Food insecurity was experienced by 11.1% of United States (U.S.) households in 2018, including households in large urban centers such as Baltimore city [1
]. According to Feeding America, about 23% of people residing in Baltimore currently experience food insecurity, including more than 30,000 children [1
]. In Baltimore, there are over 220 food pantries working with the Maryland Food Bank [2
]. Roughly half of these organizations are operated by volunteers at community-based nonprofit organizations such as churches and homeless shelters, whereas the other half are located within Baltimore City Public Schools. Clients of food pantries tend to be food insecure and may be vulnerable to nutritional deficiencies [3
]. The rise in obesity and diet-related diseases among food insecure individuals in the U.S. brings to question the nutritional quality of foods accessible to households that rely on food pantries [4
]. Additionally, the most recent updates to international food guides have also confirmed the importance of dietary healthfulness [5
Feeding America and many food banks in the United States, including the Maryland Food Bank, present the client choice method as a best practice. The main motivation for promoting client choice is to provide a dignified experience to clients. Pantries were classified by distribution method: traditional (distributing pre-packed bags) or client choice (allowing clients to make selections on the foods they receive). Additionally, client-choice can be implemented using a number of different models: the supermarket model (clients can shop like at a store), table model (food items/groups are displayed on tables), inventory list model, and food weight model (clients can select a set poundage of food), among others. A study using a Freshplace intervention included a client-choice pantry in the north end of Hartford and increased fruits and vegetables by one serving per day compared with the traditional group [8
]. However, it is not clear whether client choice influences the overall healthfulness or share of specific groups of client food selections. If it is found that a switch to client choice is independent from the availability of healthy foods in client bags, food assistance organizations will need to take additional measures beyond promoting client choice to promote healthy options to their clients. In addition, food pantry size may also matter while considering the healthfulness of foods clients receive. Pantries of different sizes may have different capacities and resources. For example, larger pantries may have more resources like fridges which are good for healthier foods including fresh fruit and vegetables to be stored. Pantry size may also matter in developing environmental strategies to try and nudge healthier client selections because we can see if different intervention strategies like client-choice or not will have different impacts in different-sized food pantries. It is also unclear whether the pantry size differentiates the healthfulness of client food selections.
The main objective of organizations like food pantries are to minimize chronic hunger. However, the managers of food pantries we spoke with in our previous studies said they were also interested in stocking and providing healthier foods if they knew clients were interested in using them, but perceived barriers associated with it [9
Previous studies have assessed the quality of food distributed in food pantries [10
]. At the pantry level, the quality and quantity of foods accessible to clients may be determined by the food distribution method used in the pantry [14
]. In urban food pantries in the Bronx, NY, USA the nutritional quality of foods available varied by item type (fresh, shelf-stable, refrigerated/frozen), sourcing, distribution method (prefilled bags and client choice), and client position in line. They found that client choice pantries in Bronx, NY, USA had healthier foods available than traditional pantries. However, at client choice pantries, earlier clients selected the less healthy options first. This suggests that a switch to client choice or stocking healthy foods alone might not be sufficient to promote healthy options to clients [10
]. However, the generalizability of these limited findings in other US settings remains unclear. Likewise, no previous studies have investigated the association with many food pantry characteristics such as pantry size and food distribution method together with nutritional quality of foods received by clients [13
]. Thus, in this study, we evaluated the association of the overall nutritional quality and the weight share of specific types of foods received by food pantry clients with food pantry size and distribution method. We assumed that both food pantry size and distribution method (separately or combined) would affect the healthiness of food clients get. By including sociodemographic factors, we also wanted to see if, in a client choice pantry, clients from certain sociodemographic backgrounds selectively preferred healthier foods. In the pantries distributing pre-packed bags, we wanted to see if pantries serving certain sociodemographics were more likely to try to distribute healthier foods. This information would help us identify clients who typically do not get healthier foods at pantries and target them in our intervention messaging in the future.
This is the first study to explore whether food pantry distribution method and size can predict the healthfulness of food obtained by pantry clients in an urban setting. Increasingly, attention is being directed to the client choice model and the need for a healthier food environment for all pantry clients. In client choice pantries, clients are not required to receive items they may already have, may not like, or cannot eat for health or personal reasons, which may decrease food waste and be more efficient for food distribution [8
]. In a past study, a food pantry intervention that involved transforming a traditional pantry to client choice saw improvements in client food insecurity and fruit and vegetable intake [17
Overall FAST scores were not significantly different among clients of different sized pantries and between clients of pantries having different distribution methods, suggesting no noteworthy differences in healthiness of foods clients received by pantry size and distribution methods founded in our study. Thus, the client-choice method may not significantly improve the nutritional quality of foods received by clients compared with traditional prepackaged methods in these seven pantries. However, outside of simply nutritional outcomes, such a method may still catalyze other positive outcomes, such as less food waste and better connection with clients, which may also have salient ramifications for nutrition and health outcomes of food pantry clients.
Among client choice pantries, clients of small pantries received healthier foods than clients of large pantries, followed by medium pantries. This suggests that the size of client-choice pantries impacts nutritional quality of foods received by clients. Small food pantry clients received the largest proportion of healthy foods. In addition, medium and large food pantries often distribute more foods and serve more clients than small ones. These findings suggest clients who visit larger food pantries may have the greatest potential to benefit from nutrition programs and interventions in pantry settings.
Clients of client-choice pantries obtained foods of significantly lower nutritional quality than clients of traditional pantries for medium-sized pantries. This suggests that, although clients are able to make their own food selections in the client choice model, the foods available in the pantry food environment and clients’ nutritional knowledge and motivation remain important determinants of the healthfulness of the client bag [13
]. From our formative research [9
], the pantry managers using the traditional distribution method we spoke with did not purposefully try to push healthier foods to clients in pre-packed bags. The reasons for preferring pre-packed bags were more related to maintaining safety, order, respect in the pantry, and for serving clients as quickly as possible. Additionally, not all client-choice pantries offered true client choice by allowing clients to select whatever they wanted, but some of them had restrictions as to the types of foods that could be selected. It’s important to note that not all client-choice pantries offered the same level and flexibility of choice.
In terms of the proportional weight of select food groups received by food pantry clients, clients across food pantries of different sizes and distribution methods received a significant portion of both healthy and unhealthy foods. In other words, clients received a variety of both fresh fruit and vegetables, beverages, desserts, and snacks, which suggests food pantries are offering a diverse range of foods. Nonetheless, while healthy food options were indeed received, this study supports past research suggesting there is still room to significantly enhance the proportion of healthy foods for food pantry clients [10
]. In our intervention studies, we were aiming to promote lean and low-sodium proteins, fruits and vegetables, and healthy carbohydrates.
This study had several limitations. Chief among these was the generalizability of the results due to sampling issues. The sample size was small, participation was low among pantries invited to participate, and the sample selection of pantries and of clients was not random. Nevertheless, our analysis provided data to consider further refinements of pantry interventions and planning of more rigorous research on factors influencing the effectiveness of pantry interventions. This also limited us from doing some regression analyses to examine if certain client characteristics were predictors of the nutrition quality of foods received. For example, if the pantry used a client-choice model, their clients’ disease history or health status may have influenced their food selection. The power of the factors studied in our study explaining the differences in healthier foods obtained by pantry clients were limited by this study’s low effect modification. These results as preliminary descriptive information require confirmation in a larger study. In particular, the client choice method does not appear to be a panacea for ensuring higher nutrition quality foods are selected, further intervention is needed to promote healthful selection. Second, we used FAST to determine the healthiness of clients’ food, instead of the Healthy Eating Index (HEI) [18
], to better target food pantry settings. Although FAST is correlated with HEI-2010 scores [12
], the HEI is commonly used to report findings on other food sources (supermarket and corner store purchases, school meals, etc.) and diets, thus limiting our ability to compare results with studies considering other sources of food for low-income, food-insecure individuals [12
]. Third, clients taking more nutritious food home from a pantry does not necessarily imply a healthier diet. For many clients, pantries supply only a portion of their overall food supply. However, for clients who usually visit pantries, those food are a very important part of diets for their daily lives. In addition, the eligibility criteria were more suitable to selecting pantries for our intervention study, which may reduce the generalizability of this sample.