24-Hour Recall

Overview

The 24-hour recall (24HR) recall is a widely used retrospective recall of dietary items consumed in the day preceding the interview (usually from 6:00 am the previous day to 6:00 am on the survey day. Typically, a trained enumerator asks the interviewee to recall all the foods and beverages consumed over the previous day, including quantities, brand names for packaged foods, preparation, and cooking (food or dish). The 24HR has been studied extensively and validated for its accuracy and reliability against the observed weighed food record (OWFR), often used as a standard for dietary assessment.

Rationale

24HR recalls are widely used to assess nutrient intakes in low- and middle-income countries (LMICs). Data on foods consumed are matched to a food composition database, with single recalls estimating mean intake and repeated recalls on non-consecutive days assessing usual intake (Saravia 2022, Freedman 2018). The enumerator does not observe the meals, so household units (e.g., cups, measuring spoons) and visual aids (pictures, models) are used to help estimate portion sizes and food weights.  

Type of data

The 24HR quantitative recall measures quantity of consumption at the food level (detailed food intake and portion sizes) with food data matched with a food composition database to derive nutrient content.

Indicators

All indicators derived from an observed weighed food record (OWFR) can also be derived from a 24-hour recall (24HR), provided the same conditions are met (e.g., repeat recalls to estimate usual intake). However, the 24HR is typically recommended over the OWFR, as it is less burdensome, more cost-effective, and less likely to influence participants’ eating behavior, whereas the act of weighing foods in real time may alter usual consumption patterns.

If using a quantitative 24HR and a food composition table (FCT): A quantitative 24-hour dietary recall (24HR) is a powerful tool for measuring both macronutrient and micronutrient intake, as well as assessing the adequacy of those intakes relative to established nutritional standards.

Quantitative intake measures (macro- and micronutrients, total energy intake, food group intake)

Quantified daily intake of macro- and micronutrients can be estimated using 24-hour dietary recall data combined with a locally adapted food composition table (FCT). The FCT provides nutrient values for each food reported, allowing calculation of energy, protein, fat, carbohydrates, and essential vitamins and minerals such as iron, vitamin A, and zinc. These intake amounts are then compared to recommended daily requirements based on a person’s age, sex, and physical activity level to assess whether their diet meets nutritional needs. 

Nutritional adequacy indicators (adequacy of energy intake, micronutrient adequacy, mean probability of adequacy (MPA), and macronutrient distribution

These indicators estimate how well a person’s diet meets their nutritional needs using food intake data from a 24HR converted into nutrient intake using an FCT. For individual-level assessment, nutrient intakes are compared to recommended daily requirements based on age, sex, and physical activity. For population-level assessments, multiple 24HRs are used to estimate usual intake and account for day-to-day variation (National Cancer Institute). Nutrient adequacy is then evaluated using methods such as the Estimated Average Requirement (EAR) cut-point method or the probability approach, which estimate the proportion of the population at risk of inadequate intake—though the EAR cut-point method is not suitable for nutrients like iron in women and children due to skewed requirement distributions. To assess overall micronutrient quality, the Mean Probability of Adequacy (MPA) is often used. It combines the probability that a person’s intake of each selected micronutrient—such as iron, zinc, and vitamin A—meets their daily requirement and averages these values to give a single score (Arimond 2010). The MPA is based on dietary intake data to estimate usual intakes and their variability^[1]. A higher MPA means better overall micronutrient adequacy. MPA does not capture excessive intake, so separate indicators are for overconsumption, such as the proportion of individuals exceeding upper intake levels.

Indicators of diet quality

Suggested indicators for tracking healthy and unhealthy food consumption—especially in urbanized settings—include dietary energy density (kilocalories per gram of food), percentage of energy from free sugars and saturated fat, and total energy intake. These data can also be used to assess overall diet quality using tools like the Global Diet Quality Score (GDQS) , which organizes all foods consumed into standardized food groupings. GDQS generates a positive score for healthy foods and a negative score for unhealthy foods. This approach supports monitoring of nutrition transitions, identification of dietary risks, and design of targeted interventions to improve diet quality. Other composite diet quality indicators include Diet Diversity Scores and Minimum Dietary Diversity for Women (MDD-W).

---

^[1] To estimate MPA, collect at least two non-consecutive 24HRs per person on a random subsample, ensuring at least 50 individuals with repeat recalls per nutrient or food group of interest (NCI Diet Primer). If resources are limited, collect one recall from all and a second from at least 30–40%, then use statistical methods to model usual intake and calculate adequacy. For nutrients or foods with high daily variation or infrequent consumption, increase the individual recalls or expand the proportion of the sample with repeat recalls.

If not using a quantitative 24HR and a food composition table: A simplified 24-hour recall can still be useful when the goal is to assess food group consumption rather than nutrient intake. In this case, quantities may or may not be collected. If quantities are included, researchers can estimate the amount of specific food groups consumed (e.g., grams of fruits and vegetables), even without calculating nutrient content. If quantities are not collected, the recall can still provide information on the frequency of consumption of specific food groups, which may be sufficient for certain indicators or for monitoring. This approach is useful when nutrient analysis is not needed but understanding dietary patterns or food group intake is still important.

Minimum Dietary Diversity for Infants and Young Children (MDD-IYC)

The Minimum Dietary Diversity for Infants and Young Children (MDD-IYC) was developed to assess dietary diversity in children aged 6–23 months (WHO/UNICEF 2021). While the original MDD-IYC indicator is validated for this age group, recent research has explored adaptations and proxy indicators for older children and adolescents in LMICs (Diop et al 2025).

Minimum Dietary Diversity for Women (MDD-W)

Similarly, the Minimum Dietary Diversity for Women (MDD-W), based on consumption from at least 5 out of 10 defined food groups (FGS-10), is a validated indicator of better micronutrient intake for adult women (15-49 years) in LMICs (FAO/FHI 2016). MDD-W has also been used in some settings with adult men (Gomez et al 2024) and shown to be a useful proxy for micronutrient adequacy. These indicators are commonly used to evaluate the impact of interventions aimed at improving diet diversity and quality.

Proportion of [population group] that consumed [food group/s] consumed

Researchers may also gain insight from estimating intake of individual food groups. A binary indicator can be generated to measure consumption of a particular food group (yes/no) in an individual’s diet. At the population-level, this can estimate the proportion of the population that consumes the food group.

Tool and indicator validation

To ensure these indicators are reliable and meaningful, validation studies have been conducted across diverse settings and populations.

Validation is essential in determining the suitability of a dietary assessment instrument, focusing on its validity, misreporting and measurement errors. Validity assesses how accurately the instrument reflects actual intake, usually in comparison with other methods. Misreporting, influenced by factors like social desirability or memory limitations, can impact accuracy. Measurement errors, either systematic (bias) or random, affect the reliability of findings. Every dietary assessment method has its own set of potential biases and errors – no method is perfect.

24HRs can underestimate energy intake, especially among overweight individuals in urban areas, but repeat recalls improve accuracy (Nightingale 2016). Validation should cover diverse populations, age groups, and data collection modes, including paper questionnaires, CAPI, phone interviews, and online tools. Phone-based 24HRs offer valid alternatives to in-person interviews (Galasso 1994), as do web-based tools for macro- and micronutrient assessments  (Timon 2016).  Despite some underreporting, 24HRs reliably assess group-level dietary intake.

LMIC children and adolescents, single 24HRs tend to underestimate intake, especially in younger ages, but accuracy improves with age and adaptation. In Burkina Faso, results were acceptable for ages 12–14 (Arsenault 2020). The INDDEX24 mobile tool showed feasibility and reasonable accuracy in Malawi’s low-literacy settings. Studies in Kenya and Bangladesh stress the importance of age-appropriate tools, trained interviewers, and repeat recalls to enhance accuracy.

Among LMIC adults, two 24HRs outperform 7-day diaries and moderately correlate with weighed records, underestimating intake by 6–12%. Examples include Ghanaian adolescents (8–12% underestimation) (Gibson 2020), Ethiopian women (~10%) (Alemayehu et al. 2011), and rural Kenyan women (6%) (Gewa et al. 2008). These findings affirm 24HR’s practicality and validity in LMIC settings when combined with proper protocols and repeated measures.

---

---

Sampling and data collection considerations

The sampling approach depends on the user’s question of interest and target population, but it is crucial to ensure that the study sample is representative of the target population. The two primary sampling approaches are probability and non-probability sampling. There are several methods of probability sampling, including:

• simple random sampling, where any member of the target population has an equal chance of being selected into the study.
• interval sampling, in which people of the targeted group are continually available and selected into a sample (i.e., consumers in a market).
• stratified sampling, which divides the target population into groups for sampling, and/or
• cluster sampling which uses groupings from which the sample population is selected.

In urban settings, administrative boundaries and enumeration areas can help organize sampling. In many countries, lists of enumeration areas can be acquired, after which a sample frame or list of households or targeted individuals from each of those areas is developed to draw the sample of households or individuals. Correcting for over- or under sampling through sample weighting is essential to improve data accuracy. If the question of interest is to assess changes at population-level in dietary quality due to a program or policy, it is critical that the sampling frame include populations that have been exposed to those interventions. Non-probability sampling methods, such as convenience and snowball sampling, can be used when ease of access is prioritized.

Careful conceptualization of the relationship between food environments and diets helps guide geographic focus and sampling strategy, ensuring more meaningful and representative results. For example, if your research question involves comparing diets across diverse levels of urbanization, you might sample households from urban, peri-urban, and rural areas. Alternatively, if you are interested in how distinct types of food environments influence diets, you could stratify your sample based on dominant food environment characteristics (e.g., areas with high density of informal vendors vs. areas with supermarkets or fast-food outlets). The choice between geographic or food environment-based sampling depends on the specific objectives of your study.

A sufficient sample size is essential when using the 24-hour dietary recall (24HR) method to account for usual intake and day-to-day dietary variability. A general rule of thumb is to include at least 200 participants, but this should be increased to 500 or more to account for design effects, non-response, and attrition—especially in urban areas, where busy schedules and high population mobility can lead to greater dropout rates. Oversampling may be necessary to ensure adequate data quality and representation. To estimate usual intake, a second recall on a non-consecutive day is typically conducted for a subsample of participants. This subsample should include at least 50–100 individuals, depending on the study design and statistical power required. In household-based sampling, the second recall is often conducted with a random subset of those who completed the first recall. Repeated recalls improve the precision and reliability of intake estimates, but there is no single “optimal” sample size. There are several ways to increase the number of repeated measures: increase the number of individuals who receive a second recall, increase the number of recalls per person (e.g., 2–3 recalls per participant), or combine both strategies to strengthen the estimation of usual intake. The optimal number of repeated recalls depends on the nutrients of interest, population variability, and available resources. For example, nutrients with high day-to-day variation (like vitamin A or iron) may require more repeated measures to accurately estimate usual intake.

Other data sources

Data Sources	Pros	Cons	Indicators
Global Dietary Database [Individual-level]	-Most are nationally representative -Urban/rural residence included in half of surveys -Harmonized data (variables, units, food definitions) for individual-level dietary data from nutrition surveys for 188 countries	-Requires nutrition and data analysis expertise -Surveys use different designs and tools -Certain food categories are excluded (e.g., poultry, dairy-based desserts, highly processed or packaged foods, mixed dishes and recipes, condiments and spice, supplements)	-Includes 51 dietary factors based on public health and chronic disease risk (e.g., foods, beverages, nutrients, dietary patterns, or qualities)
GIFT Database (FAO)   [Individual-level]	-Data are disaggregated by sex and age but inconsistent urban/rural disaggregation -Individual quantitative food consumption data coded with the FoodEx2 European standardized food classification system to code foods for dietary and exposure assessments, data are screened and formatted using R -Dashboards presenting indicators and summary statistics on foods and diets -Can link food groupings to own dietary data (dataset available upon request)	-Need nutrition and data analysis expertise, particularly as outliers and missing data have not been removed from original datasets and energy and nutrient values are provided directly from surveys (does not link food consumption datasets to food composition data) -Many datasets are old, most not nationally representative and there are no data on statistical weights -Data covers only 36 countries (Africa: Benin, Burkina Faso, Cameroon, Ghana, Kenya, Malawi, Mali, Mozambique, Nigeria, Senegal, South Africa, Tanzania, Uganda; Asia: Bangladesh, Cambodia, China, India, Indonesia, Lao PDR, Nepal, Philippines, Thailand, Vietnam; Latin America: Brazil, Guatemala, Haiti, Mexico, Saint Kitts and Nevis; Europe: Albania, Armenia, Georgia, Moldova, Ukraine; Middle East/North Africa: Egypt, Lebanon, Morocco, Tunisia; Oceania: Fiji)	-Estimated usual intakes of selected nutrients (with SPADE tool) -MDD-W (and Food group diversity score, individual food group consumption) -Food consumption (daily diet g/per person per day, proportion of food groups consumed (%), calories per person per day) -Other indicators for food safety (dietary exposure to chemicals) and environmental impacts of food consumption (emission, water, and land use) -Statistics on food consumption can be calculated for individual food items or using food groups (e.g., sources of micro- and macronutrients in the diet, macronutrient contribution to total intake)
Demographic and Health Surveys [Individual-level]	–Nationally representative data on dietary diversity -Urban/rural disaggregation	-Need nutrition and data analysis expertise -Alternatively, the DHS StatCompiler and mobile app allows for automatic indicator calculation and disaggregation	-MDD-W -IYCF practices (MAD, MDD, MMF) -Percentage consuming ten food groups[2] (PLW, WRA)
Gallup World Poll (GWP) (Global Diet Quality Project) [Individual-level]	-Global coverage and standardization (140 countries, including those that lack nutrition surveillance data) -Integration with economic, social and health indicators -Frequent updates (every 5 years) -Samples adults aged 15+ (not just women)	-Other national surveys tend to align more closely with DHS than GWP. -GWP often collects data in lean seasons, potentially underestimating MDD-W compared to DHS. -Validating MDD-W for males aged 15–49 could expand GWP’s utility. –Greater variability in GWP estimates than DHS.	-MDD-W, DDS -All-5, protective, and unhealthy food consumption -Healthy diet pattern for NCD prevention -Zero fruit or vegetable consumption -Food groups
Household consumption and expenditure surveys (HCES) [Household-level food consumption]	-Implemented at scale often with national representativeness, but made financially viable through public and donor subsidies -Conducted some countries (every 3-5 years) with a large sample -Urban/rural disaggregation -Contains other variables such as data on socioeconomic status, education, and other determinants relevant to nutrition -Often also includes acquisition data (food acquired from purchases, production, in-kind)	-Household level food consumption, does not address intra-household allocation issues that may affect household members – not ideal source of dietary data. -Modules are heterogenous across countries, making comparisons challenging -Does not differentiate between subgroups to estimate differences in probability of deficiencies in high-risk groups -Often lacks information on types of foods consumed away from home (FAFH); the information on cost of meal consumed outside the home may be available, but not the type of foods/meals. This is a problem especially in urban settings where working HH members or school-age children often eat one or more meals outside the home.	-Food variety (Household diet diversity score) -Food consumption (Food consumption score) -Nutrient availability: macronutrient and micronutrient availability per capita per day (micronutrient availability requires use of FCT), per capita energy intake. -Consumption patterns (frequency or shares of animal-sourced foods, staple foods, ultra-processed foods), however, recall periods are variable (some foods reported over a week, others a month)

Illustrative research using these tools and indicators in urban settings

24HR and Dietary Assessment-Related Resources

• Bailey, R. (2021). Overview of dietary assessment methods for measuring intakes of foods, beverages, and dietary supplements in research studies. Current opinion in biotechnology, 70, 91-96.
• Bingham, S. et al. (1994). Comparison of dietary assessment methods in nutritional epidemiology: weighed records v. 24 h recalls, food-frequency questionnaires, and estimated-diet records. British Journal of Nutrition, 72(4), 619-643.
• Data4Diets: Building Blocks for Diet-related Food Security Analysis, Version 2.0. Tufts University, accessed 2023, https://inddex.nutrition.tufts.edu/data4diets.
• Deitchler, M., Arimond, M., Carriquiry, A., et al. (2020). Planning and design considerations for quantitative 24-hour recall dietary surveys in low-and middle-income countries. Intake–Center for Dietary Assessment/FHI Solutions; Intake–Center for Dietary Assessment/FHI Solutions: Washington, DC, USA.
• Freedman, L., Midthune, D., Arab, L. et al. (2018). Combining a food frequency questionnaire with 24-hour recalls to increase the precision of estimation of usual dietary intakes—evidence from the validation studies pooling project. American Journal of Epidemiology, 187(10), 2227-2232.
• Hemsworth, J., Arimond, M., Kumwenda, C., et al. (2018). Comparison of an interactive 24-h recall and weighed food record for measuring energy and nutrient intakes from complementary foods among 9–10-month-old Malawian infants consuming lipid-based nutrient supplements. British Journal of Nutrition, 120(11), 1262-1271.
• INTAKE. Center for Dietary Assessment. FHI 360. https://www.intake.org/resources.
• National Cancer Institute. Dietary Assessment Primer. National Institutes of Health (NIH). https://dietassessmentprimer.cancer.gov/approach/table.html.

References