Food for Thought: A Qualitative Evaluation of Caregiver Preferences for Food Insecurity Screening and Resource Referral

Abstract

Background/Objective:

Although there is growing interest in screening for Food Insecurity (FI) in the clinical setting, little research exists describing the effect of screening practices on caregiver comfort and willingness to disclose social risk, or what factors affect eventual engagement with food resources.

Methods:

In this qualitative study, we conducted 40 semi-structured interviews with caregivers of pediatric patients who reported FI in the Emergency Department of an urban, freestanding children’s hospital. All interviews were digitally recorded and transcribed. We used content analysis with constant comparison to code interviews inductively and identified emerging themes through an iterative process.

Results:

Three primary themes emerged as factors affecting caregiver comfort in FI screening and engagement with resources: perceived negative repercussions associated with screening, perception of need, and effective connection with food resources. While most caregivers reported comfort with FI screening, intentional steps to provide anonymity and reduce stigma in the screening process were important in facilitating disclosure of social need. Engagement with resources was influenced by caregiver prioritization of food access within a milieu of life stressors, and the perception of personal need relative to their peers. Furthermore, caregivers suggested practices to facilitate effective connection with food resources such as use of mobile health technology, geographically based resources, and personalized connections.

Conclusions:

This study demonstrates the importance of screening methods that reduce judgement and preserves caregiver privacy, referring families to resources that are geographically accessible, and developing creative strategies that enhance families’ connection to resources, such as with use of mobile technology.

What’s New:

This qualitative study explores the effects of screening modality and location on caregivers’ comfort and willingness to disclose risk of FI, and uncovers factors that may affect eventual connection to food resources among those who are food insecure.

Food insecurity (FI), the uncertain availability of nutritionally adequate or safe food, is a pervasive and serious public health issue in the United States. FI disproportionately affects children, impacting 15.7 percent of households with children under the age of 18, and 16.4 percent of households with children under the age of 6, according to federal reports.¹ While related to poverty, FI is an independent predictor of poor health and development outcomes for children; children from food insecure households are more likely to have poor health overall with higher rates of chronic illnesses, hospitalizations, as well as behavioral, academic and psychosocial problems. ^{2, 3}

For these reasons, healthcare providers have become increasingly interested in screening for FI. Both the Academic Pediatric Association and American Academy of Pediatrics encourage pediatricians to incorporate regular screening into their practice.^{2, 4} There is little research that explores the best approach to screen for FI in a manner that maximizes caregiver comfort during screening which could affect disclosure rates and validity of responses. Previous studies of social determinants of health screening processes revealed significantly higher disclosure rates with written paper or electronic screening formats when compared to face-to-face screening, however the reason for this has not yet been explored. ^5,6 Furthermore, rates of resource engagement after screening and referral remain low, raising questions about screening without the ability to effectively intervene.^7–9

This qualitative study explores the effects of screening modality and location on caregivers’ comfort and willingness to disclose risk of FI, and uncovers factors that may affect eventual connection to food resources among those who are food insecure.

Methods

Study Setting and Participants

This qualitative, explanatory sequential study was nested within a randomized trial on FI screening modality.⁶ The initial randomized trial consisted of a sample of 1818 adult, English-speaking caregivers (i.e. parents, guardians, adults caring for the patient) presenting with pediatric patients <18 years of age, in non-critical condition, in the Emergency Department (ED) of a large, urban, freestanding children’s hospital. This was selected as the study location as the ED of academic medical centers often serves as the point of care entry for impoverished and high-risk families.^10–12 In 2017— the year of participant recruitment—this ED saw 66,952 unique patients with over a total of 99,369 encounters; 55% were African American, and 9% Latino; 3% were non-English speaking. Caregivers were approached for study participation during waiting periods in the ED and randomized to either an electronic tablet or verbal screen for FI using a validated, two-question tool (Hunger Vital Sign™) with yes/no responses.¹³ These two questions are: within the past 12 months “we worried whether our food would run out before we got money to buy more” and “the food we bought just didn’t last and we didn’t have money to get more.” All study procedures were conducted by research assistants with training in medical interviewing techniques. Each respondent was provided with a list of food resources. Those who reported FI were additionally offered direct telephone-contact by a food resource agency that assists with enrollment in federal programs and provides navigation to emergency food assistance after the ED visit, as well as participation in a follow-up phone interview. Of the 375 caregivers who reported FI—an affirmative response to either screening question— 324 agreed to participate in a follow-up phone interview. From this group of caregivers, we used purposive sampling to select 40 participants for interviews to ensure representation across each screening modality, reported preferences of screening modality, and acceptance or refusal of direct connection to a resource agency. In the event of non-response, we selected additional participants using the same criteria. We categorized each of the above separately, aiming to have a sample population that was proportionally representative of our original sample to optimize inclusion of a variety of opinions. Two members of the research team with experience in qualitative interview techniques conducted the interviews by phone after re-confirming consent. Caregivers received a $50 gift card for participating in the interview. All study procedures were deemed exempt from review by the hospital’s Committee for the Protection of Human Subjects.

Data Collection

A semi-structured interview guide was developed from review of the literature and expert opinion.^14–16 The interview guide was reviewed and modified through an iterative process after the first 3 interviews were conducted. Our intention was to explore caregiver comfort with FI screening overall, how comfort level may be impacted by screening modality and location, reasons for accepting or refusing resource connection, and experience with resources offered. Interview prompts further explored with caregivers how to better connect families with food or community resources. We conducted each interview approximately 30 days after the ED visit to allow for illness recovery and potential resource engagement. Interviews were digitally recorded, de-identified, and transcribed verbatim by a third-party transcription service.

Analysis

Transcribed interviews were entered into QSR NVivo 12 Software for coding and data analysis. We used content analysis with constant comparison to code interviews inductively and identify emerging themes. Our initial codes were established using the interview guide. We applied modified grounded theory to support identification of factors that contribute improved screening processes and resource connection. Two study team members coded the first two transcribed interviews and met to examine inter-rater reliability, resolve coding disagreements by consensus, and revise code definitions. These team members received training from a researcher with experience in qualitative research methods who met regularly with the team throughout the coding process. These study team members then used the revised codes for the next 2 interviews, and again met to examine and resolve coding disagreements and finalize code definitions. The same two team members coded the remaining 36 interviews separately, with overlap of 7 additional interviews at set intervals to ensure inter-rater reliability over time. The study team reviewed coded transcript sections to uncover emerging sub-themes, and chose direct quotations to exemplify these themes.

Results

In order to obtain a sample size of 40, we called 90 caregivers between September 2017 and January 2018; 28 did not answer the phone, 8 phone numbers were disconnected or invalid, 14 caregivers hung up during the interviewer’s introduction. The 40 caregivers interviewed were proportionally representative of all patient groups from the original 1818 caregivers screened based on their initial survey responses (Table 1). Inter-rater reliability analysis of coded transcribed interviews produced an average kappa statistic of 0.84. We identified three main categories relevant to FI screening and referral: factors that impact caregiver comfort with screening; factors that affect resource acceptance; and caregiver suggestions to improve connection with food resources. Within these categories we identified several themes that are supported by representative quotations.

Table 1:

Characteristics of Study Participants Based on Initial Survey Responses

	Total (N=1818)	Food Insecure (N=375)	Interview Sample (N=40)
Patient Age, Mean (SD)	10.04 (6.9)	9.15 (6.9)	7.21 (5.6)
Race, N(%)
American Indian/Alaska	16 (0.9)	5 (1.3)	0 (0.0)
Asian	73 (4.0)	7 (1.9)	0 (0.0)
Hawaiian/Pacific Islander	4 (0.2)	1 (0.3)	0 (0.0)
Black/African American	764 (42.0)	236 (69.2)	28 (70.0)
White	783 (43.1)	74 (19.7)	8 (20.0)
More than one	64 (3.5)	17 (4.5)	2 (5.0)
Not listed	111 (6.1)	35 (9.3)	2 (5.0)
Ethnicity, N(%)
Hispanic or Latino	150 (8.3)	51 (13.6)	2 (5.0)
Not Hispanic or Latino	1583 (87.1)	298 (79.7)	36 (90.0)
Unknown/Not Reported	81 (4.5)	25 (6.7)	2 (5.0)
Level of Acuity, N(%)
1 (sickest)	4 (0.2)	1 (0.3)	1 (0.0)
2	642 (35.3)	141 (37.6)	16 (40.0)
3	925 (50.9)	173 (46.1)	17 (42.5)
4	236 (13.0)	55 (14.7)	6 (15.0)
5 (least sick)	11 (0.6)	5 (1.3)	5 (2.5)
Direct Resource Connection
Accepted	N/A	206 (54.9)	23 (58)
Screening Modality Preference	Total with a Preference	Total with a Preference	Total with a Preference
Verbal	(N=614, 33.8%)	(N=162, 43%)	(N=16, 40%)
Tablet	104 (16.9)	25 (15.4)	2 (12.5)
	511 (83.2)	265 (84.6)	14 (87.5)

Factors that Impacted Caregiver Comfort with Screening

Perception screening helpfulness outweighed initial discomfort

Caregivers reported different levels of initial comfort with screening, and described feeling anywhere from “comfortable,” “totally fine,” and “happy” to “embarrassed,” “nervous,” and “awkward.” However, regardless of their initial comfort level caregivers frequently described the screening as necessary and helpful to both identify the problem and enhance access to additional resources. One caregiver stated: “It was embarrassing, but it seemed helpful.”

Fear of ramifications to disclosing FI

Caregivers that had a negative perception of screening feared negative consequences if they were to disclose FI. Some caregivers feared judgement by their provider: “You know that it’s not going to affect the level of care, but just to have people look at you a certain way.” Some caregivers feared being reported to child protective agencies: “You always have to worry in your mind, is someone going to remove your kids because you’re not providing enough.” Another caregiver described that spectrum of fears, “Everybody is always fearful that you’re going to report them to DHS [local child protection agency]. Or that there is some judgment there or that they need a handout.”

Tablet screening as preferred method of screening

Some caregivers noted that a tablet was a more private way to answer questions about sensitive subjects: “I’m happy it was in a device and I didn’t have to speak out loud in front of my child.” Others reported that it avoided the potential for judgment: “I’m not going to be judged by the iPad where I feel I might feel so conscious like this person is judging me.” Others reported tablet screening was easier and faster than speaking with a provider.

Relationship with primary care provider (PCP) impacts comfort with screening

We asked caregivers to consider their level of comfort with FI screening by their PCP compared to the ED setting. Several caregivers stated that their close relationship with their provider would improve their level of comfort, and some noted that it was important for their PCP to know if they were food insecure: “You shared the most personal part with you at the doctor’s office. I don’t think there’s anything bad. So, let them know financially how things are going.” Conversely, others suggested that they may feel uncomfortable disclosing FI to their PCP because they would have to see them again in the future: “So, you get a relationship with that doctor. You really don’t want them in your business or judging you or looking at you some kind of ways.”

Factors that Affected Resource Acceptance

Perception of need

While all caregivers interviewed for this study reported FI, caregivers often refused resources because they did not identify their situation as “bad enough” to require additional help. Many reported discomfort taking resources that other families needed more. One caregiver simply stated, “I don’t think that my family is as needy as other families could be.” Alternatively, some caregivers reported that in specific times of increased need they would be willing to pursue additional resources. One caregiver summarized, “I know that every now and then things get rough and we may run out of food, so them calling, giving some resources to help us out in the long run.”

Previous experiences with social assistance

Some caregivers reported negative past experiences with social resources, and expectations that their income or employment status would disqualify them from receiving additional resources. One caregiver described his rational for refusing resources: “Once they see your income, they’re going to say, ‘Oh you can afford this,’ not knowing that they don’t take a consideration that you have to pay for your insurance. You have to pay for rent. You have to pay for all these things because the system won’t help you. So, it’s just like I didn’t need to be contacted because they already shut me down.” Caregivers also reported that public assistance was not always sufficient to feed their families. One mother told her personal story about needing additional resources during a particularly difficult time in her family’s life, “Our food stamps got cut tremendously. And my husband wasn’t offered overtime at that time to compensate what he had lost in food stamps. We have three children and at the time barely any food in the house.”

Engagement with resources and competing priorities

A common theme among caregivers with FI who did not engage with resources was the presence of competing priorities: the inability to find time to make or return phone calls, visit websites, or travel to resource agencies because of other distracting or higher priority demands. One caregiver summarized, “I’m going to be honest with you with everything’s been going on the last couple of weeks since we were in there. I have not gotten the chance to take a look at it.” Other caregivers specifically listed their child’s poor health, or the actual ED visit, as a factor negatively impacting their ability to connect with resources. One caregiver described not utilizing resources offered during their ED visit because they were distracted by their son’s acute medical problem, “It was the scariest day in my life…If I do have it[resource], it’s probably in his discharge papers, but I had not seen it.” Another caregiver described their daughter’s illness as a barrier, “I did call, but then I got a little sidetracked because my daughter’s health situation just continued from there and I didn’t follow up.”

Caregiver Suggestions for Resource Connection

Electronic mode of resource connection

Several caregivers suggested that keeping track of a physical list of resources was difficult for them. One caregiver stated, “The paper, people lose it. There’s so many bunch of stuff in the trash, but emails will always be there.” Many felt that both receiving and accessing food resources might be easier if they could do so over their phone or computer. One caregiver suggested the use of a health portal normally used to message their doctor or receive test results to access food resources. Another felt screening and connection could happen via kiosk. Several caregivers mentioned that electronic options offered privacy, which was appealing.

Instructions on accessing resources should be more widely available

Many caregivers suggested that a generalized distribution of food access resources without associated screening processes would help connect families to food resources. Several different modes of communication were suggested such as television, phone calls, text messages, mailed letters, or pamphlets. Others suggested advertising resources in specific locations such as doctor’s offices, EDs, churches, and gyms. One caregiver suggested targeting at-risk communities. As one caregiver stated, “Get out there more and just pass out the paper, pass out the flyer to people.” Others also suggested more frequent reminders on how to access food resources would be helpful.

Resources should be geographically appropriate

Caregivers reported frustration that resources were not geographically convenient and emphasized individualizing resources to a family’s location. One caregiver described that her specific search for resources, “I was just basically trying to see what something is convenient in my area, like close to home. Because I don’t normally live in the area… so of course, I’m trying to find out what would be a place convenient for me to go to after work or stop in the evening or something.” Some suggested increasing food distribution locations might improve caregiver access to resources, while others listed limited options for transportation as a barrier to accessing resources: “being that I live basically on the other side of the city, and I don’t have a vehicle getting there to take advantage of that which I would love to do is very difficult.”

Discussion

With increasing recognition of the importance of addressing FI during medical visits, there is a growing need for evidence regarding how to most effectively and efficiently identify social risk and connect families with needed resources.^{4, 17} Maximizing caregiver comfort with screening mechanisms may enhance the accuracy of information collected, and identifying barriers to utilization of social resources after referral can inform the development of social resource referral processes. This study identified three primary themes as factors affecting caregiver comfort in FI screening and engagement with resources: perception of “being judged” and possible repercussions associated with screening, a family’s perception and prioritization of their social need, and effective connection with food resources.

As FI is a potentially stigmatizing condition, it may elicit concerns regarding social desirability in responses depending on format of screening. Consonant with our study findings, previous studies have demonstrated caregiver discomfort in discussing social risk in front of their children, concerns regarding involvement of Child Protective Services upon disclosure of FI, and discomfort with disclosure has been shown to be most prevalent among those with highest social risk.¹⁴ Interestingly, our study showed that a close relationship with the medical provider could be both a facilitator and a barrier to relaying social need depending on the perceived potential for “judgement.” Written screening, such as on a tablet, may help ameliorate these concerns by providing a level of anonymity in screening and reducing concerns regarding stigma while allowing for follow up discussion with a trusted provider. Fear of disclosure due to concern for punitive retaliation suggests that screening should be accompanied by language that describes its purpose and promotes a “safe space” for this dialogue.

Rates of resource engagement after screening for social determinants of health remain low, raising concerns about the ethics of screening without appropriate linkage. ^{7, 18} Consistent with prior literature, our study demonstrated that existing social programs may not fully meet the needs of food insecure families.¹⁹ Our results demonstrate that there are multiple apparent barriers to resource acceptance including previous experiences with social assistance, perception of need, and competing priorities. Caregivers discussed times that they were turned away from resources, or had needed benefits cut, and that these negative past experiences affected acceptance of further referrals. We also found that underutilization of resources was powerfully impacted by social beneficence, with multiple caregivers describing service refusal because “someone else could use it more.” It may be important to educate families that their participation in social programming will not preclude another family from doing the same. Furthermore, we found that connection with resources was impacted by prioritization of FI within a milieu of other life stressors, including the stressors associated with medical acuity in this setting. Although participants reported that the ED was an acceptable location for screening, this could limit effective resource connection after referral for some families.

Caregivers in our study suggested practices to facilitate effective connection with food resources that involve mobile health technology, “reminders” of resource availability distributed outside of the clinical setting, and geographically based resources. By better understanding reasons for poor engagement, healthcare providers and communities can develop the most effective programs and resource connection processes to address FI.

Strengths and Limitations

Our explanatory sequential study design and purposive sampling to ensure representation across a variety of opinions in terms of screening modality, location, and resource connection, provides robust data regarding preferences in FI screening and intervention. Limitations include selection bias, as those who self-selected out of interview may have differed from our study population, and we were unable to capture those who did not feel comfortable disclosing food insecurity. Furthermore, due to the low overall prevalence in our ED population, we excluded non-English speaking families from this study, thereby limiting generalizability to a more language diverse patient population.

Additionally, we acknowledge that FI is only one of many social determinants of health. Given that FI has a validated, highly sensitive and specific screening tool, and well-defined local and federal resources (i.e. SNAP, WIC, and food pantries), it is our hope that studying FI provides lessons for screening and resource connection across other SDH, but further research is needed.

Conclusions

This qualitative evaluation of FI screening in the clinical environment demonstrated the importance of screening in a manner that reduces judgement and preserves caregiver privacy, referring families to resources that are geographically acessable, and developing creative strategies that enhance families’ connection to these resources, such as with use of mobile technology (Table 2). Moreover, this study highlights the importance of involvement of families in the development and implementation of social resource referral processes.

Table 2:

Summary of recommendations for FI screening and referral protocols

Stage in Protocol	Concern addressed	Recommendation
Screening	Perceived “judgement” or potential for negative repercussion	Introductory language that describes purpose of screen and next steps
		Modality that preserves privacy (e.g.: tablet-based screening)
Resource Referral	Perception of need as compared to others	Education that use of resources will not preclude another family from doing the same
	Previously denied resources	Inclusion of eligibility criteria for resources at time of referral
	Resource accessibility	Tailor resources to family’s preferred geographic area
		Electronic-based resources (e.g. text message, email, health portal)