Testing the feasibility of a system-based approach to deliver a smoking cessation and food nudging intervention at food pantry sites

This study suggests that it is possible to train food pantry staff and volunteers to deliver health promotion programs, and that food pantry users can improve their cardiovascular disease risk because of it.

Abstract

Food-insecure adults disproportionately experience high cardiovascular risk. Guided by the Feeding America recommendations, we tested the feasibly of a system-based approach to address the cardiovascular risk behaviors of current smoking and dietary choice at food distribution sites. Food pantries affiliated with the Food Bank of Delaware organization (N = 14) were invited to take part. Pantries who agreed solicited personnel within the pantry (i.e., staff, active volunteers) to become trained as quit-smoking coaches and/or food environment “nudging” interventionists. After training, trained personnel implemented the evidence-based treatments. Across a 6-month observation period, quit coaches reported each month on the number of (a) enrolled food pantry clients, (b) total quit-smoking sessions scheduled and attended, and (c) the smoking status of clients who attended the final, third session. Trained evaluators visited participating pantries once per month across the observation period to assess adherence to nudging guidelines. One in five (21%; 3/14) invited pantries participated in the study, and five personnel were trained to deliver intervention components. Across the observation period, quit coaches reported that 86 new smoking cessation clients were enrolled, 228 quit coaching sessions were scheduled, and 187 attended (82% attendance rate). Smoking cessation rates were estimated at 19%–36%. A 100% adherence to the food nudging approach was observed. These data demonstrate the feasibility of a system-based approach to build the capacity of food distribution sites and personnel to deliver smoking cessation quit-coaching and food nudging interventions on-site.

Implications

Practice: Food distribution sites are a viable venue to address the cardiovascular risk factors of tobacco use and food selection.

Policy: Community health funders must prioritize building the capacity of food distribution sites to deliver health promoting interventions. Food distribution personnel can be trained to be effective smoking cessation and food nudging interventionists.

Research: Future research is needed to define effective strategies to engage community health funders and providers in integrated care approaches to promote cardiovascular health in high-risk food assistance populations.

Introduction

Cardiovascular diseases (CVDs) remain the leading cause of death and disability in the United States [1]. Observational data have shown that food insecurity (i.e., economic and social condition of limited or uncertain access to adequate food) [2] may be a CVD risk factor [3]. When compared with food secure adults, food-insecure adults have a 21% greater odds of hypertension [4], 32% increased odds of being obese [5], and a more than twofold greater odds of diabetes [4]. Limited CVD prevention efforts targeting food-insecure adults have been developed or tested; thus, food-insecure adults remain an understudied and overlooked population at high risk for CVDs.

Cardiovascular risk factors are also more common in food-insecure populations. For example, as a leading risk factor for CVDs, current cigarette smoking rates in food-insecure adults have been reported to be 50% [6] when compared with national rates of 14% [7]. Similarly, food-insecure adults report consuming a diet of low quality, often limited in nutrient-dense foods such as vegetables, fruit, and whole grains [8], and high in nutrient poor foods such as sweets [9] when compared with food secure adults. Food-insecure adults are also less likely to be physically active [10] and to get adequate sleep (7–8 hr per night) [11]. The common occurrence of key CVD risk factors in this population underscores the need for systemic and sustainable prevention approaches.

Another key contributing factor to the high prevalence of CVDs in food-insecure populations can be the lack of access to both primary and preventive health care services to support the management of CVDs and health behavior change, respectively. Food pantries are emerging as a centralized location through which adults at high risk for food insecurity can be systematically accessed and triaged for health care services [12]. For example, the Feeding America initiative, a nonprofit hunger relief organization whose mission is to feed America’s hungry through a nationwide network of member food banks, actively promotes “upstream” health promotion approaches that address health at food distribution sites [13]. Health-focused activities recommended at food distribution sites include the provision of healthy food, offering nutrition and health education, hosting health screenings, and facilitating the completion of health insurance and other benefits application forms [13].

The Feeding America recommendations to address health at food distribution sites are consistent with several principles of an integrated, system-based, population health approach, namely, the bundling of strategies and interventions to address multiple related risk factors, the targeting of a high-risk population subgroups using existing infrastructure, and the utilization of partnerships between multiple sectors [14]. Integrated care approaches have been more readily adopted in clinical (e.g., patient-centered medical home models) versus community settings [15, 16].

To address this gap, we sought to develop and test an integrated approach to address the key cardiovascular risk factors of current cigarette smoking and dietary intake in three food distribution sites (“pantries”) in New Castle County, DE. Consistent with the Feeding America recommendation of addressing health at food distribution sites, we sought to build the capacity of food pantry personnel to implement evidence-based approaches for smoking cessation, and healthful food selection, at their food pantry. To address the risk factor of smoking cessation, a state-funded face-to-face quit-coaching program was implemented. The program offers an alternative to the traditional telephone quit-line that has been shown to increase the odds of cessation by 40%–60% [17, 18], although the uptake is low (<1% of smokers) [19]. To address the risk factor of poor food choice, we utilized “nudging” strategies or subtle environmental changes in food distribution settings designed to make healthy choices the easy choice for food pantry clients [20]. Our goal in writing this report is to provide a proof-of-concept model for the implementation of these system-based, intervention approaches that may be adapted and refined for other contexts.

METHODS

Program goals and theoretical frameworks

The goal of this program was to use a system-based, capacity building approach to (a) train food pantry personnel to deliver smoking cessation and utilize food environment (“nudging”) strategies at food pantry sites and (b) measure smoking cessation program uptake and nudging strategies adherence in the pilot pantries. Theoretical frameworks used to guide our program development process and evaluation metrics included community-based participatory research [21] and the PROmoting School-community-university Partnerships to Enhance Resilience (PROSPER) Community-Partnership Model [22].

Consistent with community-based participatory research principles, a community-academic partnership (CAP) was formed between the University of Delaware (UD) academic institution and the Food Bank of Delaware (FBD) community organization in spring 2016. The FBD was recognized as a unit of identity and was fully engaged with UD in a collaborative, equitable partnership that involved power-sharing and colearning while being focused on a public health problem that had local relevance [21]. The initial focus of this CAP was to conduct a needs assessment of food pantry clients across six food pantry sites in the state of Delaware. Results of this needs assessment (conducted between October 2016 and March 2017) showed that in a sample of 144 food pantry clients surveyed [6], 73% were hypertensive and 50% were current smokers, and of these, 48% (n = 35) were highly nicotine dependent (as indicated by having the first cigarette within 5 min of waking) [23], and 77% were food insecure [6]. That approximately one in four respondents in this sample were food secure is consistent with Feeding America data showing that 24% of pantry client households are food secure and that 43% of pantry clients living in food secure households visit food pantries on a recurrent basis [24]. Given that poor dietary choice and intake are key predictors of hypertension [25], and that the Feeding America guidelines recommend addressing dietary choice and intake at food distribution sites [13, 20], dietary choice was chosen as an important and viable intervention target.

Results from the needs assessment also showed that the odds of current smoking was five times higher in food-insecure adults versus food-secure adults (odds ratio = 4.98; p = .006), even after adjustment for demographic factors [6]. Moreover, although a large proportion (60%) of the current smokers reported wanting to quit smoking at baseline, at 8-week follow-up assessment, only 7% called the Delaware state quit-line following a 7- to 10-min educational intervention designed to encourage uptake of the free Delaware state quit-line services. Consistent with published accounts [26], anecdotal evidence from food pantry clients who smoked suggested they did not want to use their phone minutes to call the Delaware state quit-line, and were uncertain if someone from a state quit-line would be able to help them.

These data led our CAP to recognize that system-based and capacity building approaches were needed to address the smoking cessation and food choice of food pantry clients. The PROSPER Community-Partnership Model that was originally developed to build capacity and competence in under-resourced school systems [22] posits that to successfully build a framework for intervention development and implementation of the entities of external capacity agents, internal resource agents, and linking agents are needed [22]. External resource agents are characterized by an established infrastructure (e.g., state and federal agencies) that can be used to provide critical training and technical assistance to enhance the capacity for sustained implementation of evidence-based programs. Internal resource agents are typically the primary site for the programming and contribute in terms of personnel buy-in and integration with other agencies. The linking agents connect the internal and external agents, providing interface so that programming design and implementation can be culturally sensitive and appropriately targeted.

Using this framework, three basic source organizations represent the partners involved in the current program: (a) internal agents: food pantry personnel (e.g., pantry directors, volunteers, paid employees); (b) linking agents: academic personnel (e.g., CVD, smoking cessation and nutrition content experts, community partnership expert); and (c) external agents: state providers of prevention and public health services (see Fig. 1). Under this framework, our CAP evolved to a Community-Academic-Public Partnership to include key representatives from the public sectors of the Delaware Division of Public Health, and Optum, the organization that is subcontracted to manage the Delaware state quit-line.

Fig 1.

| PROSPER linking capacity agents framework for evidence-based interventions in Delaware food pantries.

A system-based approach to smoking cessation and improved food choice

Smoking cessation program overview

The Delaware state quit-line offers a three-session, in-person, quit-coaching program that includes an 8-week course of pharmacotherapy, all at no cost to Delaware residents aged 18 or older [27]. Optum, the company who contracts with the state of Delaware to provide quit-line services, offers a comprehensive 2-day training program, so that health care professionals with a clinical license can become certified face-to-face quit coaches. Once certified, these quit coaches can function as independent contractors who can provide three quit-coaching sessions and vouchers for pharmacotherapy (i.e., nicotine patch) to treatment-seeking smokers who are enrolled in the Delaware state quit-line program. Quit coaches are paid by the quit-line $50 for the initial visit and $30 each for the second and third sessions of smoking cessation quit coaching.

Food choice program overview

To improve food choice of pantry clients, “nudging” strategies, defined by Feeding America as a subtle environment change in food distribution settings designed to make a healthy choice the easy choice, are being encouraged [20]. Research conducted as a collaboration between Feeding America and the Cornell Center for Behavioral Economics in Child Nutrition Programs has demonstrated the effectiveness of “nudging” interventions in food pantries [20]. Numerous strategies are implemented as part of “nudging” interventions including use of signage (e.g., shelf tags), priming (e.g., display of large, appealing photograph), and placement of products (e.g., fruits and vegetables at eye level) to improve food choice.

Program procedures

A convenience sample of food pantries not currently implementing “nudging” intervention strategies was invited to participate in the program by a Food Bank of Delaware representative. Several of the food pantries were affiliated with established social service agencies and community organizations such as an apartment complex catering to low-income adults, a health organization that coordinates care for adults with HIV/AIDS, and churches. Pantries interested in participating scheduled an introductory meeting with program personnel from the University of Delaware to hear more about the smoking cessation and food choice intervention components.

Smoking cessation program

Following this meeting, a representative from each food pantry provided a list of nonsmoking individuals who were interested in becoming trained as quit-smoking coaches. Quit coaches were required to have a clinical license and liability insurance. For the present study, the requirement to have a clinical license was expanded to include a previous smoking cessation training certification (e.g., American Lung Association’s Freedom from Smoking Facilitator Training), and liability insurance costs were covered by the program. Interested and qualified food pantry representatives attended the 2-day quit-coaching training. Following the training, all quit coaches provided the Delaware state quit-line’s three-session, in-person, quit-coaching program that also provided the choice of free pharmacotherapy (i.e., transdermal nicotine patch) at their respective pantries.

Pantry quit coaches advertised their services through posters and word of mouth. Food pantry participants who currently smoked (i.e., had smoked at least one cigarette in the last month) and felt ready to quit in the next 30 days were eligible to sign up for the three face-to-face smoking cessation sessions. Session 1 lasted 45 min, whereas Sessions 2 and 3 were each 30 min long. The sessions occurred 2–3 weeks apart (i.e., program lasted approximately 8 weeks). During the first quit-coaching session, clients completed a Delaware quit-line enrollment form and were provided with a Delaware quit-line workbook in which they could practice mini-quits, track their smoking habits, list distractions and substitutions, and read about all seven FDA-approved quit medications. The quit coach reviewed in detail the five key steps to quitting successfully (i.e., quit at your own pace, conquer urges to smoke, use quit medicines so they really work, control your environment, and get social support). Last, the quit coach helped the client choose a quit date and provided a voucher for free stop-smoking medication. During the second and third sessions, the quit coach asked about the participant’s quitting experience, positively reinforced successes, and problem-solved challenges. The five key steps to quitting, and medication use was reviewed. If needed, another voucher for cessation medication was provided, if needed. The content of the quit-coaching program was prescribed by the external agent (state quit-line and health department), but the methodologies for client recruitment and program marketing were the choice of the internal agents who were administering the programs (i.e., food pantry).

Nudging intervention strategies

At least one representative at each pantry included in this pilot program was identified as the point of contact for communication and implementation of nudging strategies. Prior to implementation, pantry representatives engaged in a brief training with research personnel (with nutrition expertise). The training session lasted approximately 30 min with the purpose to define nudges and discuss how each nudging strategy would be implemented given the context of the individual food pantry. Foods available in the pantry were documented by research personnel so signage and a photograph for priming could be prepared. Together research personnel and pantry staff also walked through the pantry and discussed ways to implement placement. Research personnel also conducted a baseline assessment of the food pantry nutrition environment to assess pantry style (e.g., choice pantry where clients select foods) and availability of fruits and vegetables. Photographs were taken for intervention material development, specifically signage and priming materials.

Within 1 week of the baseline assessment, pantries were provided with a variety of laminated signage (shelf tags) and a large photograph of a fruit, vegetable, or whole grain food item. The food item that was selected for the photograph was chosen based on availability at the pantry and discussions with the food pantry staff. The nudging strategies were proposed by the linking agents (i.e., academic partner) and then reviewed and modified based on feedback and needs of the internal agents (i.e., individual food pantries).

Implementation of nudging strategies was the responsibility of the food pantry personnel. Approximately, every 4 weeks, research personnel provided the pantry with a new large photograph of a fruit, vegetable, or whole grain, and pantries had the opportunity to ask for additional signage (shelf tags). An unannounced assessment was completed each month to assess placement of products (through photographs), availability of fruits and vegetables, use of signage (shelf tags), and use of priming (display of large, photograph) strategies.

Program constructs and evaluation

The purpose of the program evaluation was to measure program uptake and implementation at the food pantry level across a 6-month observation period. The constructs used in this evaluation are described below.

Internal capacity factors

Willingness of food pantries to participate in program 

A list of food pantries was generated for New Castle County by the Food Bank of Delaware and a convenience sample of 14 pantries invited to participate. The percentage of pantries who agreed to participate in the program was calculated.

Attendance to program-planning meeting 

Representatives from each food pantry were invited to attend a program-planning meeting. The percentage of pantries represented at the meeting, and the percentage of those invited, who attended, was generated.

Linking and external agent factors

Number and training of smoking cessation coaches 

The number of food pantry personnel from each participating food pantry who volunteered to be trained as a smoking cessation quit coach was tallied (see Figure 2). The proportion of these volunteers who completed the Delaware state quit-line training to become a quit coach was generated.

Fig 2.

| Overview of program process and constructs.

Uptake of smoking cessation intervention 

Smoking cessation quit coaches from each pantry reported each month on the number of new quit-smoking clients enrolled, and the number of quit-coaching sessions scheduled and attended between July and December 2018.

Smoking status at third session 

Quit coaches compiled the self-reported smoking status of clients attending their third quit-smoking session. A program quit rate was estimated in two ways. The first approach required calculating the percentage of clients who had scheduled a third treatment session and reported quitting. The second, more conservative approach, involved calculating the percentage of enrolled clients who reported quitting [28].

Number and training for food nudging 

The number of food pantry personnel from each participating food pantry who volunteered to engage in a brief training and implement the food nudging strategies was tallied.

Uptake of food nudging intervention strategies 

Each month a research staff member visited the pantry unannounced to complete the assessment. The assessment included photographs of the pantry to document foods available; availability (yes/no) of at least one fresh, frozen, or canned fruit and vegetable; presence (yes/no) of shelf tag (signage) for at least one fruit, one vegetable, and one whole grain; and display (yes/no) of priming photograph. If the priming photograph was displayed the food displayed was recorded and availability (yes/no) of that food at the pantry was recorded. Adherence was determined based on a positive (yes) response. The utilization of food nudging strategies implemented within each pantry was tracked monthly between May 2018 and November 2018.

RESULTS

Internal capacity factors

Fourteen food pantries affiliated with the Food Bank of Delaware were invited to participate in this program. Of the 14, seven (50%) did not respond, six (43%) wanted to participate, and one (7%) was unsure. Of the six who wanted to participate, two later said that they were unable to secure pantry personnel to become trained as either smoking cessation quit coaches or food nudging interventionists. Thus, out of the 14 pantries invited, four verbally agreed to participate and attend a program-planning meeting. Following the program meeting, one pantry dropped out (reason unknown), yielding a 21% uptake rate among invited pantries (3/14). From the three pantries, eight personnel (Pantry 1, n = 4; Pantry 2, n = 3; Pantry 3, n = 1) attended the informational meeting.

Linking and external agent factors

Six food pantry personnel from three food pantries (Pantry 1, n = 3; Pantry 2, n = 2; Pantry 3, n = 1) volunteered to complete the smoking cessation quit coach training, and all six completed the training (6/6; 100%; see Table 1).

Table 1.

Educational and professional credentials of quit coaches and nudging interventionists

	Setting		Clinical license	Highest educational attainment	Number of years with food pantry	Role at food pantry organization
Pantry 1	AIDS Service Organization	Quit Coach 1	LGSW	MSW	2	Medical case manager
		Quit Coach 2	CADC	BS	6	Senior case manager
		Quit Coach 3	LCSW	MSW	3	Clinical director, Mental Health Program
		Nudging Interventionist 1		MA		Director of Client Services
Pantry 2	Department of Corrections	Quit Coach 1/Nudging Interventionist 1	N/A	BS	9	Program director
		Quit Coach 2/Nudging Interventionist 2	N/A	BS	9	Community re-engagement specialist
Pantry 3	Church	Quit Coach 1	N/A	LPN (retired)	2	Blood pressure screening volunteer
		Nudging Interventionist 1

Across the 6-month observation period where enrollment in quit coaching was on-going, 86 clients were enrolled into the program, 228 in-person quit coaching sessions were scheduled, and 187 attended (82% attendance; see Table 2). When these data are examined by session, 103 scheduled and 86 completed Session 1 (84% attendance); 81 scheduled and 70 completed Session 2 (86% attendance); 44 scheduled and 31 completed Session 3 (70% attendance). When examined by food pantry, Pantry 1 enrolled 28 quit-smoking clients, Pantry 2 enrolled 58 quit-smoking clients, and Pantry 3 enrolled 0 quit-smoking clients. The attendance rate (number of sessions attended/number of sessions scheduled) for each pantry was as follows: Pantry 1 = 69/82 = 84%; Pantry 2 = 118/146 = 81%; Pantry 3 = 0/0 = 0%.

Table 2.

Number of new smoking cessation clients, quit-coaching sessions scheduled, and attended each program month

	Number of new quit-coaching clients enrolled	Number of quit-coaching sessions scheduled	Number of quit-coaching sessions attended
Month 1	11	12	12
Month 2	9	31	23
Month 3	20	47	38
Month 4	16	45	36
Month 5	17	51	41
Month 6	13	42	37
Total	86	228	187

Between July and December 2018, of the 86 enrolled quit-coaching clients, 44 had scheduled Session 3, and of these, 16 self-reported to their quit coach that they were abstinent. Thus, at the time, data analysis was conducted for this rolling admission program, program quit rates were estimated at 19%–36% (i.e., 16/86 [19%] of those enrolled in the program and 16/44 [36%] of those who had scheduled Session 3).

Four food pantry personnel from three food pantries (Pantry 1 = 1; Pantry 2 = 2; Pantry 3 = 1) volunteered to complete the brief training on food nudging strategies and serve as the key point of contact (see Table 1). All four completed the training (4/4; 100%). Two of the three pantries (Pantry 2, Pantry 3) used a choice style pantry. All three pantries were assessed each month except for 1 month when only two pantries completed an assessment due to scheduling conflicts with the third pantry. Across the 6-month time frame, pantries implemented the shelf signage nudging strategy for at least one fruit, one vegetable, and one whole grain 100% of the time. The priming photo was also displayed 100% of the time across the 6 months. Canned fruits and vegetables were always (100%) available in all three pantries, whereas only one pantry (Pantry 3) had fresh fruits and vegetables each month. Frozen fruits were available in one pantry during 1 month and frozen vegetables in one pantry on two separate months.

Discussion

CVDs remain highly prevalent in under-resourced populations such as food assistance groups. In the absence of efficacious integrated health care approaches in the community setting, reaching national goals to improve cardiovascular health by 25%, by the year 2025, will not be realized in health disparate groups [29]. Our system-based approach to improve smoking cessation rates and increase healthful food options in food assistance populations showed that 21% (3/14) of invited pantries ultimately ended up participating in our pilot study, and five personnel were trained to deliver intervention components. Across a 6-month period, 86 new smoking cessation clients were enrolled, and 187 quit-coaching sessions were attended (82% attendance rate). Moreover, of the 44 treatment-seeking smokers who had scheduled Session 3 (final quit-coaching session), 16 self-reported abstinence (36% of those scheduled for the third session; 19% of those enrolled in the program). A 100% adherence to the food nudging approach was observed over a 6-month observation period. Together, these data provide a positive-signal for a system-based approach that builds the capacity of food distribution sites to deliver health-promoting interventions.

Some of the most encouraging findings from this study centered around the quit-coaching program components. The enrollment of 86 new smokers into cessation treatment (~14 per month), an 82% quit-coaching session attendance rate, and a self-reported end-of-treatment quit rate of 19%–36% provides support for the program. In a previous study, a community-based navigator system designed to promote quit-line use in low-income smokers (n = 40) yielded a 28% program attendance rate and a 21% quit rate [30]. Smoking prevalence in under-resourced groups remains up to three times higher than the national population [6] largely because of underutilization of evidence-based treatments [31, 32]. Strategies used in the current proof-of-concept study to improve utilization were the provision of treatment by in-house food pantry personnel who were familiar to the clients, and offering quit coaching at the food pantry site where the clients were already routinely visiting. Our data suggest that these approaches may go toward overcoming barriers to treatment utilization in this population such as distrust of external health care providers [33] and lack of transportation [34].

Nudging intervention strategies are designed to be implemented within a choice style pantry. A choice style pantry is one where individuals can select the foods themselves. Two of the three pantries in this study used this food distribution style. The third pantry used a bag style method where clients were provided with a bag of preselected food items. Instead, as per the request of the site personnel, recipes related to food displayed as part of the priming were created and distributed in the food bags.

The availability of foods in a pantry is largely dependent on donations from the Food Bank of Delaware or other sources. There is great variability in what foods may be available week to week and pantries are often limited in the ability to offer fresh fruits and vegetables due to a number of constraints (e.g., equipment, distribution). Education around nutritional quality of canned and/or frozen fruits and vegetables could assist in promoting these items. Creating sustainable community connections to provide fresh fruits and vegetables could also be considered.

One of the challenges to the feasibility and scalability of the current approach is the uptake of the program. Our data showed that 21% (3/14) of invited pantries ultimately participated. Among trained volunteers, 83% (5/6) of the quit coaches successfully functioned in this role and 100% of food nudging interventionists were adherent. Consistent with previous studies that have attempted to build community capacity by using a train-the-trainer approach [35], the pantries that did not participate in our program cited lack of time and/or perceived capacity for in-house personnel to become trained interventionists. What is noteworthy about the current program is that once pantries committed, and volunteers completed the necessary training, the rates of adherence were high. To advance this model, a greater understanding of the resources and capacity needed by pantries to successfully participate in this program is needed.

Some of the challenges to program implementation centered around engaging external resource agents, and ensuring that the internal resource agents (i.e., food pantry personnel), were sufficiently resourced to complete the program. For example, the quit-line face-to-face quit-coach training program required that individuals wanting to become quit coaches have a college degree and a clinical license (i.e., RN, PharmD, LCSW) for entry into the training program. To implement the present study, we received special permission to expand these training and education entry requirements so that food pantry personnel could be trained. Such stringent training and education requirements to become a quit coach could ultimately prevent community workers from becoming trained and impede efforts to promote cessation in hard-to-reach populations. Relatedly, there was concern that using community-based smoking cessation quit coaches was not evidence based, whereas there is evidence that training community members to deliver smoking cessation treatment is effective in promoting cessation in hard-to-reach groups [36]. Trust-building and education about smoking cessation counseling approaches with the external capacity agents were vital to conducting the current program.

In terms of challenges to program implementation experienced by the internal resource agents (i.e., food pantry personnel), the Delaware state quit-line relies heavily on email and internet-based modes of communication with quit coaches. One of our program’s quit coaches did not have an email address and limited information technology (IT) literacy. This quit coach was not able to deliver services during this program’s observation period. Program staff worked with this quit coach to build her IT access and skills to allow her to maintain communication with the state quit-line. Broader dissemination and scaling of this program may have to include IT training modules, or alternative modes of communication that are more sensitive to under-resourced communities.

Our study has some strengths and limitations. Among the strengths of this study is the solidification of a long-standing Community-Academic-Public Partnership working group that is committed to improving the cardiovascular health through increased smoking cessation and improved dietary choice among food assistance populations in Delaware. Moreover, our results support the premise that among willing pantries, personnel can be trained to deliver smoking cessation quit-coaching and/or implement food nudging strategies to good effect (i.e., high program adherence, encouraging quit rates). Limitations of this study include the use of a convenience sample, the lack of clear pre- and post-data that precludes a direct assessment of improvement, or changes in study metrics that may be attributed to the intervention. We did not directly survey food pantry clients, so this precluded us from verifying attendance or smoking cessation rates, or obtaining dietary intake data for each client. The external validity of this approach has yet to be determined. Next steps for this work include education of the broader public health community about the capacity of food pantry personnel to be trained in, and adherent to, program protocols. Completion of a pre- and post-assessment of study variables, including dietary intake and blood pressure would also add to the evidence based for this system-based approach.

Funding: Research reported in this publication was supported by the National Institute On Minority Health And Health Disparities of the National Institutes of Health under Award Numbers R01MD012734 and R21AG056765 (F.P.) and by an Institutional Development Award (IDeA) Center of Biomedical Research Excellence from the National Institute of General Medical Sciences of the National Institutes of Health under grant number P20GM113125 (F.P.).

Acknowledgments: The authors gratefully acknowledge Karyl Rattay and Lisa Moore (Delaware Division of Public Health), Sarah Fitzpatrick (University of Delaware), Mackenzie Perkett (formerly University of Delaware), Tanny Dickerson (Delaware Division of the American Lung Association), and Joanne Wardell (Optum) for their valuable contributions to this work. We also acknowledge Karen Murphy, Danielle Zack, and Yatzka Hernandez for providing administrative support.

Compliance with Ethical Standards

Conflict of Interest: F.P. has received study medication from Pfizer. S.R., C.M., D.T., S.H., S.R., and R.L. have no conflicts of interest to report.

Human Rights: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed Consent: The observational nature of this study precluded the necessity for informed consent being obtained. Individual-level data were not collected as part of this study.

Welfare of Animals: This article does not contain any studies with animals performed by any of the authors.