This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR mhealth and uhealth, is properly cited. The complete bibliographic information, a link to the original publication on http://mhealth.jmir.org/, as well as this copyright and license information must be included.
Health behaviors among low-income groups have become a major issue in the context of increasing social inequalities. The low-income population is less likely to be receptive to nutritional recommendations, but providing cooking advice could be more effective. In this domain, taking advantage of digital devices can be a bonus with its own challenges.
The aim of this study was to develop and deploy NutCracker, a social network–based cooking app for low-income population, including cooking tips and nutritional advices, aiming at creating small online communities. We further determined the usefulness, perceptions, barriers, and motivators to use NutCracker.
The smartphone app, designed jointly with beneficiaries of the social emergency services, was implemented in a disadvantaged neighborhood of Magny, (Paris region, France). Once the app became available, 28 subjects, living in the neighborhood, tested the app for a 6-month period. Logs to the app and usages were collected by the software. In total, 12 in-depth, semistructured interviews were conducted among the users and the social workers to analyze their uses and perceptions of the app relative to their interest in cooking, cooking skills, socioeconomic constraints, and social integration. These interviews were compared with 21 supplementary interviews conducted among low-income individuals in the general population.
NutCracker was developed as a social network–based app, and it includes cooking tips, nutritional advice, and Web-based quizzes. We identified barriers to uses (especially technical barriers, lack of knowledge in the field of new technologies and written comprehension, and search for real contacts) and motivators (in particular, good social integration, previous use of social networks, and help of children as intermediaries). Cooking skills were both a barrier and a lever.
Targeting the low-income groups through a cooking app to promote healthier behaviors offers many advantages but has not been fully explored. However, the barriers in low-income milieu remain high, especially among the less socially integrated strata. Lessons from this intervention allow us to identify barriers and possible levers to improve nutrition promotion and awareness in deprived areas, especially in the time of social crisis.
Current health promotion campaigns that aim at motivating individuals toward healthier eating habits are inefficient among low-income populations. This situation is particularly concerning in the current context where social inequalities in health are very marked and are rising in western countries [
Moreover, public health campaigns, developed by the French National Nutrition and Health Program, focusing on the effect of a proper diet on health, were not able to reach people in low-income categories. Although the disadvantaged populations are well aware of the guidelines for healthy eating [
Consequently, we observe an unequal prevalence of obesity [
There is need of more efficient strategies to enhance nutritional awareness and promote healthy behaviors in disadvantaged populations through novel interventions specifically targeting them. Hence, the use of interventions that include a cooking component seem to be an effective way to enhance the adoption of better eating behaviors [
Digital technologies could offer a good platform to help build and disseminate appropriate individualized recommendations targeting special social groups. As described by the European Commission’s Green Paper on mobile health (mHealth) [
The project took place in disadvantaged neighborhoods of Magny (Yvelines, Paris Region, France). This city had a total population of 32,639 inhabitants in 2015 [
In this study, we aimed to (1) present the coconstruction of NutCracker, specifically designed for low-income people, promoting a healthy diet; (2) describe the results of its usage and determine the barriers and levers of this process; and (3) discuss the main lessons that can be drawn from this study.
Our study was composed of 2 parts—an app design and an app evaluation—and was based on 2 subsamples.
We worked with the disadvantaged households of Magny, Paris region, France to design a mobile app for healthy cooking and nutrition. The app was jointly designed with its future beneficiaries in a 2-step process. First, questionnaires were distributed among 46 participants with the help of the local social workers. The questionnaire focused on the socioeconomic status of the respondents, the foods commonly consumed, food supply, expectations and general perception toward food, importance of eating, importance granted to nutritional issues, level of culinary skills, and the availability of cooking equipment. Second, a collective discussion with 8 participants was organized to discuss about a new mobile phone app providing healthy and affordable cooking advices and recipes to determine the participants’ expectations toward such a device. The objective of this discussion was to launch a collective dynamic as well as to reach a consensus on the overall design of the mobile app and provide general recommendations on its features. We used the
When this app was released, it was tested by 28 volunteers (8 participants from the collective discussion plus 20 supplementary volunteers) from the same neighborhood and recruited in collaboration with beneficiaries of 3 social emergency services in Magny.
Data on individual app usage (logs and visited features) were collected, and a qualitative study was performed to assess this app.
We conducted 33 semistructured interviews using the 32-item Consolidated Criteria for Reporting Qualitative Research checklist [
Along with this sample, and to compare with digital practices in low-income population, 21 interviews were conducted in the Paris region, among individuals of the lower strata (
These interviews, conducted by the 2 project sociologists, lasted on average of 1 hour, were face-to-face interviews, were recorded (except for 1 participant who refused), transcribed in their entirety by a team of transcription consultants, and anonymized. Analyses of the data were double-checked by the 2 sociologists, and then discussed and validated with the team involved in the project (ie, the 4 researchers who were directly involved and the scientific committee of 5 experts). We developed a content analysis, and the 3 main themes investigated using interview guide were as follows: (1) uses; (2) barriers and levers in using NutCracker; and (3) digital devices in the field of cooking, diet, and other related areas. Among the themes, subthemes were identified in line with the themes in the interview guide, and new themes derived from the data collected were also included.
Our study was based on 2 subsamples (
In the NutCracker sample (
For the collective discussion, participants were informed about the purpose of the discussion before the interview. For the semistructured interviews, the goals of this study were explained to the interviewees, and individual consent was obtained for the recording. Personal data from the interviews, including the name of the place the project took place in, were anonymized.
Subsamples of the study.
Sociodemographic characteristics of the participants.
Sociodemographic variables | NutCracker target population (n=8), n (%) | NutCracker social workers (n=4), n (%) | Low-income milieus in the general population (n=21), n (%) | Total | |
Male | 0 (0) | 2 (50) | 3 (14) | 5 | |
Female | 8 (100) | 2 (50) | 18 (86) | 28 | |
19-29 | 0 (0) | 1 (25) | 7 (33) | 8 | |
30-39 | 2 (25) | 2 (50) | 5 (24) | 9 | |
40-49 | 5 (63) | 1 (25) | 5 (24) | 11 | |
50+ | 1 (13) | 0 (0) | 4 (19) | 5 | |
Intermediate professions | 0 (0) | 4 (100) | 0 (0) | 4 | |
Employee | 2 (25) | N/Aa | 12 (57) | 14 | |
Manual workers | 0 (0) | N/A | 4 (19) | 4 | |
Unemployed | 6 (75) | N/A | 2 (10) | 8 | |
Students | 0 (0) | N/A | 3 (14) | 3 | |
Total | 8 (100) | 4 (100) | 21 (100) | 33 |
aN/A: not applicable.
The results from analyzing the 46 questionnaires emphasized the importance of food abundance and variety, the desire to please children with food, and the value put on national brands as markers of food quality. Although generic nutrition recommendations are very well known, the importance of nutritional issues, such as the criteria for food choices, appeared to be less important for the low-income population than for people with higher social ranking. Respondents declared the frequent use of mobile technologies, with 46 owning a smartphone (ie, 72%, whereas the national average was 58% in 2015 [
Once the app was created, it was released and made available for download to all study participants. A total of 28 households who met the selection criteria mentioned above were selected to use the mobile app over a period of 6 months (from May 2016 to October 2016).
The postintervention field research highlighted a very limited use of NutCracker among its intended target population, namely, the women (
By contrast, the app was much more successful among the general population, where it was shared through a network of interconnections. Between May 2016 and October 2016, 378 connections were made, 600 recipes were viewed, 18 recipes were shared via the app, and responses to 1200 quiz question were supplied by the users.
Generally, participants from Magny had more limited access to the internet and to new technologies than participants from the low-income milieus in the general population (
On the basis of the inductive thematic analysis methodology described in the study by Peng et al [
The most frequent barrier cited during the interviews was the technical instability of the prototype NutCracker leading to recurrent interruptions during the sessions. Another factor was the necessity to provide user name and password at every session as well as not being able to quickly reset the password.
Uses of NutCracker and new technologies by the NutCracker sample and the low-income milieu population.
Uses | Verbatim | NutCracker target population (n=8), n (%) | Low-income milieus in the general population (n=21), n (%) |
NutCracker app | “[NutCracker], I don’t have it on my phone anymore, I don’t know why. I had downloaded it and it was really interesting.” | 4 (50) | Not applicable |
Cooking apps or websites | “Marmiton website...it’s complete, and it works really well.” | 5 (63) | 14 (67) |
Web-based trade | “I sell things on eBay.” | 0 (0) | 5 (24) |
Web-based games | “I play poker on the Internet.” | 0 (0) | 3 (14) |
Administrative Web-based tasks | “My bank accounts, I have everything, the RATP, I have WhatsApp, etc. and I even have Carrefour, Wish, I have my accounts, I have Mappy to get around, I have music stuff.” | 4 (50) | 11 (52) |
Diet or physical activity self-tracking | “I use the app MyFitnessPal.” | 0 (0) | 1 (5) |
Barriers to using NutCracker and the new technologies by the Nutcracker sample and the low-income milieu population.
Barriers | Verbatim | NutCracker target population (n=8), n (%) | Low-income milieus in the general population (n=21), n (%) |
Technical barriers | “I don’t think my telephone is good enough...I will change it, but I don’t know when [...] It’s another expense.” | 3 (75)a | 10 (48) |
Unfamiliarity with new technologies | “Administration services computerize, but people are lost. Not everyone has an email address.” | 4 (50) | 5 (24) |
Comprehension and written barriers | “We can speak French well, we know how to write, but there are things that we have trouble deciphering.” | 4 (50) | 2 (10) |
Time constraints | “As I work mornings and evenings, I don’t have a lot of time. [...] I don’t have the time.” | 3 (38) | 3 (14) |
Competition from television | “I watch (the Samira channel) all the time - as soon as my children leave, I watch it.” | 3 (38) | 2 (10) |
Search for real contacts | “Cooking workshops give you the opportunity to get out from of home and meet other people.” | 8 (100) | 1 (5) |
Live social network | “So, do you like it (what I published)? I’m the only one who publish, it’s a shame.” | 1 (13) | Not applicable |
Fear of being stalked on the internet | “There are always dangers. And I’m scared of...[...] I don’t like it, it’s scary. And...well [...] it can cause damage.” | 4 (50) | 4 (19) |
Interest for cooking and cooking skills | “I have everything in my head.” | 6 (75) | 13 (62) |
aWe only considered the 4 participants who used the app.
These technical barriers, although unique to NutCracker, reflect an exacerbated trend of limited use or underperformance, if it was not a defective equipment in the low-income milieus. A total of 9 individuals mentioned these issues, which were essentially caused by the insufficient availability of memory or an overtaxing app, because most individuals preferred to keep what space they had for personal photos or videos:
I don’t have many apps on my telephone, I have to buy a card, I don’t have enough memory. All this stuff takes space!
A lack of knowledge on new technologies was also an obstacle. In the least well-integrated fringes of low-income categories (unemployed women and education level lower than a bachelor degree), everyday internet tasks are neglected, unless they attend specific training programs in social centers where they learn how to create and use an email account and a password on a computer before moving to the mobile phone usage.
All women in the NutCracker intervention group were foreigners, and some of them only had a basic command of French or were not comfortable with the written French (4 of 8). These 2 barriers explained both the low rate of usage of NutCracker and the overall use of digital technologies:
I am taking French courses [...] because there are some words that I don’t understand. I can read, but I don’t understand. [...] I have some difficulty.
This lack of knowledge regarding the new technologies did not appear as a real barrier to our sample in the general population.
The project participants who did not download NutCracker were characterized by a few financial constraints as they were either unemployed or earned a very low salary. They also had time constraints. For the unemployed women, domestic tasks, most notably those with large families, were cited as barriers to using NutCracker. Eveline and Hawa were employed, but they had long commutes to work, and their workdays as housekeepers were split in 2 (early mornings and evenings).
In Magny, 3 of the women interviewed expressed a preference for passive media (television), which allowed them to have their hands free, unlike a smartphone, tablet, or computer. It gave them the opportunity to do household chores, with the television playing in the background.
Regarding cooking more specifically, and for the most underprivileged participants in our study, television was the main source of culinary expertise: television channels, such as the Algerian channel Samira, which is entirely devoted to cooking and gastronomy, were greatly valued.
Among the participants from the general population, the television rarely prevented them from using digital devices.
The NutCracker app was aimed to connect women from underprivileged areas to Web-based micronetwork; however, the interviews with these women showed that they preferred to socialize
This refusal of being part of virtual socializing was observed in only 1 participant from the low-income milieus sample; she was better positioned socially, but her real-life socializing was affected by a break-up. Following her divorce, Elga was seeking stronger social ties:
There are already so many things that we can buy on the internet, if we add food to that, we will never leave the house. I need to get out of the house.
The small number of members of a living social network prevented the creation of an active and thriving social network. Individuals, such as Martine (NutCracker, 39 years, housewife, 3 children), who were the most enthusiastic, found themselves to be quite alone in their investment in NutCracker, which led them to stop using the app. For example, in her interview, Martine said that her enthusiasm waned once she realized that she was one of the few active people in the network and that no one was responding to her posts.
The NutCracker app depended on women publishing their recipes; however, women expressed reluctance to publishing recipes and to sharing their own experience. Furthermore, putting one’s own recipes on the Web implies exposing oneself, as food is closely connected to identity, whether individual or social [
This also explains why publishing recipes as part of a group in a workshop was much easier than doing so individually. Instead of involving risky personal exposure, the collective posting was perceived as developing a collective identity:
...putting up their recipes, they were happy, it was gratifying.
In the general population, participants more rarely mentioned the fear of a digital tracking (4 of 21), and in these cases, individuals expressed a fear of being overwhelmed by the internet and becoming addicted to digital tools. Furthermore, participants from a better social position presented limiting the use of digital tools as a choice; thus, the refusal of digital technologies was chosen, and not endured.
The results below show some levers for using NutCracker and new technologies (
As expected, the participants who were more familiar with new technologies were the ones who most easily took to the NutCracker app, which was related to the social position of the individuals. They had a slightly higher social status (most notably through their husbands’ professions in the case of the 2 most active users); a higher level of education (high school qualification), which in itself requires a certain familiarity with digital and information technology devices; a functional equipment (a powerful smartphone or a home personal computer); and a keen interest in cooking. This outcome was confirmed by our general population sample, where participants had higher social positions (in terms of occupation or level of education), and almost all participants used the internet and new technologies.
The participants who were most active in NutCracker were also those who were already familiar with social network apps most used in low-income areas (
Furthermore, Martine characterized herself through her position as an intermediary, which drove her to invest herself in the digital culinary realm, whether on NutCracker or, previously, other sites or social networks (especially Marmiton or her personal Facebook page where she published recipes). Originally from Senegal, Martine arrived in France a few days after she was born:
I’m familiar with French culture, as well as my homeland’s culture, so I quite enjoy a kind of mixing of both cultures and all, yes, that’s me.
This distinguished her from the other participants who had arrived in France much later in life, and often as adults. This sense of double belonging, associated with pride in her 2 cultures, led her to see cooking as a way of exchanging and discovering new cultures. This positive self-image also helped her feel more comfortable publishing her personal recipes on the Web.
The interview study highlighted the role of children as essential intermediaries for digital technologies as they were regarded as the experts in the field. All the more, as there is a tendency for the underprivileged to please their kids with the symbols of consumer society [
I know how to use it, but there are a lot of things we don’t know [...] but young people, they know everything.
Eveline’s case was interesting as well:
It’s my daughter who knows, because she’s the one who installed my [Viber] app (...) She knows how to do everything on the internet.
Among the low-income milieus in general, children appear as mediators, and they may help in case of technical difficulties, but their role was not as essential as in the deprived households, and they were rarely mentioned in the discourses we collected (once only).
Levers for using NutCracker and new technologies by the NutCracker sample and the low-income milieu population.
Levers | Verbatim | NutCracker target population (n=8), n (%) | Low-income milieus in the general population (n=21), n (%) |
Social integration | “I’m an outgoing person...I’m too outgoing [...] I’m organizing a pastry competition (at school) for the mums.” | 2 (25) | 17 (81) |
Knowledge of new technologies | “Compared to before, we had no Internet...we just had a blank shit [...] Now, everything is | 4 (50) | 18 (86) |
Familiarity with other social networks | “I already had a Facebook page precisely on cooking, I was used to publish recipes every week.” | 6 (75) | 10 (48) |
Children acting as intermediaries | “It’s the kids who know about that.” | 2 (25) | 2 (10) |
Digital and modernity | “I’m a geek of my smartphone.” | 1 (13) | 5 (24) |
Videos on internet | “But there’s everything on YouTube [...] You just type “chicken” or “Tajine” and it gives you [...] there are tons of different videos.” | 3 (38) | 7 (33) |
Interest for cooking and cooking skills | “Maybe I’ll find some ideas (on the app) in order to make my children eat vegetables.” | 6 (75) | 15 (71) |
Those participants who were the least comfortable with the French language preferred using visual material on the internet or in apps to increase their culinary expertise (3 of 8 in the NutCracker group); watching cooking videos on YouTube was preferred over using websites or apps that were predominantly text-based. Thus, Fatima (Magny), who had difficulty with the French language, used YouTube videos—the links were sent to her by her friends via WhatsApp:
I start the video, I put it on full screen (...) it’s faster, it’s easier.
In addition, when she looked something up on the internet, she used the
This approach was also a trend we observed in low-income milieus in the general population (eg, Rabia, 38 years, employee: “I watch videos (on YouTube) for recipes”) that was exacerbated in Magny.
Cooking skills appear as both a barrier and a lever. The 2 women who actually used NutCracker had a keen interest in cooking—for Martine, the appeal was more on the
Maybe I’ll find a way to get my children to eat my food, vegetables included. [...] There are a lot of moms who have trouble getting their kids to eat vegetables.
On the other end of the spectrum, great culinary traditions based on oral transmission rendered the use of the internet unnecessary and unattractive when it came to cooking. Eveline echoed this when explaining why she did not need the internet to nourish her culinary inspiration:
It’s all in my head.
Exactly the same words (“it’s all in my head”) were expressed in the general population by Savina (45 years, old manual worker).
Furthermore, in this specific social milieu, cooking was not necessarily connected to the digital domain, as emphasized by a social worker in Magny:
...making the connection with the app is very, very difficult.
Cooking is the realm of the tangible and the emotional, which is disconnected from the digital world. It is difficult to determine, at this stage of our investigation, whether this separation is specific to people from underprivileged groups or not.
Finally, and somewhat unexpectedly, those who were completely uninterested in cooking were the ones who demonstrated the most interest in the app; they figured that the digital tool could provide them with ideas to turn what they saw as a tedious task into something more fun. For Clarisse, cooking was like:
a chore; when we get home from work, we’re tired.
Therefore, regarding an app:
Yes, why not? It could provide some ideas.
NutCracker, developed as a social network-based app, includes cooking tips, nutritional advices, and Web-based quizzes. We identified barriers to its uses (including technical barriers, lack of knowledge in the field of new technologies and written comprehension, and search for real contacts) and motivators (in particular, good social integration, previous use of social networks, and help of children as intermediaries). Cooking skills were both a barrier and a lever. Although the designers of the project had great expectations and foresaw a high membership rate, the real results may appear disappointing. Several explanations can shed light on this discrepancy. To begin with, a social desirability bias was linked to the group’s situation in the collective discussion, where the participants found it difficult to express reluctance toward a project concerning a pleasant and well-liked area such as cooking and the symbols of modernity, that is, digital tools. Furthermore, the participation of these women in this project revealed a desire to move upwards socially in very underprivileged areas by gaining access to information deemed good and desirable (eating healthy, at a low cost) as these recommendations by the National Program for Nutrition and Health have been widely promoted since 2001 [
The reality of the field study itself also revealed some barriers. First, those related to the slowness or capacity issues of the devices, especially because smartphones were often the only internet access for the households in our study. Our work corroborates previous studies, which noted that the speed the apps use was critical to the satisfaction of users [
We have also added some new elements to the digital divide such as the difficulties experienced by our participants, which reflect the working class’s unfamiliarity with new technologies. Our results demonstrated that the lowest percentile of the low-income population, which is the least socially integrated, are at greater risk of digital inequalities, which is in line with previous reports [
Our study confirms a gap in the use of digital self-tracking tools. Although individuals from well-off backgrounds choose to use these tools [
This research also confirms the ability of this app to act as a culinary lever to promote healthier eating habits [
However, the connection between cooking and the digital world is very weak and should be explored further in future. Our results highlight the widespread reliance of the target population on videos, which allow them to overcome the written-language barrier and to rematerialize cooking through direct access to tips and tricks, emphasizing the significance of visual material. The use of videos to develop cooking skills [
Finally, we have demonstrated the difficulty experienced by women of very underprivileged backgrounds to integrate and participate in online social networks, barring women who were already well-integrated in their real world and were active on social media. Social inequalities were increased by gender inequalities. Our results corroborate studies performed on digital technologies that highlight the complementarity and overlapping between real-world and Web-based sociability [
The app was disseminated among a small sample of participants: this prevented the creation of an active social network. Our current research project FacilEat4All—a continuation of NutCracker—is based on a broader community (approximately 100 volunteers). Another limitation of the NutCracker project is that the choice of the participants, which was limited to individuals from underprivileged backgrounds, led to the selection of people with heavy combination of constraints (budget, social integration, and language). The FacilEat4All project now includes people from modest categories with less difficulties (white or blue collars, in socially more diverse areas).
Finally, the method of face-to-face interview was particularly interesting for the postintervention study to have access to individual opinions on the app. On the contrary, collective discussions appeared as useful tool to launch a collective dynamic for the project, albeit the group’s discussion led to a general consensus regarding the expectations concerning the app itself.
In the most underprivileged milieus, it was particularly difficult to motivate individuals to test the app over a long term (3 months) and to participate in the social network. One conclusion of this limitation is that in those underprivileged milieus, digital devices are more easily accepted if used as collective tools, in workshops for example.
The NutCracker and FacilEat4all projects bring new elements to a theme that had not been studied closely until now. They promote healthy eating through culinary levers and digital tools. The NutCracker app and the study of the use and perception of digital
Our study also shed light on the levers we rely on, such as prior experience and use of social networks, which led certain participants to become leaders. We also observed the importance of children as intermediaries of new technologies. Finally, our research has demonstrated the importance of culinary levers in the development of digital tools for people in the low-income categories. Further interventions should assess the advantages of a cooking-based communication platform to promote healthier behaviors.
Screenshots of a website; SPSS files containing original data.
mobile health
nongovernmental organization
The authors are very grateful to the
FR developed the protocol for the field study, performed a part of the interviews, analyzed the interviews, and wrote the manuscript. MD conceived the pedagogical program, prepared, and distributed the questionnaires, prepared and conducted the collective discussion, prepared the app-design. ND developed the protocol for the questionnaires and collective discussion, and wrote the manuscript. CA performed a part of the interviews, analyzed the interviews, and participated in writing the manuscript.
None declared.