Addressing food insecurity in HIV care: perspectives from healthcare and social service providers in New York state

Abstract

Ending the HIV epidemic in the United States will require addressing social determinants contributing to poor care engagement among people living with HIV (PLH), such as food insecurity. Food insecurity is associated with poor care engagement among PLH. Yet, few studies have examined the perspectives of healthcare and social services providers on addressing food insecurity in HIV care. Guided by the Social Ecological Model, we conducted semi-structured interviews with 18 providers in New York State to understand barriers and facilitators to addressing food insecurity in HIV care. Thematic analysis illustrated eight themes across various levels of the Social Ecological Model. At the patient-level, providers perceived patients’ feelings of embarrassment, shame, and judgement, and low health literacy as barriers. At the provider-level, challenges included limited time. Facilitators included fostering strong, patient-provider relationships. Barriers at the clinic-level included limited funding, while clinic resources served as facilitators. At the community-level, challenges included intersecting stigmas arising from community norms towards PLH and people who receive food assistance and limited access to healthy food. Findings suggest the need to incorporate their insights into the development of interventions that address food insecurity in HIV care.

Introduction

Ending the HIV epidemic in the United States (U.S.) will require increasing engagement in care for people living with HIV (PLH). Despite evidence illustrating strong associations between retention in care and positive HIV outcomes (Yehia et al., 2014), optimal engagement among PLH remains a challenge (Centers for Disease Control and Prevention, 2022). With approximately 1.1 million PLH in the U.S. in 2020, nearly 80% were linked to care; yet, only 51% were retained in care and 65% achieved viral suppression (Centers for Disease Control and Prevention, 2022). Poor care engagement among PLH is predictive of increased HIV viral loads (Berg et al., 2005; Zinski et al., 2015) and is a primary driver of the HIV epidemic; more than 60% of new HIV infections in the U.S. are transmitted by PLH who are not retained in care (Frieden et al., 2015; Skarbinski et al., 2015). Similar trends are observed in New York State (NYS). Among the estimated 111,000 PLH in NYS in 2021, 82% were linked to care, 64% were retained in care, and 73% were virally suppressed (New York State Department of Health AIDS Institute, 2022).

Strategies for improving HIV care must address social determinants contributing to poor care engagement among PLH. One such determinant is food insecurity (FI), defined as having limited access to nutritionally adequate foods due to a deficiency of resources or the inability to attain foods in socially acceptable way (National Research Council, 2006; U.S. Department of Agriculture, 2021). FI has been shown to disproportionately affect PLH. It is estimated that 25–80% of PLH are food insecure (Anema et al., 2011; Feldman et al., 2015; Weiser et al., 2013), compared to only 11% of the U.S. population (Coleman-Jensen et al., 2021). Among PLH, FI is associated with poor care engagement (Bleasdale et al., 2022; Bleasdale et al., 2023; Colasanti et al., 2017; Lee et al., 2021; Whittle et al., 2016; Young et al., 2014). A retrospective case–control study illustrated that unretained PLH were significantly more likely to report FI and unmet food needs compared to PLH retained in care (Colasanti et al., 2017). Further, FI is associated with negative HIV outcomes, including poor ART adherence (Leddy et al., 2020; Singer et al., 2015), lower CD4 cell counts (Aibibula et al., 2016), and incomplete viral suppression (Aibibula et al., 2017).

While there are well-documented associations between FI, care engagement, and related HIV outcomes, less is known about effective strategies to alleviate FI among PLH (McKay et al., 2017). Recently, there has been an increased focus on “Food is Medicine” programs in healthcare settings (Downer et al., 2020; Mozaffarian et al., 2022). The “Food is Medicine” framework describes various healthcare- and population-based food and nutrition interventions, including medically tailored meals and groceries, and fresh produce prescriptions (Downer et al., 2020; Mozaffarian et al., 2022). Research has illustrated the potential of “Food is Medicine” interventions to improve health outcomes and care utilization (Downer et al., 2020). Yet, there is limited research on applying these programs in HIV care, particularly to address FI among PLH (Aidala et al., 2015; McDermott et al., 2003). Integrating these programs into HIV care will require a greater understanding of how to address FI among PLH from multiple perspectives. Currently, there is limited research exploring the experiences of FI among PLH in high-resource countries (McKay et al., 2017). Most work has explored patients’ experiences (Brunner et al., 2001; Miewald et al., 2010) or has assessed the feasibility and effectiveness of community-based food assistance interventions for PLH (Palar et al., 2017; Yu et al., 2022).

An understudied area of research on addressing FI among PLH is the perspectives of multidisciplinary HIV providers (i.e., healthcare and service providers). While research has examined providers’ perspectives on food and nutrition programming at community-based HIV organizations (Slater et al., 2015) and clinics (Brunner et al., 2001), and the use of food as incentives for care (Gagnon et al., 2021), the extant body of research has yet to capture their perspectives on the facilitators and barriers to addressing FI in HIV care. Understanding HIV providers’ perspectives is essential to integrating food assistance interventions, such as “Food is Medicine” programs, into routine HIV care. To address this gap, the current study qualitatively explored HIV healthcare and social service providers’ perspectives on the importance of food and nutrition in HIV care, and barriers and facilitators to addressing FI in HIV care. Such viewpoints have the potential to offer valuable insights into the development of interventions that seek to address FI as a barrier to successful engagement in HIV care.

Materials and methods

Participants and procedures

Participants were recruited from August-September 2022 using convenience sampling methods. First, we distributed study information to the NYS Ending the Epidemic listserv, a voluntary, state-wide mailing list of HIV researchers, healthcare providers, and social service providers, encouraging members to participate and distribute study information within their networks. Second, study flyers were distributed to local HIV health clinics in Western New York. Interested persons completed an eligibility screener and a brief demographic survey. Participants were eligible for the study if they were (1) ≥18 years old, (2) able to read, write, and speak English, (3) a healthcare (e.g., physician, pharmacist, physician assistant, nurse practitioner, nurse, dietician) or social service (e.g., social worker, community health worker, health educator, counselor, retention specialist) provider, and (4) working with PLH in the past 3 months. Eligible participants were sent a study information sheet and invited to complete an interview.

Interviews occurred with the Principal Investigator (PI) via phone or Zoom using a semi-structured interview guide, which directed discussion about the role of food and nutrition in HIV care, experiences discussing food insecurity with PLH, and barriers and facilitators to addressing food insecurity at each level of the Social Ecological Model (Bronfenbrenner, 1977, 1986, 1992; McLeroy et al., 1988). Participants received a $50 Amazon electronic gift card. Data collection occurred until saturation was achieved, and no new information emerged. All study procedures were approved by the University at Buffalo Institutional Review Board with a waiver for written informed consent.

Data analysis

Interviews were audio-recorded, transcribed verbatim, and averaged 38 min in length. We performed a thematic content analysis to report semantic level themes regarding barriers and facilitators to addressing FI in HIV care (Braun & Clarke, 2006, 2012). The analytic process occurred in five phases. In phase one, three analysts with expertise in qualitative data analysis (JB, KC, SP) familiarized themselves with the data and developed the initial codebook based on levels of the Social Ecological Model (e.g., patient, provider, clinic, and community). In phase two, two analysts (JB, KC) applied the initial codebook and coded a subset of transcripts (n = 6). The three analysts met regularly to revise the codebook and resolve any coding discrepancies through arbitration and mutual agreement (Weis & Fine, 2000). In phase three, the two analysts recoded the initial subset of transcripts using the revised codebook to ensure consistency (interrater reliability = 90%). In phase four, the two analysts independently coded the remaining subset of transcripts (n = 12). In phase five, the three analysts conducted a cross analysis to identify themes across levels of the Social Ecological Model through observed patterns in the qualitative data. The analysts met to discuss all themes and select representative quotes.

We applied criteria set forth by Guba (1981) to assess the rigor and trustworthiness of the data. To assess credibility, the PI reviewed the interview guide prior to any data collection and appraised the authenticity of participant responses by conducting post-interview assessments (i.e., documentation of general notes, concerns or problems, and reflections on the PI’s positionality). The PI ensured dependability and confirmability by adhering to all study protocols. The analysts upheld all criteria during the analytic phase through weekly meetings to ensure reliability, reconcile discrepancies, and reach consensus.

Results

Sample characteristics

Of the 18 participants, four were healthcare providers and 14 were social service providers, representing five healthcare agencies across NYS (herein, both healthcare and social service providers are referred to as providers). The majority identified as non-Hispanic White (50.0%) and female (72.2%). Age ranged from 23 to 66 years (M = 39.6 years, SD = 12.5). More than three-quarters of providers (83.3%) were employed at a community health center and nearly all (94.4%) had experiences discussing food and nutrition-related concerns with PLH. Additional sample characteristics appear in Table 1.

Table 1.

Sample Characteristics (N = 18).

Characteristic	n (%)
Age (years [M (SD)]	39.6 (12.5)
Race
Hispanic White	1 (5.6)
Non-Hispanic White	9 (50.0)
Hispanic Black	2 (11.1)
Non-Hispanic Black	5 (27.8)
Other	1 (5.6)
Sex at birth
Female	13 (72.2)
Male	5 (27.8)
Gender Identity
Female	13 (72.2)
Male	4 (22.2)
Genderqueer	1 (5.6)
Sexual Identity
Straight/heterosexual	13 (72.2)
Gay/lesbian	3 (16.7)
Bisexual	1 (5.6)
Pansexual	1 (5.6)
Provider Type
Healthcare	4 (22.2)
Social Service	14 (77.8)
Level of Education
College degree	11 (61.1)
Graduate/professional degree	7 (38.9)
Employment Setting
Community health center	15 (83.3)
Hospital	1 (5.6)
Physician practice group	1 (5.6)
Private clinic	1 (5.6)
Years working with people living with HIV (M [SD])	8 (6.1)
Experience discussing food and nutrition-related concerns with people living with HIV	17 (94.4)

Barriers and facilitators by level of the social ecological model

Thematic analysis highlighted eight themes illustrating barriers and facilitators to addressing FI in HIV care across various levels of the Social Ecological Model (Figure 1). Exemplar quotes for each primary and secondary theme appear in Table 2.

Figure 1.

Barriers (−) and facilitators (+) by level of the social ecological model.

Table 2.

Primary and secondary themes to address food insecurity in HIV care.

Primary Theme	Secondary Theme	Illustrative Quotes
Patient-level barriers	Embarrassment, shame, and judgement	"A lot of times, patients are ashamed that they are in this situation, and they don’t want to openly talk about it.” (31-year-old social service provider) "My patients have this shame and embarrassment about asking for a food pantry referral, which makes it hard for me to address [it].” (33-year-old social service provider) "[Patients] are not open and honest about not having food. They're embarrassed about it. They don’t want to come in here talking about having a hard time struggling to get food because they don’t want to be judged. I always tell them, I'm here for you, but I can’t help you if you don’t tell me.” (66-year-old social service provider)
	Health literacy	"They didn’t get infected with HIV because of food insecurity, so patients don’t understand the importance of how food security helps with HIV treatment and having a healthy immune system.” (33-year-old healthcare provider) "A lot of patients generally appear healthy and fit. That’s one thing that makes them put less weight into the importance of food security. Even if they don’t eat every day, it’s not showing in their physical appearance, so they think don’t need to prioritize that.” (31-year-old social service provider)
Provider-level barriers	Time	"Honestly, lack of time .... There's so much going on for these patients, that it’s hard to fathom adding one more thing to do. it’s hard to take all of the social elements, like food insecurity, into consideration.” (33-year-old healthcare provider) "The biggest barrier is time. Assuming the patient arrives on time, once they are checked in, the nurse has to do their updating, so the provider might get 15–20 min and has a list of medical things they need to address. The last thing that they are there to talk about is their food insecurity.” (39-year-old social service provider)
Provider-level facilitators	Patient-provider relationships	"One of the biggest things is coming from us as the providers is having a huge line of communication so that patients feel comfortable enough to tell us and ask us for the things that they need of, like food.” (49-year-old social service provider) "Having a relationship with them, knowing them and having met them before, that definitely sets the foundation where I can learn more about their background, their home life, and feel comfortable addressing their food situation. Similarly, if patients feel safe and comfortable coming to our clinic, they'll be more likely to like share more vulnerable or embarrassing topics like food insecurity.” (36-year-old healthcare provider)
Clinic-level barriers	Funding	"Not having the funding to be able to provide those immediate needs to address those food barriers definitely hinders. We have to fight hard to be able to even get the little funding we have.” (38- year-old social service provider) "With grant funded programs, things have to be done so specifically and the requirements are so much that it makes it a barrier for patients to engage in services.” (66-year-old social service provider) "We get funding to promote things for the patients, like food services, and they [funding agencies] make it difficult to get things. 'No, no, you don’t have the money for that.' Well, why not? Why can’t we get it to get to the patients? 'Well, that’s not really what it’s for.' Well, that’s what it should be for.” (54-year-old healthcare provider)
Clinic-level facilitators	Onsite resources	"Having our onsite food pantry certainly helps with what we need for addressing food insecurity among our patients.” (28-year-old social service provider) "I think the fact that there's a food bank makes a really big difference because there's instant access and supplies to address their barriers.” (32-year-old healthcare provider) "We can get patients set up with the food pantry to get them food and get some future appointments so they don’t have to stress like, 'will I have food next month?' Yes, you will. We're going to get you an appointment so you can get that food for next month. We definitely see how easily the weight just lifts off their shoulders knowing that they won’t have to worry about where their food is coming from that next month.” (23-year-old social service provider)
Community-level barriers	Community norms and intersecting stigmas	"We see it a lot with our folks who are HIV positive, that they still feel that stigma in the community. They're too afraid to kind of ask for help. They're afraid to come forward with their food insecurity needs because they don’t want people to find out about their diagnosis.” (28-year-old social service provider) "Some patients don’t like saying 'yes, I need help' because there is stigma in the community associated with using a food pantry.” (39-year-old social service provider) "There are a lot of compounding and stigmatizing factors that keep patients from receiving food services. A lot of my patients have a lot of nervousness about that. If their HIV status comes up, how are they going to be looked at and treated?” (28-year-old social service provider)
	Limited access to healthy food	"On the east side, there's really only one grocery store, and people within many square miles who don’t have a car or transit access have no access to healthy foods. That really hinders my ability to help.” (33-year-old social service provider) "It’s the lack of grocery stores. Public transportation is terrible. It’s a matter of if you don’t have a car and the nearest full-service grocery store is five or six miles away and the bus runs once an hour, and you have to catch two buses to get there. I can’t help if my patients don’t have physical access to food.” (31-year-old social service provider)

Patient-level barriers

Embarrassment, shame, and judgement.

Providers shared that PLH were reluctant to discuss FI. This hesitancy often arose from negative affect associated with receiving assistance, describing that most patients were embarrassed or uncomfortable to ask for help procuring food. These feelings of humiliation and shame often made it difficult for providers to alleviate food-related challenges. Providers also discussed that PLH were occasionally dishonest about their true needs. Often, this dishonesty was rooted in fears of judgement. Patients were ashamed of not having enough food to eat and feared their provider would place judgement on them.

Health literacy.

Providers perceived low health literacy and limited nutritional knowledge as a patient-level challenge to addressing FI. They felt that patients were unaware of the importance of proper nutrition in HIV care. Most providers discussed that disease management is the highest priority for PLH. Yet, the importance of nutrition in that process is frequently overlooked. Often, patients equated HIV health with physical appearance. These misperceptions made it difficult for providers to alleviate their patients’ food-related concerns when instances of FI arose.

Provider-level barriers

Time.

Despite nearly all providers acknowledging the importance of screening and addressing FI in HIV care, a commonly reported challenge was lack of time. Providers described the various complexities of their patients’ HIV treatment and care which limited their ability to screen for and address FI concerns. Brief medical appointments often required providers to prioritize medical over social concerns. Providers were not able to discuss or address their patients’ FI concerns because of these time restraints.

Provider-level facilitators

Patient-provider relationships.

Providers described strong patient-provider relationships as a facilitator to addressing FI among PLH. Building and maintaining rapport allowed patients to feel comfortable disclosing their food-related concerns with providers. Providers also viewed these relationships as having reciprocal benefits. They described how having good rapport with patients makes it easier to engage in discussions about FI.

Clinic-level barriers

Funding.

Providers frequently discussed how limited funding and bureaucratic requirements attached to funding sources hindered their ability to ameliorate FI. For example, lacking financial resources was a commonly reported obstacle to establishing and maintaining food programs at clinics. Among providers whose workplaces offered food-related programing, insufficient funding prohibited the expansion of these programs, despite patient need. Providers also described that documentation processes required for grant funding were arduous and hindered patient engagement. These funding-related restrictions often limited providers’ abilities to address FI among PLH.

Clinic-level facilitators

Onsite resources.

Providers described that onsite services were important facilitators to mitigating FI. Clinic-based programs such as food pantries helped to address FI concerns in real-time. Providers discussed how onsite services often removed many challenges to addressing patients’ food-related concerns, and nearly all providers emphasized the positive effects of these resources. Nearly all providers described that patients felt a sense of relief knowing that they could receive food from the same place where they receive their HIV care.

Community-level barriers

Community norms and intersecting stigmas.

Providers described the ways in which community norms towards PLH and people who receive food assistance created intersecting stigmas that served as a challenge to ameliorating FI. While providers described feelings of embarrassment, shame, and judgement as a patient-level barrier, providers noted that HIV stigma often operated at the community-level. Providers described that HIV stigma is still prevalent in many communities. This stigma often leads patients to be reluctant to accept help, resulting in their refusal of various HIV care-related services. Providers also described community-level stigma towards people who need help procuring food. These community norms created intersecting stigmas that kept PLH from receiving food services. These intersecting and compounding stigmas made it difficult for providers to address FI among PLH.

Limited access to healthy food.

Providers noted that most of their patients live in areas with limited access to healthy food. Providers discussed how many PLH lived in areas with no grocery stores, or lived in areas where healthy and nutritious foods were too expensive. This was influential in their ability to address FI among PLH. Providers described how otherfactors, such as socio-structural determinants, including inadequate transportation and limited income sources, contributed to the development of neighborhoods with limited access to healthy food. These factors often made it difficult for providers to intervene for their patients.

Discussion

In this study, providers highlighted several barriers and facilitators to addressing FI in HIV care at various levels of the Social Ecological Model. First, providers in our study noted that PLH were reluctant to discuss their challenges accessing food. These barriers often rose from PLH feeling embarrassed and ashamed about needing assistance procuring food. Qualitative investigations with healthcare providers in other clinical settings have demonstrated similar results (Douglas et al., 2020; Pooler et al., 2018). Providers recognized FI as a sensitive topic and understood that patients may be reluctant to discuss it (Douglas et al., 2020). It is possible that patient hesitancy to discuss FI is a function of stigma associated with food assistance at the community-level (Douglas et al., 2015; Middleton et al., 2018).

Second, low health literacy regarding the importance of nutrition in HIV care was also a patient-level challenge. Consistent with previous work (Pooler et al., 2018), providers in our study described that PLH often lacked basic nutritional education, which limited their ability to relieve food-related concerns. In HIV care, low health literacy has been associated with worse HIV outcomes (Reynolds et al., 2019). Our results suggest that low health literacy among PLH may extend beyond HIV outcomes and impact the ability of providers to address broader social needs. Given the importance of nutrition education in comprehensive HIV care (Food and Nutrition Technical Assistance Project, 2004), our findings have implications for incorporating nutrition education along each step of the care continuum to address low food-specific health literacy.

Providers often described how the complex health needs of their patients limited their time to screen for and address food-related concerns whereby medical needs took priority over social concerns during brief clinic appointments. Research among primary care providers has illustrated time as the most common barrier to screening and addressing FI (Kostelanetz et al., 2022; Pooler et al., 2018; Schickedanz et al., 2019). Given most of our study sample were social service providers, our findings highlight time as a universal challenge to amending social needs, such as FI, in HIV care regardless of provider type (i.e., healthcare or social service). Our results suggest that HIV providers may benefit from leveraging workflows that incorporate FI screeners and resources, as previous qualitative work found this strategy to be effective at minimizing time-related barriers (Adams et al., 2017; Palakshappa et al., 2017). For example, incorporating FI screeners into electronic health records has reduced time-related constraints in other healthcare settings (Adams et al., 2017).

Establishing and fostering strong patient-provider relationships supported providers’ abilities to address FI in our study. Previous literature illustrated how screening for FI has the potential to improve patient-provider relationships (Adams et al., 2017; Palakshappa et al., 2017; Schickedanz et al., 2019). The patient-provider relationship plays an important role in long-term care engagement for people with chronic and infectious diseases (Ciechanowski et al., 2001; Rogal et al., 2016). Similarly, robust interpersonal connections and established rapport between patients, healthcare, and social service providers have been shown to promote successful engagement in HIV care (Flickinger et al., 2013; Langebeek et al., 2014). Our results suggest that establishing and maintaining strong patient-provider relationships has the potential to extend beyond disease management to include addressing FI among PLH.

Onsite food banks and pantries were described as clinic-level facilitators to addressing FI. This finding is unique, given that research among clinicians in other settings has consistently cited a lack of facility resources as a challenge to mitigating FI concerns (Kostelanetz et al., 2022; Pooler et al., 2018; Schickedanz et al., 2019). Our finding may be a function of HIV-specific funding, such as the Ryan White HIV/AIDS Program (RWHAP) Food and Nutrition Services category, which provides federal funding for “Food is Medicine” programs and services, including medical nutrition therapy, home-delivered meals, and food pantries (Aidala et al., 2015; Health Resources & Service Administration Ryan White HIV/AIDS Programs, 2022a, 2022b). Yet, consistent with previous work (Pooler et al., 2018), providers in our study noted insufficient funding as an obstacle. Providers recounted instances where clinics were unable to establish or expand programs due to limited or restricted funding. In FY2016, 77% of RWHAP-Part A and 58% of RWHAP-Part B grantees allocated support services funding for food banks and/or home-delivered meals (Health Resources and Services Administration, 2018a, 2018b). Our results suggest the need to optimize funding for food programs in HIV care.

While research has begun to evaluate the impact of community-based food support interventions among PLH (Palar et al., 2017; Yu et al., 2022), few studies have extended this work to clinical settings (McKay et al., 2017; Tanner et al., 2022). Clinics may benefit from evaluating current food-related programs and implementing cost-effective, ecological interventions to address the needs of food insecure PLH. Further, given that PLH have increased contact with health-related services, our results suggest that integrating FI programs into routine care may incentivize care engagement; a finding echoed in previous qualitative work (Gagnon et al., 2020; Gagnon et al., 2021).

In our study, providers described intersecting stigmas related to one’s HIV status and food procurement as common obstacles to addressing FI. HIV-related stigma remains prevalent in many communities and has been shown to influence care engagement for PLH (Tan et al., 2023; Tarfa et al., 2023). Further, stigma associated with food pantries has been demonstrated to reduce use of these services in communities (Bruckner et al., 2021; Douglas et al., 2015; Middleton et al., 2018). Our results provide qualitative insights into how the intersecting and compounding roles of these stigmas may exacerbate FI among PLH, impact engagement in the HIV care continuum, and reduce providers’ abilities to address FI in HIV care.

Lastly, providers in our study noted that patients lacked access to affordable and nutritious food options. Research among PLH has illustrated inadequate access to food as a primary barrier to HIV care engagement (Bleasdale et al., 2022; Lee et al., 2021) and a contributor to poor HIV outcomes (Pellowski et al., 2016; Weiser et al., 2020). Further, providers described how inadequate public transportation systems compounded their ability to address FI among PLH. Previous work among providers has described unreliable and insufficient transportation as a barrier to engaging PLH in HIV care and other services, such as mental health and housing (Doraivelu et al., 2022; Roland et al., 2022). Our results extend the current body of literature by illustrating the role of community factors in affecting providers’ abilities to alleviate food concerns among PLH.

Limitations

Our study should be interpreted within the context of its limitations. First, our results are limited by our small sample size (N = 18), despite achieving data saturation. Second, a greater number of social service providers participated in this study compared to healthcare providers. A more diverse sample of healthcare providers is essential for future work that seeks to examine successful strategies to address FI in HIV care. Third, this study relied on convenience sampling from one state in the U.S.; therefore, our findings may not be transferrable to other geographic locations. Lastly, data analyzed in our study relied on semi-structured interviews to facilitate discussions, which may present social desirability and recall biases.

Conclusion

In this study, providers highlighted several barriers and facilitators to effectively addressing FI in HIV care. Our results add to the burgeoning body of literature highlighting the need to integrate FI interventions into HIV care for PLH using an ecological approach. Study findings have the potential to generate new lines of scientific inquiry for mixed methods and quantitative investigations and may assist with the evaluation and implementation of clinic-based “Food is Medicine” programs to address FI in routine HIV care.

Data availability statement

The interview guide is available from the corresponding author upon reasonable request.