Discussion

In this study within a medium-sized southeastern city of 63 adults from diverse, low-income communities at high risk for FI, we found some dietary preferences in common: chicken, pasta, and beef were the most popular home-cooked dishes among adults and children alike. However, there were variations in preparation methods between English- and Spanish-speaking participants. Of all produce offered, adults enjoyed more vegetables, while reporting that children preferred more fruit. Lastly, families had little interest in learning new recipes or cooking methods at home and primarily used stoves.

Compared to individuals living in food-secure households, individuals living in food-insecure households have less healthy dietary patterns. Individuals in food-insecure households generally have a higher consumption of more palatable foods such as high-fat dairy and salty snacks, more red/processed meat, decreased frequency of fat-lowering behaviors, and less vegetable consumption [Reference Mello, Gans, Risica, Kirtania, Strolla and Fournier3, Reference Hanson and Connor7, Reference Leung, Epel, Ritchie, Crawford and Laraia28, Reference Kirkpatrick and Tarasuk29]. Our sample population had food preferences similar to what is described in the literature, with a high intake of calorie-dense foods and low consumption of vegetables. This could be because these foods, in comparison to fresh produce, may be cheaper, have longer shelf lives, provide more sustenance, and are easier to access within food deserts. Children have similar dietary patterns to adults in the same households, as seen in our study. Prior research has shown that children from food-insecure households have a higher likelihood of eating fast food, and can develop unhealthy eating patterns due to intermittent availability of food and increased stress levels [Reference Widome, Neumark-Sztainer, Hannan, Haines and Story30–32]. While much of the food consumed was similar between English- and Spanish-speaking households, it was often prepared differently. Meat is typically held in high regard by many cultures, while other food items may have varying levels of significance [Reference Allen and Baines33]. It is important to understand the food preferences of particular groups of people, including culturally relevant foods (based on language, race/ethnicity, or religion, for example), so that community-based hunger relief programs, such as HOPE, can provide not only healthy, but culturally acceptable food.

Fresh produce preferences differed among adult and pediatric populations. We found that adults listed more vegetables, while children overwhelmingly liked fruit. This is consistent with literature showing that younger children rate fatty foods, sugary foods, and fruit the highest, and generally dislike vegetables [Reference Cooke and Wardle34, Reference Edwards and Hartwell35]. While the relationship of fruit and vegetable intake to weight has shown mixed results, various studies have shown an inverse association with intake and type II diabetes mellitus risk, visceral fat, liver fat, and insulin resistance in Hispanic/Latino children, thus making this a potentially important variable to evaluate in pediatric health [Reference Cooper36, Reference Cook, O’Reilly, Goran, Weigensberg, Sprutijit-Metz and Davis37]. It has been shown that food preferences of parents, early introduction, and increased exposure of foods such as vegetables are important in increasing children’s preferences for such foods [Reference Skinner, Carruth, Bounds and Ziegler38, Reference Fildes, van Jaarsveld, Wardle and Cooke39]. Fruit and vegetable intake has also been directly correlated with its accessibility and availability at home, school, local markets, and the inclusion of children in cooking meals at home with their families [Reference Edwards and Hartwell35, Reference Chu, Storey and Veugelers40, Reference Bere and Klepp41]. In this manner, HOPE and other mobile meal programs may attempt to increase accessibility, availability, and visibility of fresh produce to encourage increased consumption by children and adults alike. Further research is needed to understand how community-based hunger relief programs can provide better access to fresh produce and encourage the intake of produce among food-insecure families.

One surprising finding in our study was that participants expressed little interest in learning new recipes or cooking methods. This may be due to financial constraints, time, and effort needed to learn to cook new recipes, and no guarantee that everyone in the household will eat a novel meal. In fact, one study showed that FI was associated with a lack of mealtime planning [Reference Fiese, Gundersen, Koester and Jones42]. Other studies, such as Landers et al. [Reference Landers and Shults43], found that food-insecure families want to learn low-cost recipes such as soups based on canned stock or bouillon cubes, stir-fry, and one-dish skillet dinners that can be prepared in 30 min or less. Lack of supplies may also be a barrier to new methods of cooking, as literature has found that food-insecure households typically own fewer kitchen supplies than their food-secure counterparts [Reference Oakley, Nikolaus, Ellison and Nichols-Richardon44]. The presence of food preparation supplies is associated with increased family meal frequency and child consumption of home-prepared dinners. Improving access to these supplies could be an important target in improving food-insecure families’ diets, as eating more family meals at home has been associated with the reduced caloric beverage and greater fruit and vegetable intake [Reference Appelhans, Waring, Schneider and Pagoto45]. Literature shows that there may be areas of intervention such as teaching time-saving and cost-effective recipes, educating about meal planning, and providing more basic kitchen supplies to families. Further research is still needed to understand how families with limited budgets and time can eat healthier meals. Given the current literature and the results of our study, HOPE is encouraging local farmers to grow the produce that families reported they preferred and modifying the produce they provide to better address the needs of the families that utilize their services; furthermore, HOPE has made changes to the prepared meals in response to participant feedback such as ingredients in sandwiches and snacks. Additionally, a future development for HOPE is to begin offering cooking classes with recipes incorporating delivered produce that can generally be prepared within 30 min. Future studies and incentive programs could also consider distributing cooking utensils with related educational programming to encourage families to prepare more meals at home if they have successfully used the tools and successfully made a meal they enjoy.

There are several limitations to our study that should be acknowledged. First, although we tried to identify a diverse sample of adults from neighborhoods at high risk of FI, our small sample size may not be generalizable to other communities. Second, parents who agreed to participate in the study may not be representative of the populations in these neighborhoods due to selection bias. Third, because we were conducting interviews at the time participants were receiving services at HOPE, we limited the number of demographics collected for each participant to reduce the time of the interviews. We are unable to draw conclusions about differences in food preferences by demographics such as race/ethnicity. Lastly, due to stigma, we may have underestimated the risk of FI in participants when asking if they could afford the type of food they enjoyed, as they were asked verbally with other community members in proximity. Furthermore, we did not screen for FI using a validated instrument.