Exploring the use of Moringa oleifera as a vegetable in Agua Caliente Nueva, Jalisco, Mexico: A qualitative study

People in Mexico speak of the medicinal properties of Moringa oleifera (“moringa”), but they know less about its nutritive properties. A qualitative, participatory approach was utilized to explore the use of moringa as a vegetable in Agua Caliente Nueva, Jalisco, by enrolling 14 female participants (mothers over 18) who were nutritional gatekeepers of the family and were responsible for preparing family meals using moringa leaves. Participants prepared meals with moringa and participated in semi-structured interviews to record their perceptions of foods, shopping habits, and knowledge and experience using moringa. The study objectives were to assess the regional acceptability and feasibility of introducing a readily available and nutritious plant into diets, and to examine whether moringa could potentially serve as part of a solution to the double burden of malnutrition in Mexico. Findings suggest that given its ability to thrive in this dry tropical region, moringa has the potential to be recognized as a viable component in the traditional diet in this small “ejido” communal farming community.

TA B L E 1 Nutritional composition of Moringa oleifera (dried) leaves. The macronutrients, vitamins, and minerals have been frequently reported and widely reviewed. This table presents consensus values along with the references from which those values were chosen. Only the predominant glucosinolate, glucomoringin, has been reliably measured across a wide number of cultivars, and we report that value herein. A variety of other bioactive phytochemicals, predominantly flavonoids, have also been reported, but not across a wide selection of cultivars or wild accessions Nutrient Amount in one serving (10 g) dried moringa powder  Gopalakrishnan et al. (2016). § Olson et al. (2016). ¶ Chodur et al. (2018).
Although there is existing literature that supports the use of moringa as a vegetable and potential weapon against malnutrition in south Asia and parts of Africa, less is known about its usage in Latin America, specifically in Mexico. Mexico is of interest because moringa likely first reached the Americas via the Pacific coast of Mexico as part of the galleon trade between Manila and Acapulco, with the earliest known records of its arrival in the Americas being from Mexico (Olson & Fahey, 2011). Mexico is also of interest because the double burden of malnutrition is prevalent in Mexico, where nearly three quarters of the population aged 15 years or older are identified as overweight or obese (OECD, 2019;Popkin et al., 2019). Food environments, specifically food deserts and food swamps, serve as optimal conditions for obesity to manifest due to the convenience of empty calorie, shelf-stable packaged and highly processed foods (Chen et al., 2016;Cooksey-Stowers et al., 2017). A nutritious plant such as moringa, with low maintenance needed for its rapid growth and drought resistance, has the potential to make a significant impact in Mexico today. However, more research needs to be done to identify areas of opportunity for introducing it to help combat the complicated manifestations of malnutrition and obesity among the country's population.
The ideal investigation would aim to fill in the gaps about moringa usage in Mexico and evaluate whether residents would be willing to incorporate the plant into their diets for nutritional benefits.
Researchers would visit a community in Mexico where moringa was prevalent and engage with residents about their exposure or knowledge about the plant. This would allow researchers to assess existing knowledge about moringa, its use among a small subpopulation in Mexico, and whether it would be possible for future interventions to introduce moringa as a vegetable in the diet. This investigation would target women, ideally those with children, because mothers tend to be the nutritional gatekeepers of the household (a nutritional gatekeeper is the main individual in the household who makes decisions related to food purchasing and preparation) (Allen & Sachs, 2007;Wansink, 2003). By interviewing this group, researchers would be able to gather insights regarding perceptions of "good" versus "bad" foods and whether these perceptions would impact food shopping habits.
The ideal investigation would include a participatory component where these women would be given moringa leaves and be encouraged to prepare meals with them. This would allow researchers to assess the acceptability and feasibility of introducing moringa as a vegetable to households throughout Mexico as a potential method to tackle the double burden of malnutrition in the country.
To provide such data, the objectives of this study were to assess how nutritional gatekeepers in a small town within an ejido (a collectively owned agricultural community, recognized by the Mexican government) made food decisions based on perceptions of nutrition, availability of food, and shopping habits, and how Moringa oleifera was used in Mexico, specifically among populations in small towns or villages. Additionally, this study aimed to ascertain whether this group possessed knowledge of moringa's health-promoting or medicinal properties, and whether the participants were amenable to incorporating a new element to their meals in efforts to improve the nutritional values of their family diets. This study may function as a gateway example for future, larger-scale evaluations of the potential to deliberately use and introduce Moringa oleifera to as a solution for malnutrition in Mexico.

METHODS
This was a study to explore the perceptions of foods and diets in the household, and the feasibility and acceptability of Moringa oleifera as a vegetable in Agua Caliente Nueva, an ejido in Jalisco, Mexico, with a total population size of approximately 750 residents, based on satellite canvassing using Google Earth and door-to-door canvassing conducted by the researchers in 2015 (INEGI, n.d.). A qualitative, mixed-methods approach was applied, using semi-structured interviews conducted in Spanish, and a community participatory function similar to the Trials of Improved Practice (TIPs) approach, developed by the Manoff Group. TIPs is a formative research method that invites a limited number of participants to test a new practice and share their feedback with researchers before an intervention is implemented (Dickin et al., 1997;Gittelsohn et al., 2006;The Manoff Group, 2005;The Manoff Group, n.d.). These participants, also known as TIPs consultants, provide valuable insight, which helps researchers adjust and refine an intervention before it is launched on a wider scale. Observations were held to witness the process of preparing meals with moringa. The observations took place after a cooking demonstration was conducted to depict how to use moringa in traditional Mexican meals. The cooking demonstration and observations were modeled on the TIPs approach (Dickin et al., 1997;The Manoff Group, 2005; The Manoff Group, n.d.).
Purposive sampling was used for this study to intentionally target a subgroup in this community. Inclusion criteria included women, specifically mothers over the age of 18 who resided in Agua Caliente Nueva, who: (a) were the main food preparers of their household, (b) agreed to attempt incorporating moringa leaflets into their meals, and (c) were available to participate in an interview after using moringa.
This method of sampling was chosen because in this community and in Mexico women traditionally make food purchasing and preparation decisions in their household (Allen & Sachs, 2007;Bee, 2014;Kimoto et al., 2014;Quisumbing et al., 2014). Sixteen female participants were recruited to take part in the study, with ages ranging between 24 and 72 years. Two participants dropped out when one moved to a different city and one had a family emergency. Two participants were interviewed together (mother and daughter), for a total of 13 interviews. All participants elected to participate in the study without compensation or incentives.
One participant who had prior experience cooking with moringa leaflets was identified and recruited to conduct the cooking demonstration. The demonstration was scheduled to be held in mid-July 2017, and two in-person invitations and reminders were provided one week and three days prior to the event. Four participants attended the cooking demonstration, one of whom was the demonstrator herself, who prepared quesadillas using moringa as portrayed in Figure 1. All the participants in the study were provided two loosely packed quartsized bags of moringa leaves, one after the cooking demonstration F I G U R E 1 Tortillas as prepared during the cooking demonstration. Ingredients used in this demonstration included flour tortillas, cheese, and moringa leaflets stripped from complete leaves and another a few days prior to each scheduled interview. They were instructed to attempt preparing a palatable dish or "culinary creation" using the leaflets, based either upon the demonstration or through their own improvisational skill as food preparers/cooks. Fresh moringa leaves were harvested by the study team from a mature local moringa tree previously identified (Figure 2a), one compound leaf containing many leaflets, per participant, and they were placed in quart-sized cloth canvas bags. The leaves were distributed to each participant immediately after harvesting to ensure that they were freshly picked off the tree. Upon distribution of the moringa leaves, interview and observation dates were scheduled with the 14 participants. A few days prior to each scheduled interview, each participant was provided a second leaf of moringa to prepare something different or to use in case they did not prepare the first bunch they received while they were still fresh. The researchers were able to observe six participants as they picked the leaflets off the leaves and used them to prepare meals for their families ( Figure 2b). Four participants did not end up using the moringa leaves that were offered to them but agreed to participate in their interviews.
The interviews and observations took place between July and August 2017 in Agua Caliente Nueva.
The interviews were conducted in Spanish and consisted of a total of 53 questions (Supporting Information Material S1), with the interviews ranging from 25 minutes to one hour long. Some questions were skipped based upon answers to the previous question, and whether the interviewer judged them relevant to a particular household situation. Two main topics were discussed during the interview: (a) food and diets within the home and (b) moringa in Agua Caliente Nueva and in F I G U R E 2 (a) Moringa oleifera leaves on a twig. (b) Moringa leaves are compound (they have many small leaflets borne on a branching frame). Both the petiole (the "stalk" that holds the whole leaf) and the main rachises (the main "branches" of the leaf) are tough and generally not eaten. Thus, when preparing fresh moringa leaves in a food product, one separates the leaflets from the rachises and uses the leaflets in cooking. Variation in how much rachis is included in a preparation is probably a major contributor to variation in nutritional content across moringa samples. Scale (spaces between black lines on background) in b = 10 cm the home. Data collection spanned four weeks after the initial rapportbuilding, participant selection, and cooking demonstration were performed.
Interviews were conducted in Spanish and all by the same interviewer. As outsiders to the community, the research team deemed it appropriate not to require written consent to participate in the study.
Thus, in accordance with IRB instructions, consent forms were verbally explained and participants provided their oral consent prior to the interviews and observations. Participants were made aware that at any point they could skip any question or choose to withdraw their responses and terminate their participation in the study.

Analysis
Data for analysis included English-translated transcripts of audio recordings from the semi-structured interviews as well as written field notes from observation activities. Of the 13 interviews conducted, 12 were audio-recorded using the Voice Memos mobile application on a password-protected iPhone 6 mobile device (2007). After each interview was conducted, the audio recording was immediately relabeled and saved with the participant's identification number and uploaded onto a password-protected computer. Once each file transfer was complete, the voice memo recording was deleted. The audio recordings were then used to perform simultaneous translation to create Englishtranslated transcripts.
Transcripts were initially coded using the qualitative software program Atlas.ti Version 1.6.0 (Atlas.ti, n.d.) and a codebook was developed. Line-by-line coding of these transcripts produced over 100 initial codes. Coded text was organized by category and read multiple times using constant comparisons between participants to identify themes across the interviews on the subjects of both their perceptions of nutrition and moringa (Charmaz, 2006). After reviewing the transcripts, codes, and written field notes, analytic memos were prepared to facilitate reflecting and identifying emergent patterns, categories, and themes (Miles et al., 2015;Saldaña, 2016).

Food and diets within the household
During discussions regarding food and diets within the household, participants were asked to think of "good" versus "bad" foods, referenced in Table 2. As examples of "good" foods, 85% and 77% of participants named vegetables and fruit, respectively. Following this, 38% of participants (5 out of 13 interviews) thought fish and dairy were considered good foods as well. In contrast, when probed on examples of "bad" foods, 88% of participants thought of red meats such as pork and/or beef and 77% of participants thought of instant, processed, or canned foods. Direct quotations from participants (below) are italicized and are followed by the participant's number in parentheses (see Table 3).
Approximately one-third of participants believed starches were bad for one's health but continued to incorporate tortillas into meals. In one interview, a participant stated that she "knew tortillas were bad," but continued to prepare meals that incorporated tortillas anyway.
Another participant attributed consumption of these foods to weight Participants were asked about the types of foods eaten at home, and more specifically what their partners and children consumed. As the main food preparers, the participants shared coping mechanisms to encourage their family members to eat more "good" foods such as vegetables. One participant shared that she disguised vegetables by blending them into sauces, juices, and soups to ensure that children would eat the vegetables. Because a diverse range of fruit was not readily available in the town, participants utilized food preservation methods (such as freezing) when they encountered seasonal fruit at grocery stores outside of town or during produce truck visits.

Moringa in Agua Caliente Nueva and in the home
Previous studies reveal that moringa arrived on the Jalisco coast through two mechanisms: through traditional horticulture that has taken place over centuries and through recent imports from India.
In probing what the study participants understood about how the plant arrived in their town, most participants shared that they first learned about moringa from one of the study team (MEO), who maintains the International Moringa Germplasm Collection. One participant shared an anecdote of a man from a town called Arroyo Seco, in the same ejido as Agua Caliente Nueva, who traveled to Navojoa, Sonora, a city approximately 16 hours north along the coast from Agua Caliente Nueva, and returned to the town to distribute Moringa seeds to town residents. According to her, these residents shared seeds with their social networks and moringa eventually arrived in Agua Caliente Nueva. Herbarium records document the presence of the plant being cultivated as an ornamental in the area for at least 50 years (Lott, 1985).
All the participants were aware that moringa existed in the community and knew how to identify the plant when walking around the town, referenced in Table 3. Some shared that they would bring moringa seed pods home to take the seeds as pills, while others stated that their families or friends would bring seeds for them to take.
When prompted about uses for moringa, participants provided many different ailments for which moringa could be used, documented in

Food in Agua Caliente Nueva
Nearly all participants reported that fruit and vegetables were considered "good" foods. They mentioned that vegetables were good for digestion, preventing illness, and one participant perceived vegetables as having properties that helped expel mercury from the body if one consumed seafood. Some other "good" foods that were named included legumes, cereals, and eggs.
Participants perceived a variety of foods as "bad," and although some shared the perception that tortillas were unhealthy, they were still an essential component to household meals. Traditionally, tortillas and beans are a central part of the Mexican diet, which was demonstrated by their availability in the corner stores and the tortilla delivery service in the town (Kimoto et al., 2014). Carbohydrates (such as flour, corn, or bean tortillas for instance) are a necessary component to one's diet, both for human development and for energy to complete daily activities (Hardy et al., 2015). Some participants correlated excess simple carbohydrate consumption with overweight or obesity prevalence in their community, but incorrectly assumed that consumption of tortillas was bad in all regards. This may be due to participants' information-seeking behaviors and identifying sources to which they deemed reliable.
The global food system has led to some shifts in the Latin American food environment, including the retail sales of food. Packaged and processed foods are produced in large volumes and featured more prominently in stores, which contribute to the shifts in diets, increased consumption of processed foods such as canned goods, and subsequent rise in obesity rates in countries like Mexico (Corvalán et al., 2017;Popkin, 2014;Popkin & Reardon, 2018). Interviews with participants in this study indicate a willingness to incorporate other nutritious components such as moringa to make their meals healthier.
There were different sources from which participants obtained information regarding food; some learned through the health clinic in town, others from the internet, and also from knowledge passed down from family members such as their mothers and grandmothers. This information-seeking behavior indicates initiative taken by the study population to look up information about food but raises concerns about the validity of the information that they gather. Additional factors to consider include the varying levels of access to sources of information, whether credible, to identify "good" versus "bad" foods. Although the internet can serve as a method for accessing accurate information, it may also be a potential avenue for spreading inaccurate data, or misinformation, to individuals.

Women's roles in the household
Throughout the data collection process, the participants' responses indicated that they were the decision-makers when it came to food purchases from the corner stores in town and from the produce sellers who drove their small truckloads of fresh fruit and vegetables into town twice a week, as well as when it came to meal preparations for their children and partners.
Many dishes were prepared based on what was available, such as the tomatoes and onions that were sold at the corner stores, or fresh seafood caught from the neighboring seacoast. Participants shared a range of methods to incorporate "good" foods into household meals, and tried similar strategies when incorporating moringa leaflets to their dishes. Their knowledge of their household members' dietary preferences, food preparation prowess, and creativity demonstrated through this study suggest that researchers should engage women for future nutrition interventions (Allen & Sachs, 2007;Bee, 2014;Kimoto et al., 2014;Quisumbing et al., 2014).

Strengths
There is demonstrable evidence that moringa is widely recognized in the community for its medicinal properties, and this study suggests that it can be recognized for its nutritive properties as well. The Agua Caliente Nueva community of fewer than 800 residents is close-knit, allowing for the sharing of information and knowledge about moringa.
Although moringa was introduced to this community, it has the ability to thrive with only minimal care. As researchers continue developing cost-effective, sustainable nutrition interventions to improve the nutritional status of vulnerable populations, they should consider identifying and utilizing indigenous plants like moringa to address the micronutrient deficiencies that are so rampant in those areas and that are so nicely addressed by adding moringa to the diet.

Limitations
Due to the time constraints of this project, the research was conducted within a two-month time frame. Given the time of year for this study (July to August 2017), some residents were not in town and unavailable to participate in the study, resulting in a smaller source population from which to recruit eligible participants. With this small sample size, it is plausible that data saturation was not reached and that new information and themes could emerge from additional data with additional participants. However, small sample size notwithstanding, the population of Agua Caliente Nueva is so small, close-knit, and relatively homogeneous that these findings may be useful for other members of the community seeking inspiration for future small-scale nutrition interventions. Larger sample sizes and longer duration interventions will aid in assessing the feasibility that this community and others like it may embrace moringa as a nutritious vegetable that can be incorporated into local cuisine.
As a new resident of the community, the primary researcher onsite (BCC) spent the first three weeks building rapport by visiting local tiendas and homes to allow the local community members to familiarize themselves with her presence and the reason for her stay. During the interviews, it was sensed that social desirability bias came into play with some participant responses; participants may have provided answers to interview questions based on what they determined was relevant to the researcher and the study. These instances were noted in the written field notes and reviewed during the analytic memo writing sessions.
Participants were provided two loosely packed quart-sized bags of

DEDICATION
The researchers dedicate this contribution to Balbir Mathur, the visionary founder of Trees for Life International.

AUTHORS' CONTRIBUTIONS
BCC, JWF, and MEO co-designed the study aims and objectives. BCC developed the data collection tools, collected the data in the field, and prepared the first draft of the manuscript. MEO and JWF initiated the study. All authors refined the study and participated in the data analysis and revisions of the manuscript, and all authors approve of this manuscript.

CONFLICTS OF INTEREST
BCC and MEO declare that they have no conflicts of interest to disclose. JWF has consulted for both food and supplement companies in the past year.