Study desing and sampling

This qualitative study analyzed results of 12 focus groups performed between June and August of 2015 in the largest and most populated cities of Mexico (Guadalajara, Mexico City, Monterrey, and Querétaro). A purposive sampling of adults exiting supermarkets, who answered an in-site questionnaire (not reported in this paper), were invited to participate in focus groups. Three focus groups were conducted per city and were stratified by SES (low, medium, and high SES). We selected a sample from a list of all supermarkets in each city to stratify supermarkets, hypermarkets, membership food stores, and convenience stores by SES. Those supermarkets were mapped using a geo-reference system to determine the Basic Geostatistical Areas (AGEB by its acronym in Spanish) where they were located. AGEBs are a proxy-estimation of socio-demographic characteristics of the area in each city, they are specific and delimitate urban areas with 2 500 inhabitants or more, and are used to locate specific socio-demographic conditions such as living, commercial, and industrial usage, among others. They are a proxy estimation of the socio-economic characteristics of areas in each city.³³ The supermarkets in each AGEB were randomly selected and proportional to the distribution of three levels of marginalization defined by the National Institutes of Statistics and Geography on a scale of low, middle, and high.^34,35 Trained staff with health background (dietitians) recruited the participants and collected the data for the study. The training was conducted during a whole day session. From 9 am to 5 pm researchers instructed fieldworkers about how to approach subjects. They performed two exercises to practice before doing the fieldwork. Fieldworkers asked participants if they make their usual grocery shopping in that supermarket, if they answered that they did, and their self-reported family income (reported in quantitative survey not reported in this paper) matched the SES of the supermarket they were recruited to participate in the focus groups. We looked for a homogeneous sampling in each focus group in order to make comparisons between statuses and observe differences. Each focus group was integrated by 8 to 10 participants of similar characteristics (age, education, and SES).

Snowballing sampling technique was performed in three focus groups as some of the participants asked to attend with a family member or a friend. Participants were eligible for inclusion only if they were willing to participate, were literate, and were older than 18 years old. We excluded participants who did not shop in grocery stores, those who were older than 65 years old, and those who were visually impaired or blind. Nutrition researchers and nutrition students from the University of Guadalajara, the Mexican National Institute of Public Health, the Autonomous University of Nuevo Leon, and the Health Promotion Department of the Ministry of Health of Querétaro helped with sample recruiting, provided the venue, and transportation. Date and time of the sessions were given in a written format to potential participants outside supermarkets. Focus groups were performed in universities’ classrooms. In Mexico City they were performed in the Mexican National Institute of Public Health and in Querétaro they were performed in the meeting room of the health promotion department. Participants were offered food and beverages during the focus group session for their time and participation.

Ethical considerations

The study was approved by the Ethics and Biosafety Committees of the Mexican National Institute of Public Health (Approval number: 1153). Before starting the focus group session, oral informed consent was obtained from the participants, and was audio-recorded prior to participating in the research. Participants were informed that their identity and data would be anonymous; also, they were informed that they could stop participating at any time of the study.

Focus group guide and tools

A semistructured guide was developed a priori guided by the health belief model³⁶ and the theory of planned behavior^37,38 (supplementary appendix 1, Spanish version).³⁹ The aim of the guide was to prompt questions and explore perceptions towards the use and comprehension of nutrition labeling including GDA, NFT, NS, and claims. Emergent themes during the thematic analysis were also included in results. The focus group guide was piloted with two focus groups. One with participants from high SES recruited in the waiting room of a medical doctor private practice. The other was helded with janitors of a primary school. The guide was revised and corrected accordingly using field notes taken by one researcher and by the moderator. Questions were revised one by one in a full-day meeting. Discussion were helded when the moderator and the assistant agreed that the question was not fully understood. Figure 2 shows the different labels visually presented to participants during the focus group sessions: 1) GDA, 2) Claims, and 3) NS. The real size of the labels was one page A4. Labels were presented in Spanish. Main themes included attitudes towards nutrition labeling, and beliefs and opinions of different labeling formats.

1) GDA: defined according to the Mexican regulation, NOM-051, which states that the mandatory front of the package labeling must display the following nutrients of concern per package: saturated fat, other fats, sugar, sodium and energy. 2) Claims: defined as any representation which states, suggests or implies that a food has particular characteristics relating to its origin, nutritional properties, nature, production, processing, composition and any other quality. 3) Nutritional Stamp: defined according to the Mexican regulation NOM-051. A packaged food will be able to carry the stamp if complies with the nutritional criteria established by the Ministry of Health.

Figure 2 Labels presented to participants during focus group sessions 1) Guideline Daily Amount (GDA), 2) Claims, and 3) Nutritional Stamp (NS) 

Focus groups and data analysis

Three researchers moderated the focus groups (CM, CN, AC). They were trained to guide the focus group similarly and follow the same methodology in all the sessions. One is a medical doctor and the other two are dietitians. An observer was present at all times to help the moderator and take the corresponding field notes. All focus group sessions were audio-recorded with two devices, and lasted between one and two hours. The audiotapes were transcribed verbatim. Transcripts were checked against the recordings to ensure accurate reporting. Data was coded with an inductive approach^.40 Three researchers coded and analyzed the data individually (CM, CN, AC). Coding decisions were made in a consensus meeting. At least two researchers had to agree on coding to make the decision. The key themes were determined before, but researches also took into account themes that emerged from the focus groups (table I). A thematic analysis based on a positivism perspective was conducted.⁴¹ We identified main key themes, within the main themes, several sub-topics emerged as perceptions of the use and comprehension of the different nutrition labeling systems and claims.

Participants’ characteristics

Each of the 12 focus groups had on average eight participants (min five, max 12). Table II shows the sociodemographic characteristics of participants stratified by SES. Most of the participants were female. Participants who reported less years of formal education were mainly in the low SES, while participants in the middle and high SES groups reported to attended high school, college or had a post-graduate degree.

Perception towards nutrition labeling

In the present study, regardless of the participants SES an unfavorable perception emerged towards the current nutrition labels displayed in foods. In general, participants did not acknowledged any outcomes of reading the labels. They mostly reported not to read them because they do not understand them; nevertheless, some of them, acknowledged the importance of reading the “food you eat”.

Perceived advantages/favorable characteristics of nutrition labels

Despite participants’ background and negative perception of the current nutrition labeling system (GDA and NFT in the front and back of the package), participants acknowledged the relevance of having clear information on packaged foods to make healthier choices:

“It is important [to know what a product contains] because at certain age you begin to value more your health and taking care of what has the food you eat” (high SES).

“It is important to know what you are getting into your body” (medium SES).

“It tells you what to eat and the expiration date” (low SES).

Regardless their SES most of the participants mentioned that the information displayed is important due to several factors; some of them, mentioned the age and the concurrent presence of NCDs.

Perceived barriers when using nutrition labels

Regardless of the format, several barriers for using the nutritional labels (GDA and NFT) were identified by participants. These included the complex language used on the labels, the font size, the amount of information displayed on labels, and the lack of trust in the claims made on the package.

First, participants reported that the information is very technical, for example one participant mentioned: “I think most of the times they use other names, more technical names and then people gets confused and doesn’t know what it really is” (medium SES). In addition, participants reported that the font size is barely readable: “It is difficult to understand and the information has very small fonts” (high SES), “Only ants can read it” (low SES). This becomes an issue especially among older adults, as the main characteristics of the product could be difficult to read and to interpret by older population, and people with impaired vision.

Regardless the SES of participants, the opinion of any type of label on packaged foods was perceived as “confusing”, and “difficult to understand”. They also mentioned the lack of trust for claims a constraint for using nutrition labels. Participants perceived the nutrition labeling as a whole, including the GDA, the NFT, and the claims displayed on packaged foods. A member from a low SES focus group mentioned: “They [nutrition labels] are useless, we don’t understand them”. Particularly when referring to the GDA and the percentages displayed for each nutrient, a high SES participant mentioned “for me this percentage is confusing…it doesn’t says well, we don’t know percentage of what…”. A middle SES participant mentioned to partially understand the labels: “I only understand the calories” (medium SES). It was mentioned that some training or knowledge on the topic is needed to use the labels. For example, a participant reported: “If someone knows about this, maybe they can use it and understand it, but if not, it is very misleading” (high SES).

There are many constraints for the use and understanding of the labeling system implemented in Mexico, for both types of labeling, the GDA, and the NFT in the back of the package. Regardless of SES, participants made clear they do not understand the labeling; therefore, they find it useless to read.

Perceptions about the claims on the FOPL

Participants from medium SES and low SES consistently reported to mistrust the claims: “In reality we don’t know if the product has that characteristic or not…” (low SES), “Sometimes the information on the label might be accurate, but I think, most of the times is not reliable” (medium SES), “it might say it has cero fat, but it is obvious that any packaged food has something of that” (low SES).

Likewise, other participants from the high SES and medium SES are not sure that the information is accurate. Consumers from the high SES reported that they have no way to prove or verify the accuracy of such claims. “There is not a monitoring system implemented by the government to check if food products comply with claims they make on the front of the package” (high SES). Some other participants expressed that such claims are misleading: “People think that they can eat big amounts of that because it says light and they are eating fewer calories…” (medium SES). Also, among the high SES, the participants consistently mentioned: “They want to promote their product, with claims, cartoons, or bright colors. The information is about that, and not accurate (high SES)”.

Participants from the low and medium SES generally reported to mistrust the claims displayed in packaged foods. Data from the high SES suggest reliability of claims is compromised with marketing strategies. Participants tended to believe that the main driver for claims is marketing and promotion of a product.

Perceptions about the GDA in the FOPL

When participants were asked about their opinion of the GDA, high SES participants in general mentioned it as a better alternative than the NFT displayed on the back of the package. Nevertheless, they still mentioned it was difficult to understand, especially the percentages: “what I don’t like about them is the percentage; it doesn’t say anything and it is not standardized” (high SES). Participants from medium SES and low SES also believed it was easier to focus your attention and easier to read than the one in the back. In these two groups, their opinion changed when a GDA figure was bring into discussion (supplementary appendix 1),³⁹ and participants were asked to explain it. Participants from the middle SES groups mentioned they understood how many calories the product had; however, they did not know how to interpret the information about sugar, fats and other fats. The term “other fats” caused confusion:

“Other fats content is that these aren’t natural or are more saturated?…” (medium SES).

Most of the participants from the low SES mentioned that the only thing they understand about the GDA is what portion means, but they defined it as “the total amount they should eat in one meal”. Some of the participants were aware of controlling their sugar intake by reading the GDA, mainly people who reported having diabetes, and consequently already attended workshops on diabetes care. Nevertheless, it was difficult for them to report how many grams the product contains, versus the amount of sugar they are allow to eat:

“One portion is the amount you can eat, more portion equals more sugar and more calories, but there is an established portion for diabetics, you cannot eat more than such portion, if you do, your glucose will rise” (low SES).

Participants from the high SES group reported to understand better the GDA labeling; nevertheless, they still found it confusing when looking at the GDA figure:

“Apparently it doesn’t have many calories but it says six portions…when it says calories per portion you may think that 6 is the whole package, or grams or I don’t know” (high SES).

Because of the confusion towards the GDA, participants do not consider this information drives their decision when buying foods. One participant from a low SES focus group declared “Sincerely, no [I don´t use it] because I don’t trust it I don’t use it” (low SES) and a high SES group participant mentioned: “Unless I fully understand [a nutritional label] I don´t use them” (high SES). Barriers to the use and understanding of the GDA were a common theme among all groups regardless their SES. Data show that the GDA in FOPL is not used nor comprehended as intended.

Perceptions about the “nutritional stamp” logo

The majority of participants across SES reported that the NS logo resembled a food plate. Most of them interpreted the stamp as a sign that the product “was healthy enough for you to put it in your plate”. Other examples for the perception of the stamp are:

“For me it means it is inside the nutritional plate, it is good” (low SES).

“I see it like something that is good and healthy” (high SES).

Some participants in the medium SES thought that the stamp resembles the signal of restaurants zones: “with the cutlery I think about restaurants” (medium SES).

Other participants in high SES and medium SES reported it was difficult to understand because the stamp does not say anything by itself.

“It’s easy but it doesn’t mean anything” (medium SES).

“It’s difficult; does not have any sense and says nothing” (high SES).

When asked about the kind of foods that could display this stamp, participants in high SES said prepared foods, canned foods and food for kids, while participants in medium SES and low SES said healthy foods. Some participants among all groups did not mention any foods because they did not know what the stamp actually meant. Researchers were unable to find an agreement among the different SES groups, participants simply do not agree about a meaning for the nutritional stamp.

Conclusion

In general participants have a negative perception towards the nutrition labels regardless their SES, they perceived several barriers to the use and understanding of the GDA; like the technicalities of the terms used, and the format. Also, claims are mistrusted especially among high SES. This data might call attention to policymakers in order to implement a simpler and easier to comprehend nutrition-labeling system that is understood across SES.