristics | |
| Survey-weighted estimates for Guatemala | |
| Characteristic | Value |
|---|---|
| Sample size (individuals) | 46,017 |
| Unique households | 10,964 |
| Departments covered | 22 |
| Population estimate (weighted) | 17.24 million (SE: 0.36) |
Total and Maize Daily Intake
| Total and Maize Daily Intake | |
| Survey-weighted population estimates with 95% confidence intervals1 | |
| Nutrient/Metric | Mean (95% CI) |
|---|---|
| Food weight (g/day) | 1165.6 (1108.1 - 1223.1) |
| Energy (kcal/day) | 3204 (3006 - 3402) |
| Protein (g/day) | 90.1 (84.9 - 95.3) |
| Iron (mg/day) | 26.51 (25 - 28.03) |
| Zinc (mg/day) | 18.05 (16.85 - 19.25) |
| Maize weight (g/day) | 613.7 (559.7 - 667.7) |
| 1 Estimates account for complex survey design (clustering, weights). | |
Maize Contribution to Nutrient Intake
The table below shows mean daily intake for key nutrients, separated by source. The “% from Maize” column is the biofortification-modifiable fraction: the share of intake that would increase if households consumed biofortified varieties. For protein, biofortification does not change the total protein content of the grain — it changes the amino acid profile (lysine, tryptophan), and with it the utilizable protein fraction.
| Maize Contribution to Total Nutrient Intake | |
| Percentage of daily nutrients from maize products | |
| Nutrient | % from Maize (95% CI) |
|---|---|
| Energy contribution | 49.6% (48.4 - 50.8) |
| Protein contribution | 46.6% (45.4 - 47.9) |
| Iron contribution | 45.2% (44 - 46.4) |
| Zinc contribution | 55.5% (54.3 - 56.7) |
Intake by Age Group
The table below describes the distribution of mean daily nutrient intake across age groups under the Adult Male Equivalent allocation, where household food is distributed to each member in proportion to energy requirements.
| Daily Nutrient Intake by Age Group | |||||||||||
| Survey-weighted population estimates | |||||||||||
| Age Group | N | Energy (kcal) | Energy (SE) | Protein (g) | Protein (SE) | Iron (mg) | Iron (SE) | Zinc (mg) | Zinc (SE) | Maize Energy (%) | Maize Energy (SE) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| de 0 a 5 meses | 492 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
| de 6 meses a 1 año | 434 | 912.7 | 81.9 | 25.3 | 2.2 | 7.6 | 0.6 | 5.2 | 0.5 | 53.6 | 1.9 |
| de 1 a 2 años | 881 | 1,237.3 | 104.2 | 34.8 | 2.7 | 10.5 | 0.8 | 7.0 | 0.6 | 51.2 | 1.4 |
| de 2 a 4 años | 1,832 | 1,859.9 | 145.7 | 51.4 | 3.8 | 15.3 | 1.1 | 10.6 | 0.9 | 53.2 | 1.0 |
| de 4 a 7 años | 3,316 | 2,097.6 | 112.4 | 58.6 | 2.9 | 17.4 | 0.9 | 11.9 | 0.7 | 51.2 | 0.9 |
| de 7 a 10 años | 3,219 | 2,647.6 | 163.1 | 74.0 | 4.3 | 21.8 | 1.2 | 15.2 | 1.0 | 52.4 | 0.9 |
| de 10 a 14 años | 4,261 | 3,063.9 | 191.8 | 85.6 | 5.0 | 25.4 | 1.4 | 17.4 | 1.2 | 52.6 | 0.8 |
| de 14 a 18 años | 3,681 | 3,669.4 | 205.0 | 102.4 | 5.3 | 30.1 | 1.6 | 20.6 | 1.2 | 52.0 | 0.9 |
| de 18 a 30 años | 9,210 | 3,737.8 | 192.8 | 104.4 | 5.0 | 30.8 | 1.5 | 21.1 | 1.2 | 48.9 | 0.7 |
| de 30 a 60 años | 14,259 | 3,574.1 | 111.3 | 101.6 | 3.0 | 29.7 | 0.9 | 20.1 | 0.7 | 47.9 | 0.7 |
| más de 60 años | 4,432 | 3,474.5 | 161.4 | 97.9 | 4.3 | 28.8 | 1.2 | 19.2 | 1.0 | 45.5 | 0.9 |
Diet Composition
Plant-based diets predominate in Guatemala (see Table 6 for the population distribution across plant-based food share categories). The pattern is consistent with limited access to animal-source foods in rural and low-income populations.
| Distribution of Plant-Based Diet Categories | ||||
| Survey-weighted population proportions | ||||
| Plant-based Foods (%)1 | Sample N | Population % | 95% CI Lower | 95% CI Upper |
|---|---|---|---|---|
| 0-50% | 1,229 | 2.8% | 2.3% | 3.2% |
| 50-70% | 5,841 | 13.1% | 11.9% | 14.3% |
| 70-90% | 19,881 | 41.9% | 40.4% | 43.4% |
| 90-100% | 17,072 | 36.6% | 34.7% | 38.5% |
| NA | 1,994 | 5.6% | 4.6% | 6.7% |
| 1 Categories based on percentage of grams from plant-based foods. | ||||
TipPlant Percentage Application
The distribution of the plant-based food percentage across the population is reported here as a descriptor of dietary composition. This distribution is also referenced in the iron absorption decision, where it supports the choice of a uniform 5% iron absorption factor for the target population.
Data Quality
Data completeness is high across key variables. Infants under 6 months are assigned ame_porcentaje = 0 under the exclusive breastfeeding assumption and are excluded from household-level food allocation — these are not treated as missing values in the analysis.
| Missing Values by Key Variable | ||
| Data completeness assessment | ||
| Variable | N | % |
|---|---|---|
| AME percentage | 0 | 0.00% |
| Total intake | 763 | 1.66% |
| Maize intake | 1,259 | 2.74% |
| Plant-based percentage | 1,994 | 4.33% |
Extreme values are present in a non-trivial share of records, consistent with unusual household compositions or recall error. Handling of these values in subsequent analyses is described in Modeling Data Preparation.
| Extreme Value Detection | ||
| Potential data quality issues or outliers1 | ||
| Quality Check | N | % |
|---|---|---|
| Very high energy (>5000 kcal) | 7,060 | 15.34% |
| Very low energy (<500 kcal) | 5,047 | 10.97% |
| Very high protein (>200g) | 3,964 | 8.61% |
| Zero food intake | 1,172 | 2.55% |
| 1 Extreme values may indicate data entry errors, unusual household compositions, or recall bias. | ||
NoteData Quality Considerations
- Infants under 6 months: Have
ame_porcentaje = 0, resulting in zero calculated intake from solid foods under the exclusive breastfeeding assumption. - Missing values: Typically occur for individuals in households with no food consumption data reported. These are genuine missing data, not calculation errors.
- Extreme values: May reflect data entry errors, unusual household compositions (e.g., large families, guests), or reporting period issues (recall bias).
References
Recomendaciones dietéticas diarias del INCAP (Segunda Edición). (2012). Instituto de Nutrición de Centro América y Panamá, INCAP.
Weisell, R., & Dop, M. C. (2012). The Adult Male Equivalent Concept and its Application to Household Consumption and Expenditures Surveys (HCES). Food and Nutrition Bulletin, 33(3_suppl2), S157–S162. https://doi.org/10.1177/15648265120333S203