Development and validation of an online tool for assessing dietary intake, diet quality, and environmental impact in Mexico

Introduction

Diet is a crucial lifestyle-related risk factor for chronic diseases (1) and a primary contributor to environmental degradation (2). Food production is associated with substantial greenhouse gas emissions, land occupation, biodiversity loss, and, notably, high water consumption (2,3). Globally, the agricultural sector accounts for approximately 70% of total freshwater withdrawals, highlighting its significant role in water resource management (3). This disproportionate use has resulted in widespread water scarcity, especially in countries like Mexico, where over 85% of the territory experiences hydric stress. Thus, assessing water usage in food production is essential to mitigate the environmental impact of dietary patterns (4).

The water footprint (WF) is a widely used indicator to quantify the volume of freshwater required to produce a good or service, including food. The WF index consists of three components: green, blue, and grey WF. In food production, green WF represents the volume of rainwater stored in soil and used by crops through evapotranspiration; blue WF refers to surface and groundwater used for irrigation; and grey WF estimates the volume of water required to assimilate pollutants, such as wastewater from food processing and water used during food preparation (e.g., boiling, washing, or cooking). The sum of these three components constitutes the total WF of a product, providing a comprehensive measure of its water resource use (5).

Reliable dietary assessment tools are indispensable for nutritional epidemiology and understanding population dietary intakes, diet quality, and diet-disease relationships (6). Furthermore, they are essential for evaluating the environmental impact of diets and promoting sustainable dietary patterns (7). Traditional dietary assessment methods face challenges regarding validity and reliability, which have been widely discussed in the literature (8-11). Technological advancements have been recognized as effective strategies for improving dietary assessments and reducing data collection time in research (9-11).

Internationally, several online dietary assessment tools have been developed. However, few integrate environmental metrics alongside dietary evaluations (9-12). Contextualization is critical when assessing diet and its environmental impact, as different regions present unique challenges and consumption patterns (8). In Mexico, despite the availability of some digital dietary tools, none incorporate ecological aspects such as the WF (13-15). Additionally, to the best of our knowledge, no existing tool simultaneously assesses dietary intake, diet quality, and environmental impact.

Given the growing need for novel digital solutions that address these gaps, this study aimed to develop and validate Nutriecology^®, an online Mexican tool designed to evaluate diet through a 24-hour recall and a validated Food Frequency Questionnaire (FFQ) and automatically calculate its WF. The software also includes the newly proposed ‘Alternate Mexican Diet Quality Index (IACDMx)’, allowing for automatic scoring through the FFQ. The validation of Nutriecology^® included two studies to assess reliability and reproducibility, both by self-administrating the software and by being applied by a nutrition expert.

This innovative platform provides a unique solution for integrating dietary and environmental assessments, enhancing the efficiency and standardization of data collection and analysis in both individual and population-based studies, offering a comprehensive framework for evaluating food consumption and its environmental impact, promoting the integration of sustainable dietary practices into research and public health policies in water-stressed regions like Mexico.

Methods

Software design

The software, named Nutriecology^®, was developed using the waterfall life cycle methodology, a structured multi-stage process that began in June 2019 and concluded in December 2019, with final testing completed in January 2020. A multidisciplinary team, including software developers, computer engineers, nutritionists, psychologists, and environment experts, collaborated on its creation. The development process consisted of five key stages: (I) formative research, (II) definition of software features, (III) software development, (IV) internal validation testing, and (V) launch of the live site. Since Nutriecology^® was designed specifically for the Mexican context, the entire platform is in Spanish. The software is registered with the National Institute of Copyright (INDAUTOR) under the registration number: 03-2022-012812203100-01.

Stage 1: formative research

A thorough literature review was conducted to identify existing online nutritional and ecological software tools. This review guided the identification of core features to be included in Nutriecology^®. Internationally, the most notable tool identified was myfood24^®, which serves several countries, including the UK, USA, Germany, and others (10,12). In Mexico, existing tools such as Nutrimind^® (15), Aznutrition^® (13), and Nutricloud Nutrición Digital^® (14) were reviewed, though these primarily support nutritional consultations rather than research applications. Notably, Nutricloud^® (14) integrates a validated FFQ, 24-hour recall, and a diet quality index (16), making it the closest reference for developing Nutriecology^®. However, Nutricloud^® lacks environmental impact assessments, which served as a gap that Nutriecology^® aimed to fill. Additionally, diet quality did not consider sustainability dimensions. Moreover, at this stage, informal interviews with nutrition and software experts provided additional insights into desired functionalities.

Dietary assessment tools: 24-hour recall and FFQs

To identify dietary assessment tools available in Mexico, we consulted the Nutritools^® platform, which provides a comprehensive database of international dietary instruments, including over 20 types of 24-hour recalls and 79 FFQs (17). Among these, only one FFQ had been validated specifically for the Mexican context (17). Additionally, other FFQs used at state and national levels in Mexico were identified, such as those developed by Macedo-Ojeda et al. (16) and Denova-Gutiérrez et al. (18), which have been employed for both dietary assessment and WF estimations (19).

Diet quality indexes

Globally, the most widely used diet quality indices are the Healthy Eating Index (HEI) (20,21), the Alternate Healthy Eating Index (AHEI) (22), and the 2015 Dietary Guidelines Adherence Index (DGAI) (23). However, these indices were developed for specific populations outside of Mexico. In Mexico, various indices have been created, including the Mexican Diet Quality Index (MxDQI) (24), the Mexican Alternate Healthy Eating Index (25), and the ICDMx (26). Although all these indices are based on Mexican dietary guidelines, each presents specific limitations. For example, they do not differentiate between meat products, and only the ICDMx includes food sub-groups and has been recently validated (26,27). The ICDMx, integrated into Nutricloud Nutrición Digital^® (the basis for this project) (14), considers the concept of a “correct diet” (except for the adequate dimension), which represents current dietary policy in Mexico (28) and aligns with the following criteria:

• Sufficient: Meets nutritional needs in terms of energy, iron, calcium, fiber, and water.
• Balanced: Proportional intake of macronutrients (proteins, lipids, and carbohydrates).
• Complete: Includes foods from the three main groups: a) vegetables and fruits, b) cereals, and c) legumes and animal products.
• Varied: Incorporates a diverse range of foods within each group, such as fruits and vegetables of various colors (e.g., red: tomatoes, strawberries; green: spinach; yellow-orange: carrots, papaya) and different cereals (e.g., corn, wheat, rice).
• Innocuous: Ensures regular consumption does not pose health risks, particularly regarding fats, sodium, and alcohol.

Adequate: Considers cultural preferences, tastes, and food affordability. This last criterion, however, is not currently included in the ICDMx (26,27).

The ICDMx provides a maximum score of 100 points, where higher scores indicate better diet quality. Diet quality can be calculated using an FFQ or a 24-hour recall, although for this study, only the FFQ was used due to its robustness in assessing habitual dietary intake (8).

Despite the strengths of Nutricloud Nutrición Digital^®—such as its inclusion of a 24-hour recall, FFQ, and the ICDMx—the software has notable limitations. The principal issue is that the included FFQ omits several foods commonly consumed in the traditional Mexican diet, such as pozole, tacos, sopes, chilaquiles, and enchiladas (5,19). Moreover, the ICDMx has significant limitations, particularly in the aspects of being “complete,” “innocuous,” and “adequate.” The “complete” component does not differentiate adequately between animal-based and plant-based foods, evaluating them at the same level, despite their varying nutritional compositions (26,29). For instance, fish and legumes have been associated with reduced risks of chronic diseases (30,31). While red meat consumption has been linked to higher cancer risks (32). Additionally, from an environmental perspective, the WF of red meat is up to six times higher than that of fish and legumes. For example, a kilogram of fish has a WF of 3,110 liters, beans 5,789.87 liters, and beef 21,566 liters (5,33).

Furthermore, the “innocuous” component of the ICDMx does not consider limits on sugar intake, even though excessive sugar consumption is a well-known risk factor for chronic disease development (34,35). The index also fails to incorporate the “adequate” criterion, which evaluates whether a diet meets the economic and cultural needs of its consumers (26,27). This criterion is fundamental in defining sustainable diets, which should integrate health, environmental, economic, and social dimensions (36).

Environmental impact of diet

In terms of environmental impact assessment tools for dietary analysis, the most prominent software solutions identified were Optimeal^® and Agri-footprint^® (37). Optimeal^® facilitates the design of sustainable diets through optimization, while Agri-footprint^® enables the assessment of multiple environmental impacts, including terrestrial acidification, freshwater eutrophication, land use, and water consumption (38). However, these tools do not integrate nutritional assessment with environmental evaluation, making them less suitable for a comprehensive analysis of diet quality and sustainability. Furthermore, they rely exclusively on the Life Cycle Assessment (LCA) method, which, while robust, does not account for other relevant perspectives, such as the Water Footprint Assessment (WFA) approach—currently the most widely used methodology for evaluating the WF of dietary patterns (5).

Additionally, both softwares were developed for European contexts, limiting their applicability to the Mexican diet due to differences in commonly consumed foods and local dietary patterns. In Mexico, a new methodological framework based on the WFA has been proposed (5). However, to date, no dietary assessment tool has been developed that simultaneously evaluates diet quality and environmental impact, specifically tailored to the Mexican context.

Stage 2: software design and features

The software is structured into two main components: a user section and an administrator section, each designed to support different functionalities, as outlined below.

Users section

The user segment is structured into five sections: (I) Registration and consent, (II) Sociodemographic and nutritional evaluation, (III) 24-hour recall, (IV) FFQ, (V) diet quality, and (VI) environmental impact assessment. Each section is designed to streamline data collection and analysis, ensuring standardized and efficient processing for research purposes. Features like self-reporting options, visual food portions, and automated dietary calculations enhance user interaction and data accuracy. The content of this section is presented in the supplementary table 1 (available at https://cdn.amegroups.cn/static/public/mhealth-25-16-1.pdf).

(I) General data

In Section I, the software provides the option for users to register, log in, and give informed consent. Section II collects general demographic information, including name, surname, phone number, email address, age, date of birth, sex, place of birth, and residence history. The socioeconomic level (39) is classified according to education (40) (Supplementary Material 1.1 in the Appendix: https://cdn.amegroups.cn/static/public/mhealth-25-16-1.pdf) and occupation (41) (Supplementary Material 1.2 in the Appendix: https://cdn.amegroups.cn/static/public/mhealth-25-16-1.pdf). Questions regarding job type, work schedule, number of workdays, and monthly income were also included.

Questions about food purchasing behaviors, monthly food expenditure, primary shopping locations, frequency of dining out, and types of food consumed outside the home were also included. Physical activity is evaluated using the International Physical Activity Questionnaire (IPAQ), which records the type, frequency, duration, and intensity of physical activities (low, moderate, or high) (42).

An additional open-ended question regarding health status was included to capture information on disease presence, type of condition, duration since diagnosis, and current medication use. Furthermore, Nutriecology^® allows for the recording of anthropometric and body composition data. If users have not been measured by a researcher, they can provide self-reported weight and height. When a nutritional evaluation is conducted by the research team, detailed measurements are recorded, including height, weight, body mass index (BMI), body fat percentage, muscle mass, total body water, basal metabolic rate, metabolic age, visceral fat, waist and hip circumferences, and bone mass. The platform accommodates both researcher-collected and self-reported data, enabling flexibility in data entry depending on study requirements.

(II) Dietary assessment

Section III captures dietary intake through a detailed 24-hour recall, covering five common mealtimes in Mexico and recording time, place, ingredients, cooking methods, and water consumption. Visual aids with portion sizes were included to enhance reporting accuracy (43).

Section IV integrates an adapted FFQ based on two validated instruments (16,18), and expanded to 248 food items reflecting both traditional and emerging dietary patterns (5,18,26,44). The detailed structure and food list are provided in Supplementary Materials 2,3 (available at https://cdn.amegroups.cn/static/public/mhealth-25-16-1.pdf). Consumption frequencies and portion sizes were standardized, and daily intake was calculated using a formula specifically developed for this study (Supplementary Material 3). Nutrient composition was derived from national food databases (29,45) and supplemented by food label analysis (Supplementary Material 4). Food classification also incorporated environmental impact criteria (5,34), detailed in Supplementary Material 2. Nutriecology^® automatically estimates 33 nutritional and environmental parameters, as summarized in Supplementary Material 5 (available at https://cdn.amegroups.cn/static/public/mhealth-25-16-1.pdf).

(III) Diet quality

In Section V, diet quality was assessed using the IACDMx, adapted from the ICDMx (28) to incorporate sustainability considerations. The IACDMx includes six components and evaluates both traditional and Westernized food consumption patterns, following national recommendations (8). Detailed scoring methodology and classification thresholds are provided in Supplementary Materials 2 and 6 (available at https://cdn.amegroups.cn/static/public/mhealth-25-16-1.pdf). The FFQ was the sole source for diet quality assessment, allowing a maximum total score of 100 points.

(IV) WF calculation

For the dietary environmental impact assessment, Nutriecology^® incorporates the calculation of the dietary WF, including total, green, blue, and grey WFs (5,46). The method proposed by Lares-Michel et al. (5), based on the WFA approach (46), was implemented in Nutriecology^® to automatically calculate the four WFs of each of the 248 foods included in the FFQ. The first step in the WF calculation involves identifying the daily intake of each food (in grams or milliliters) using data from either the 24-hour recall or the FFQ. The software then determines whether the food requires pre-consumption processing, such as washing, cooking, or peeling. For example, a potato must be washed, peeled, and cooked while milk is consumed as purchased. If a food item is already processed, conversion factors (e.g., 1.45 for meats) provided by Lares-Michel et al. (5) are applied to revert to its raw form. The complete list of foods to which correction factors were applied is detailed in Supplementary Material 2.

After adjusting to raw weight, the software identifies whether additional water is needed for washing or cooking and incorporates the grey WF values reported by Lares-Michel et al. (5): 10 liters/kg for cereals and legumes, 1 liter/kg for meats, and 14.44 liters/kg for other foods. Foods requiring these adjustments are also listed in Supplementary Material 2. The final WF for each food item is calculated using the green, blue, and grey WF values provided by Lares-Michel et al. (5). For composite foods with multiple ingredients, the individual WFs are summed to produce a total WF value, or pre-existing WF data for traditional Mexican dishes is used (5).

Accounting for these detailed aspects in the WF calculation is crucial to accurately assess the environmental impact of diets. Previous studies often calculated WF values based on raw food weights without considering processing losses, leading to discrepancies of up to 135% in WF estimations depending on whether correction factors and processing water usage were considered (5,47). For instance, a 145 g portion of raw beef reduces to 100 g after cooking, which would result in a WF of 2,156.60 liters if uncorrected, compared to an actual WF of 3,127.07 liters when processing water loss is included (5).

Administrator section

The administrator section of Nutriecology^® was designed to manage and customize the software’s database. As outlined in in the supplementary table 7 (Supplementary Material 7, available at https://cdn.amegroups.cn/static/public/mhealth-25-16-1.pdf), administrators can access the system using a pre-established account and password, which provides a menu with options to view, edit, add, or delete foods, food groups, and sub-groups. Additionally, a feature for importing and exporting data via Excel sheets is available, allowing for flexible management and updating of the food database.

Stage 3: development and coding

The coding phase involved continuous feedback from the research team, focusing on interface usability, database integration, and data export capabilities. The development followed a structured workflow, ensuring the software met all predefined specifications (Figure 1). The platform allows automatic calculations for energy, macronutrients, micronutrients, and the three components of WF: green, blue, and grey WF (48).

Figure 1 Process followed for the development of the nutritional ecologic software Nutriecology^®. FFQ, Food Consumption Frequency Questionnaire.

Validation procedures

Two independent studies were performed to validate Nutriecology^®, which are described below.

Study 1

Study design

A cross-sectional study was conducted with 174 adults (18–74 years) from the Metropolitan Zone of Guadalajara, Mexico. Participants were invited to participate in two governmental dependencies and a university. The participants who agreed to be included in Study 1 were gathered in a computer room and were asked to use Nutriecology^® on one occasion to self-report their dietary intake via a 24-hour recall and an FFQ. Validity was assessed by comparing the energy intake, macronutrients, micronutrients, and WFs obtained from the FFQ and the 24-hour recall.

Statistical analysis

Spearman correlations were performed in STATA 12, and Bland-Altman plots were designed in GraphPad Prism 10 (16). It is important to mention that as Nutriecology^® does not classify food obtained from 24-hour recall, nor does it calculate diet quality from this tool, food groups and sub-groups, as well as diet quality, were not included in this study.

Study 2

Study design

A 6-months longitudinal study was conducted with 87 adults (18–35 years) from the South of Jalisco, Mexico. Participants were invited to be included in study 2 at a university, consisting of assisting in nutritional consultation on two occasions, at baseline and after six months. Participants who agreed to participate assisted in a baseline face-to-face nutritional consultation, where they were interviewed by a trained nutritionist who used Nutriecology^® to assess their diet through all the sections of the software (one 24-hour recall and an FFQ). After six months, the procedure was repeated. Validation was also assessed by comparing the energy intake, macronutrients, micronutrients, and WFs obtained from the FFQ and the 24-hour recall.

Statistical analysis

Spearman correlations and Bland-Altman plots were performed (16). Reproducibility was assessed by correlating both 24-hour recalls and both FFQ and its related data (energy intake, macronutrients, micronutrients, and WFs) by Spearman correlations. For this study, food groups, sub-groups, and diet quality were assessed against FFQ 1 and FFQ 2. Analyses were performed in STATA 12 and GraphPad Prism 10.

Ethical considerations

Study 1 was approved by the University of Guadalajara’s Ethics Committee (No. CEICUC-PGE-004) in 2019, and study 2 was approved by the Ethics Committee of the Center for Studies and Research in Behavior of the University Center for Biological and Agricultural Sciences (CUCBA) from the University of Guadalajara (No. CUCBA/CEIC/CE/002/2022) in 2022. Both studies adhered to the Declaration of Helsinki and its subsequent amendments, and complied with the Federal Law on Protection of Personal Data Held by Private Parties. All participants provided written informed consent. In both studies, participants received nutritional advice at the end of the study as a reward for their participation.

Results

Users section

Nutriecology^® was officially launched on the website www.Nutriecology.com. To access the user section, participants must navigate to www.Nutriecology.com/invitados. Figure 2 shows the interface of the software. When accessing, users are required to register using their email address and create a password (Figure 2A), enabling them to save their progress and resume the survey at a later time if needed. Upon registration, users are presented with a general informed consent form outlining the purpose and scope of the questionnaires. If they agree to participate, they can proceed by clicking “accept.”

Figure 2 Nutriecology^® interface. (A) Users’ registration in the users’ section of Nutriecology^®. (B) General data of user’s section in Nutriecology^®. (C) 24-hour recall of user’s section in Nutriecology^®. (D) FFQ of user’s section in Nutriecology^®. (E) Administrator section in Nutriecology^®. (F) Exportable excel sheets in Nutriecology^® administrator section. FFQ, Food Consumption Frequency Questionnaire.

After registration, the platform requests general demographic information from the user (Figure 2B). Next, Nutriecology^® guides users through the 24-hour recall section (Figure 2C), where they input their dietary intake for a single day. Once the 24-hour recall is completed, users are directed to the adapted FFQ (Figure 2D) to record their habitual intake over a longer period. After completing all items in the FFQ, a thank you message and a farewell screen are shown. Calculations for diet composition, quality, and WF are performed internally and are only accessible to the administrator (Figures 2E,2F).

Administrator section

The administrator section can be accessed through www.Nutriecology.com (Figure 2E,2F) using a designated email and password created specifically for administrative purposes. Upon login, a menu is displayed, offering options to view, edit, add, or delete entries for foods, food groups, food sub-groups, and dietary sub-group classifications. Additionally, the administrator can access sociodemographic, nutritional evaluation, 24-hour recall, and FFQ data. There is also a dedicated section for importing and exporting data in Excel format.

When selecting any option (e.g., foods, food groups), the current data entries—along with the complete nutrient and WF information previously uploaded—are displayed. All data are editable and can be updated or deleted as needed, and new entries can be added to the system.

The data export section includes options for downloading up to nine different Excel sheets, corresponding to the following data sets:

• User Information: Registration details, sociodemographic data, and anthropometric and body composition parameters.
• FFQ by Food Item: Nutritional composition and WF breakdown for each food item reported in the FFQ.
• FFQ Summary: Total nutritional composition and WF for the complete FFQ.
• Food Group Classification: Nutritional composition and WF for each food group from FFQ only.
• “Adequate” Food Group Classification: Nutritional composition and WF based on the adequacy criteria from FFQ only.
• Food Sub-group Classification: Nutritional composition and WF for each food sub-group from FFQ only.
• 24-hour Recall by Food Item: Nutritional composition and WF breakdown for each food item recorded.
• 24-hour Recall Summary: Total nutritional composition and WF for the complete 24-hour recall.
• Diet Quality Scores: Total diet quality score and scores for each diet quality component.

Study 1

The mean age of participants was 39.40±15.45 years, and 49% were men. The majority had a high educational level (56%), while occupational status was predominantly categorized as low to medium (83%) (Table 1).

Validation of Nutriecology^® (Study 1)

Energy intake was highly valid, with a correlation coefficient of 0.804 (P<0.0001). Carbohydrates, proteins, and lipids also exhibited strong correlation values, all exceeding 0.64 (P<0.0001), with lipids having the highest correlation among macronutrients (rho =0.71, P<0.0001). For micronutrients, correlation values were lower, reaching a maximum of 0.57 (P<0.0001). Among vitamins, niacin presented the highest correlation (rho =0.43, P<0.0001), while heme iron showed the highest correlation among minerals (rho =0.50, P<0.0001).

High validity was observed for the total WF and the green, blue, and grey components, with significant correlations obtained across all categories. Detailed correlation values and significance levels are presented in Table 2.

The Bland-Altman plots for validity (Figure 3) present data from Study 1 and Study 2. The plots display the average intake values for corresponding items measured by the FFQ and 24-hour recall on the horizontal axis, and the differences between the two methods on the vertical axis. Most data points fall within the limits of agreement (±1.96 SD from the mean difference) for energy, macronutrients, and WF, indicating strong agreement between methods.

Figure 3 Bland-Altman plots for validity showing differences in mean values of energy, macronutrients, and water footprint obtained from FFQs and 24-hour recalls in Studies 1 and 2. FFQ, Food Consumption Frequency Questionnaire.

Study 2

The mean age of participants was 24.58±4.29 years, and 70% were women. The majority had a medium educational level (63%), while occupational status was predominantly categorized as low (68%) (Table 1).

Validation of Nutriecology^® (Study 2)

Regarding validity, energy intake (rho =0.4237; P<0.0001) and macronutrients (rho >0.3; P<0.05) showed an adequate correlation between methods (24-hour recall vs. FFQ) in both evaluations (baseline and month 6). Regarding WF, blue WF correlated the highest between methods with a rho =0.3480 (P<0.001) (Table 3).

With respect to reproducibility, both tools (24-hour recall and FFQ) showed a high level, but FFQ highlighted with values rho >0.6 (P<0.0001), including energy, macronutrients, micronutrients, and WF (Table 3). Most food groups (Table 4) and sub-groups (Table 5), both regarding intake and WF, showed high significant correlations between FFQs, especially vegetables (rho >0.7; P<0.0001), milk (rho >0.68; P<0.0001), and fast food (rho >0.7; P<0.0001). However, only a few sub-groups, like oatmeal, obtained low or no statistical significance correlations. The Bland-Altman plots for reproducibility (Figure 4) display the average intake values for corresponding items measured by the two FFQ and two 24-hour recall on the horizontal axis and the differences between the same method after six months on the vertical axis. Most data points fall within the limits of agreement (±1.96 SD from the mean difference) for energy, macronutrients, and WF, indicating strong agreement between the instruments through time.

Figure 4 Bland-Altman plots for reproducibility showing differences in mean values of energy, macronutrients, and water footprint obtained from FFQ 1 vs. FFQ 2 and 24-hour recall 1 vs. 24-hour recall 2 in Study 2. FFQ, Food Consumption Frequency Questionnaire.

Regarding diet quality, Table 6 presents the mean and standard deviations of the scores obtained from the IACDMx. Each component has a maximum possible score of 20 points, with a total maximum score of 100 points. The lowest score was observed in the innocuous component (6.99±2.92 points), while the highest score was recorded in the Varied component (19.70±0.89 points). Most components obtained moderate to high correlations (rho >0.5; P<0.0001). Notably, the sufficient component (rho >0.5; P<0.0001), particularly calcium and water (rho >0.6; P<0.0001) showed high reproducibility. The balance component, specifically protein (rho >0.5; P<0.0001), as well as the complete component regarding fruits and vegetables and fish and seafood (rho >0.6; P<0.0001) showed high correlations. The innocuous component also showed high correlations, particularly regarding sodium, alcohol, and sugar (rho >0.5; P<0.0001).

Table 7 presents the adequate classification of the IACDMx both from Study 1 and Study 2. As can be seen, most of the population regularly consumes at least 30 g of traditional Mexican foods, being classified as ‘Adequate A’. This was also observed with the Mexican traditional dishes, as most of the population (>80%) consumed at least 180 g of these dishes a week (‘Adequate B’). Regarding Western foods and dishes, most people regularly consume at least one ultra-processed food (‘Inadequate C’) or dish food (‘Inadequate D’).

Discussion

This study aimed to develop and validate Nutriecology^®, a novel Mexican online nutritional and ecological software designed to simultaneously assess dietary intake, diet quality, and environmental impact using a 24-hour recall and an FFQ. This tool was created to facilitate the efficient collection and analysis of nutritional and environmental data at individual and population levels, thereby improving research standardization and usability. At the time of its development, no existing tool—nationally or internationally—combined automated dietary, environmental, and diet quality assessments within a single interface.

In recent years, international tools have begun to incorporate sustainability metrics into dietary platforms. For example, myfood24^® has integrated environmental indicators such as greenhouse gas emissions (GHGE), land use, and water use into its food composition databases (49,50). However, its core functionality remains centered on nutrient intake estimation, without automated evaluation of diet quality or food group classification. In contrast, Nutriecology^® not only includes environmental impact through WF estimation, but also integrates a culturally adapted diet quality index and systematic food classification, allowing for a more comprehensive analysis of sustainable dietary patterns in the Mexican context.

Other approaches, such as the mobile food record (mFR) developed by Harray et al. (51), have explored sustainability-oriented dietary assessment through image-based records, focusing on indicators like ultra-processed food consumption, packaging, and plate waste. Although conceptually valuable, this method was not developed into a functional, validated software for real-time dietary analysis. By contrast, Nutriecology^® enables immediate data processing, incorporates national dietary guidelines, and offers automated conversion of food intake into nutritional and environmental indicators.

The Food4Me project represents another significant effort in online dietary assessment, with validated FFQs demonstrating high correlation with the EPIC-Norfolk FFQ and moderate agreement with weighed food records. However, similar to other tools, it lacks integration of environmental metrics (52,53). While recent versions of Food4Me focus on personalized nutrition, Nutriecology^® remains distinctive in its dual-purpose design, combining automated assessment of diet quality and environmental impact. This integrated approach addresses a critical gap in digital tools for nutrition research and public health, particularly in the context of Mexico.

Strengths of Nutriecology^®

Each feature of Nutriecology^® was designed to enhance the capabilities of existing tools. The software not only collects detailed sociodemographic, physical activity, and body composition data but also integrates these aspects for a more comprehensive analysis (10,14). This allows for a holistic view of nutritional and environmental interactions, which is often lacking in similar tools. Furthermore, the 24-hour recall module includes flexible options for preparation methods, mealtimes, and consumption settings, enabling both retrospective and prospective dietary analyses (8).

For the FFQ, Nutriecology^® provides a more detailed assessment by expanding food item options and refining the frequency categories. This feature addresses the limitations of conventional FFQs, which often overlook the environmental impact of infrequently consumed items, such as holiday foods (e.g., Christmas turkey), that contribute significantly to dietary WF (54). The combination of daily, weekly, monthly, and annual frequency options, along with portion size specifications, allows for more accurate estimations of both dietary intake and environmental impact.

The incorporation of the IACDMx enhances traditional diet quality assessments by incorporating sustainability principles. The “complete” component distinguishes between plant-based and animal-based foods, prioritizing the intake of legumes, fruits, vegetables, and whole grains, while limiting red meat and dairy. This modification aligns with public health guidelines and sustainability targets, making it particularly relevant for research on diet-related disease risks and environmental sustainability (31,34,55,56). By distinguishing between different animal protein sources (e.g., beef vs. chicken), Nutriecology^® captures the substantial variations in WF, with beef exceeding 20,000 liters per kilogram compared to 4,000 liters for chicken and legumes (5,33,57).

Additionally, the tool improves on the “innocuous” component by including an evaluation of sugar intake, addressing a critical gap in previous indices. High sugar consumption has been linked to multiple non-communicable diseases, and this inclusion allows for a more comprehensive assessment of dietary risks (58). The new “adequate” component evaluates diet adherence to cultural and economic norms, making Nutriecology^® a versatile tool for studying nutrition transitions and sustainable diet patterns in Mexico (59).

A unique strength of Nutriecology^® is its capacity to automatically calculate WF, including adjustments for cooking and food preparation losses. Previous studies have shown that failure to account for these factors can result in WF variations of up to 135% (5,47). The built-in WF database also accommodates complex Mexican dishes with multiple ingredients, which is essential for accurately assessing traditional diets.

Another key advantage is the flexibility of the software’s database. Nutriecology^® allows administrators to add new foods, update nutritional data (e.g., vitamins, phytochemicals), and incorporate additional environmental indicators (e.g., greenhouse gas emissions, land use, biodiversity impact). This adaptability extends its usability beyond Mexico, enabling it to support dietary assessments in other cultural contexts, such as Mediterranean diets, by simply updating the database with region-specific foods and metrics.

Limitations of Nutriecology^®

Despite its strengths, Nutriecology^® has some limitations. Currently, the software only supports WF as the environmental impact index, limiting its ability to provide a broader environmental profile. Moreover, the extensive number of items in the FFQ and the detailed nature of the 24-hour recall may increase respondent burden, potentially affecting data accuracy if participants do not complete the surveys thoroughly. This limitation suggests that trained personnel should administer the tool whenever possible to minimize self-reporting errors.

Another limitation is that the IACDMx score is currently only calculated using FFQ data, with no diet quality assessment available for 24-hour recall entries. This restricts the tool’s flexibility in studies that rely on multiple dietary assessment methods. Additionally, new indices such as the Global Diet Quality Score (60) and the Traditional Mexican Diet Index (61) have been developed since the creation of Nutriecology^®, which could offer alternative approaches for evaluating diet quality and adherence to traditional dietary patterns in Mexico.

Finally, the present study did not assess user experience or perception regarding the software’s usability. Understanding user satisfaction and ease of use is crucial for ensuring broader adoption and effective application of digital dietary assessment tools. Future research should address this aspect to identify potential improvements and optimize Nutriecology^®’s functionality and accessibility.

Validation of Nutriecology^®

The validation results support that Nutriecology^® is a robust and reliable tool for assessing dietary intake and environmental impact. Beyond demonstrating significant correlations across energy, macronutrients, micronutrients, and WF components, its high reproducibility highlights its potential for longitudinal research applications. Compared to other studies that have validated dietary assessment tools (16,26), the correlation values obtained indicate that Nutriecology^® meets or exceeds the standards typically reported, supporting its reliability and practical applicability.

Moreover, Nutriecology^® addresses a critical gap by enabling the simultaneous evaluation of nutritional quality and environmental sustainability, aligning with the global need for integrated approaches in dietary assessment. This dual functionality strengthens its utility in studies aiming to promote sustainable diets and to evaluate both health and environmental impacts comprehensively.

While other validation studies have primarily compared FFQs against dietary records or weighed food records (62), future research could incorporate biomarkers and additional advanced validation methods to further enhance the robustness of Nutriecology^® (63).

Future directions and software maintenance

Given the waterfall life cycle methodology used, continuous maintenance and updates are essential (48). Planned updates include separating the FFQ into smaller sections to reduce respondent burden and introducing additional environmental impact metrics, such as greenhouse gas emissions (kg CO₂/eq), land use (m2*y), and fossil energy use (megajoules). A new weighted score, the pReCiPe score, is also under consideration for comprehensive environmental impact assessment (64).

Other planned improvements include adapting the IACDMx to align with the Global Diet Quality Score (60) and the Traditional Mexican Diet Index (61), expanding the database to include food additives, bioactive compounds, and economic aspects, and enhancing the evaluation of the “adequate” component. Moreover, since by the time of the design of Nutriecology^®, the planetary diet was not launched, future updates of the software will consider the recommended amounts of consumption to better align with sustainability goals (2).

Based on the above, the development of a new dietary sustainability index could be added to Nutriecology^®, enhancing the capabilities of the software to assess diet quality and environmental sustainability simultaneously. Expanding Nutriecology^® to other countries and cultural contexts will be a priority for future collaborations, enabling it to become a global tool for sustainable diet assessment.

Conclusions

We developed Nutriecology^®, the first nutritional-ecological software specifically designed for the Mexican context that enables simultaneous assessment of dietary intake, diet quality, and environmental impact through WF. The adapted diet quality index, IACDMx, refines the original ICDMx components to better align with current dietary recommendations and promote sustainable food choices.

Nutriecology^® is the first tool to offer automated dietary WF assessment that incorporates correction factors for food preparation (e.g., cooked vs. uncooked, peeled vs. unpeeled) and includes water used during food washing and cooking. By accounting for the cumulative WF of ingredients in composite dishes, it provides a comprehensive environmental evaluation rarely addressed in existing tools.

Validation results confirmed the tool’s reliability, accuracy, and reproducibility for assessing both nutritional and environmental dimensions. Its implementation in research, clinical, and surveillance settings could support Mexico’s public health system by enabling standardized monitoring of dietary quality and sustainability. Moreover, its integration into nutrition programs and policies could guide evidence-based strategies to encourage healthier and more environmentally sustainable eating patterns at the population level.

Ongoing updates and future developments will enhance its functionality by incorporating additional environmental indicators—such as greenhouse gas emissions and land use—further expanding its applicability in national and international sustainability research.

Acknowledgments

We thank the National Council for Science and Technology (CONACYT) for scholarship [number 717186 (CVU 934420)].

Footnote

Data Sharing Statement: Available at https://mhealth.amegroups.com/article/view/10.21037/mhealth-25-16/dss

Peer Review File: Available at https://mhealth.amegroups.com/article/view/10.21037/mhealth-25-16/prf

Funding: This research was partially funded by the SNI-SNCA Permanence Scholarship Program (PROSNI) by the University of Guadalajara. However, the financial support did not influence the design and results of the study.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://mhealth.amegroups.com/article/view/10.21037/mhealth-25-16/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Study 1 was approved by the University of Guadalajara’s Ethics Committee (No. CEICUC-PGE-004) in 2019, and study 2 was approved by the Ethics Committee of the Center for Studies and Research in Behavior of the University Center for Biological and Agricultural Sciences (CUCBA) from the University of Guadalajara (No. CUCBA/CEIC/CE/002/2022) in 2022. Both studies adhered to the Declaration of Helsinki and its subsequent amendments, and complied with the Federal Law on Protection of Personal Data Held by Private Parties. All participants provided written informed consent.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Hasenböhler A, Denes L, Blanstier N, et al. Development of an Innovative Online Dietary Assessment Tool for France: Adaptation of myfood24. Nutrients 2022;14:2681. [Crossref] [PubMed]
2. Willett W, Rockström J, Loken B, et al. Food in the Anthropocene: the EAT-Lancet Commission on healthy diets from sustainable food systems. Lancet 2019;393:447-92. [Crossref] [PubMed]
3. Deng Z, Hu Y, Wang X, et al. Transitioning to healthy and sustainable diets has higher environmental and affordability trade-offs for emerging and developing economies. Nat Commun 2025;16:3948. [Crossref] [PubMed]
4. National Aeronautics and Space Administration. Sequía generalizada en México. 2021 [cited 2022 Jan 21]. Available online: https://ciencia.nasa.gov/ciencias-terrestres/sequia-generalizada-en-mexico/#:~:text=M%C3%A9xico%20est%C3%A1%20viviendo%20una%20de,para%20beber%2C%20cultivar%20y%20regar
5. Lares-Michel M, Housni FE, Aguilera Cervantes VG. A quantitative estimation of the water footprint of the Mexican diet, corrected for washing and cooking water. Food Sec 2021;13:849-74.
6. Ambrosini GL, Hurworth M, Giglia R, et al. Feasibility of a commercial smartphone application for dietary assessment in epidemiological research and comparison with 24-h dietary recalls. Nutr J 2018;17:5. [Crossref] [PubMed]
7. Ghammachi N, Dharmayani PNA, Mihrshahi S, et al. Investigating Web-Based Nutrition Education Interventions for Promoting Sustainable and Healthy Diets in Young Adults: A Systematic Literature Review. Int J Environ Res Public Health 2022;19:1691. [Crossref] [PubMed]
8. Food and Agriculture Organization of the United Nations. Dietary assessment: a resource guide to method selection and application in low resource settings. 2018. Available online: https://openknowledge.fao.org/server/api/core/bitstreams/3dc75cfc-9128-4f29-9d76-8d1f792078f0/content
9. Eldridge AL, Piernas C, Illner AK, et al. Evaluation of New Technology-Based Tools for Dietary Intake Assessment-An ILSI Europe Dietary Intake and Exposure Task Force Evaluation. Nutrients 2018;11:55. [Crossref] [PubMed]
10. Carter MC, Albar SA, Morris MA, et al. Development of a UK Online 24-h Dietary Assessment Tool: myfood24. Nutrients 2015;7:4016-32. [Crossref] [PubMed]
11. Dao MC, Subar AF, Warthon-Medina M, et al. Dietary assessment toolkits: an overview. Public Health Nutr 2019;22:404-18. [Crossref] [PubMed]
12. myfood24®. Quick and easy to use nutritional analysis software 2022 [cited 2022 Feb 28]. Available online: https://www.myfood24.org/
13. Aznutrition. AZNUTRITION 2022 [cited 2022 Jul 12]. Available online: https://www.aznutrition-oficial.com
14. Nutricloud 2022 [cited 2022 Feb 28]. Available online: https://www.nutricloud.mx/
15. Nutrimind. Software de Nutrición 2022 [cited 2022 Mar 14]. Available online: https://www.nutrimind.net/
16. Macedo-Ojeda G, Vizmanos-Lamotte B, Márquez-Sandoval YF, et al. Validation of a semi-quantitative food frequency questionnaire to assess food groups and nutrient intake. Nutr Hosp 2013;28:2212-20.
17. Nutritools. Dietary Assessment Tools 2022 [cited 2022 Jul 11]. Available online: https://www.nutritools.org/tools
18. Denova-Gutiérrez E, Tucker KL, Salmerón J, et al. Relative validity of a food frequency questionnaire to identify dietary patterns in an adult Mexican population. Salud Publica Mex 2016;58:608-16. [Crossref] [PubMed]
19. Lares-Michel M, Housni FE, Aguilera Cervantes VG, et al. Eat Well to Fight Obesity… and Save Water: The Water Footprint of Different Diets and Caloric Intake and Its Relationship With Adiposity. Front Nutr 2021;8:694775. [Crossref] [PubMed]
20. Gao SK, Beresford SA, Frank LL, et al. Modifications to the Healthy Eating Index and its ability to predict obesity: the Multi-Ethnic Study of Atherosclerosis. Am J Clin Nutr 2008;88:64-9. [Crossref] [PubMed]
21. Kennedy ET, Ohls J, Carlson S, et al. The Healthy Eating Index: design and applications. J Am Diet Assoc 1995;95:1103-8. [Crossref] [PubMed]
22. McCullough ML, Feskanich D, Stampfer MJ, et al. Diet quality and major chronic disease risk in men and women: moving toward improved dietary guidance. Am J Clin Nutr 2002;76:1261-71. [Crossref] [PubMed]
23. Sanjeevi N, Freeland-Graves J, Hersh M. Food insecurity, diet quality and body mass index of women participating in the Supplemental Nutrition Assistance Program: The role of intrapersonal, home environment, community and social factors. Appetite 2018;125:109-17. [Crossref] [PubMed]
24. López-Olmedo N, Popkin BM, Gordon-Larsen P, et al. Cross-sectional association between diet quality and cardiometabolic risk by education level in Mexican adults. Public Health Nutr 2020;23:264-74. [Crossref] [PubMed]
25. López-Olmedo N, Popkin BM, Mendez MA, et al. The association of overall diet quality with BMI and waist circumference by education level in Mexican men and women. Public Health Nutr 2019;22:2777-92. [Crossref] [PubMed]
26. Macedo-Ojeda G, Márquez-Sandoval F, Fernández-Ballart J, et al. The Reproducibility and Relative Validity of a Mexican Diet Quality Index (ICDMx) for the Assessment of the Habitual Diet of Adults. Nutrients 2016;8:516. [Crossref] [PubMed]
27. Diario Oficial de la Federación. DOF - Diario Oficial de la Federación. 2012 [cited 2022 Jun 8]. Available online: https://www.dof.gob.mx/nota_detalle.php?codigo=5259329&fecha=13/07/2012#gsc.tab=0
28. NOM-043-2012.pdf. 2012 [cited 2021 Jan 27]. Available online: https://www.gob.mx/cms/uploads/attachment/file/138258/NOM-043-servicios-basicos-salud-educacion-alimentaria.pdf
29. Pérez-Lizaur AB, Berenice Palacios González, Ana Laura Castro Becerra, Isabel Flores Galicia. Sistema Mexicano de Alimentos Equivalentes. Ogali; 2014.
30. Dinter J, Bechthold A, Boeing H. Fish intake and prevention of selected nutrition-related diseases. Ernahrungs Umschau 2016;63:148-54.
31. Jenkins DJ, Kendall CW, Augustin LS, et al. Effect of legumes as part of a low glycemic index diet on glycemic control and cardiovascular risk factors in type 2 diabetes mellitus: a randomized controlled trial. Arch Intern Med 2012;172:1653-60. [Crossref] [PubMed]
32. Genkinger JM, Koushik A. Meat consumption and cancer risk. PLoS Med 2007;4:e345. [Crossref] [PubMed]
33. Mekonnen MM, Hoekstra AY. The green, blue and grey water footprint of crops and derived crop products. Hydrol Earth Syst Sci 2011;15:1577-600.
34. Clark MA, Springmann M, Hill J, et al. Multiple health and environmental impacts of foods. Proc Natl Acad Sci U S A 2019;116:23357-62. [Crossref] [PubMed]
35. World Health Organization (WHO). Guideline: sugars intake for adults and children 2015 [cited 2022 Jun 21]. Available online: https://www.who.int/publications-detail-redirect/9789241549028
36. Food and Agriculture Organization of the United Nations, World Health Organization. Sustainable healthy diets: guiding principles. 2019 [cited 2021 May 14]. Available online: http://www.fao.org/3/ca6640en/ca6640en.pdf
37. Agri-footprint. Blonk Sustainability. 2022 [cited 2022 Mar 14]. Available online: https://blonksustainability.nl/tools/agri-footprint
38. Optimeal. Blonk Sustainability. 2022 [cited 2022 Mar 14].Available online: https://blonksustainability.nl/tools/optimeal
39. Lares-Michel M, Housni FE, Cervantes VGA, et al. The Relationship between Consumption, Socioeconomic Level and Reasons of Tomato Intake in Mexico. Agricultural Sciences 2018;9:771-91.
40. Autonomous University of Mexico (UNAM). Plan Educativo Nacional 2015 [cited 2021 Jan 27]. Available online: https://www.gob.mx/cms/uploads/attachment/file/207252/Modelo_Educativo_OK.pdf
41. Instituto Nacional de Estadística y Geografía [INEGI]. 2015 [cited 2021 Jan 27]. Available online: https://www.inegi.org.mx/default.html
42. International Physical Activity Questionnaire [IPAQ]. IPAQ scoring protocol2002 [cited 2021 Jan 27]. Available online: https://sites.google.com/view/ipaq/score?authuser=0
43. Romero-Gwynn E, Gwynn D. Dietary patterns and acculturation among latinos of mexican descent. The Julian Samora Research Institute. Universidad del Estado de Michigan; 1997 p. 12. Report No.: 23.
44. Denova-Gutiérrez E, Ramírez-Silva I, Rodríguez-Ramírez S, et al. Validity of a food frequency questionnaire to assess food intake in Mexican adolescent and adult population. Salud Publica Mex 2016;58:617-28. [Crossref] [PubMed]
45. de Chávez M, Solano JÁL, Villasana AC, et al. Composición de alimentos Miriam Muñoz de Chávez Valor nutritivo de los alimentos de mayor consumo. Second edition. McGraw-Hill; 2021.
46. Aldaya MM, Chapagain AK, Hoekstra AY, et al. The Water Footprint Assessment Manual: Setting the Global Standard. 1st ed. Routledge; 2012.
47. Blas A, Garrido A, Willaarts BA. Evaluating the Water Footprint of the Mediterranean and American Diets. Water 2016;8:448.
48. Widiaty I, Riza L, Gafar Abdullah A, et al. Web-Based Digital Learning Application of Iconic Batik in Batik Learning at Vocational High School. Journal of Engineering Science and Technology 2019;14:2475-84.
49. Williams G, Hasenböhler A, Hamilton A, et al. Sustainability metrics of the UK diet using myfood24. Proceedings of the Nutrition Society 2024;83:E484.
50. Adam. Measuring Sustainability with myfood24. myfood24. 2022 [cited 2025 May 10]. Available online: https://www.myfood24.org/blog/you-can-now-measure-sustainability-with-myfood24/
51. Harray AJ, Boushey CJ, Pollard CM, et al. A Novel Dietary Assessment Method to Measure a Healthy and Sustainable Diet Using the Mobile Food Record: Protocol and Methodology. Nutrients 2015;7:5375-95. [Crossref] [PubMed]
52. Forster H, Fallaize R, Gallagher C, et al. Online dietary intake estimation: the Food4Me food frequency questionnaire. J Med Internet Res 2014;16:e150. [Crossref] [PubMed]
53. Fallaize R, Forster H, Macready AL, et al. Online dietary intake estimation: reproducibility and validity of the Food4Me food frequency questionnaire against a 4-day weighed food record. J Med Internet Res 2014;16:e190. [Crossref] [PubMed]
54. thesra.org. The sustainable Christmas dinner. 2014 [cited 2022 Jul 14]. Available online: https://www.sustainabilityexchange.ac.uk/files/christmasfactsheet2014.pdf
55. Barnard N, Levin S, Trapp C. Meat consumption as a risk factor for type 2 diabetes. Nutrients 2014;6:897-910. [Crossref] [PubMed]
56. Briggs MA, Petersen KS, Kris-Etherton PM. Saturated Fatty Acids and Cardiovascular Disease: Replacements for Saturated Fat to Reduce Cardiovascular Risk. Healthcare (Basel) 2017;5:29. [Crossref] [PubMed]
57. Mekonnen MM, Hoekstra AY. A Global Assessment of the Water Footprint of Farm Animal Products. Ecosystems 2012;15:401-15.
58. Prada M, Saraiva M, Garrido MV, et al. Perceived Associations between Excessive Sugar Intake and Health Conditions. Nutrients 2022;14:640. [Crossref] [PubMed]
59. Lares-Michel M, Housni FE, Aguilera Cervantes VG, et al. The water footprint and nutritional implications of diet change in Mexico: a principal component analysis. Eur J Nutr 2022;61:3201-26. [Crossref] [PubMed]
60. Castellanos-Gutiérrez A, Rodríguez-Ramírez S, Bromage S, et al. Performance of the Global Diet Quality Score with Nutrition and Health Outcomes in Mexico with 24-h Recall and FFQ Data. J Nutr 2021;151:143S-51S. [Crossref] [PubMed]
61. Valerino-Perea S, Armstrong MEG, Papadaki A. Development of an index to assess adherence to the traditional Mexican diet using a modified Delphi method. Public Health Nutr 2021;24:4387-96. [Crossref] [PubMed]
62. Steinemann N, Grize L, Ziesemer K, et al. Relative validation of a food frequency questionnaire to estimate food intake in an adult population. Food Nutr Res 2017;61:1305193. [Crossref] [PubMed]
63. Lucassen DA, Brouwer-Brolsma EM, Boshuizen HC, et al. Validation of the smartphone-based dietary assessment tool “Traqq” for assessing actual dietary intake by repeated 2-h recalls in adults: comparison with 24-h recalls and urinary biomarkers. The American Journal of Clinical Nutrition 2023;117:1278-87. [Crossref] [PubMed]
64. Grasso AC, Olthof MR, van Dooren C, et al. Effect of food-related behavioral activation therapy on food intake and the environmental impact of the diet: results from the MooDFOOD prevention trial. Eur J Nutr 2020;59:2579-91. [Crossref] [PubMed]