Studies consistently show that food insecurity increases vulnerability to HIV and worsens health outcomes for people living with HIV. While federal and state programs in the U.S., such as the Supplemental Nutrition Assistance Program (SNAP), provide food assistance to low-income people, income guidelines and other restrictions can keep such programs out of reach for people impacted by HIV.
In a recent study, Aaron Richterman, M.D., M.P.H., an infectious diseases fellow at the University of Pennsylvania, and colleagues explored the relationship between access to SNAP and HIV health outcomes. The study found that people with food insecurity make pressured choices—and for people living with HIV, Richterman says, that translates to prioritizing their short-term food needs over their long-term health. Another key finding: Increased access to SNAP was associated with decreases in HIV diagnoses.
Richterman presented the study in October at the IDWeek 2020 medical science conference. I spoke with him about the findings and broader takeaways, given that Richterman’s research generally focuses on the relationship between poverty, food insecurity, and individual and community health outcomes for infectious diseases, with particular attention to people living with HIV, tuberculosis, and cholera.
This transcript has been edited for clarity and length.
Terri Wilder: Thanks so much for speaking with me today.
Aaron Richterman: Thanks so much for chatting. I’m always happy to talk about this subject.
Wilder: During your IDWeek presentation, you stated that the Supplemental Nutrition Assistance Program is the primary program in the United States that addresses food insecurity. What is food insecurity, and how does it intersect with HIV?
Richterman: Food insecurity is defined as a lack of stable access to food in adequate quantity or quality. That’s the technical definition, but in a more general sense, you can think of it as people who are living in a situation where they don’t know where their next meal is going to come from.
When you look at food insecurity and health, basically every single health outcome or condition that we have looked at with relation to food insecurity, we found that living in a food insecure state has negative impacts on health outcomes.
Probably the condition in which this has been best studied is HIV. People who are living with HIV who are food insecure have lower adherence to their antiretroviral therapy. They have higher rates of not having their viral load suppressed. They’re more likely to resort to high-risk coping strategies, including exchanging sex for food or money.
It’s not as simple as just: “They don’t have enough food, so their nutrition is not good.” Nutrition is one component of this. A lot of it is the ways that living in this state affects people’s behavior. People who are unsure of where their next meal is going to come from will tend to prioritize their short-term food needs over their long-term health. They’ll make pressured choices. Their risk calculations become altered. And all of this has both near and far impacts on their health outcomes.
In addition to that, just living in this food-insecure state causes people to have this heightened stress response, where their stress hormones are elevated. Their levels of chronic inflammation, when you actually measure the blood, are elevated. Both stress and chronic inflammation have been implicated in outcomes for HIV.
So, what we think is that really that this food-insecure state is very important for people with HIV, in terms of affecting their ability to control the infection and also to be in control of their lives.
Behind the Study: Why Researchers Explored the Link Between SNAP Benefits and HIV Health Outcomes
Wilder: Tell me about the research that you and your colleagues conducted on this topic.
Richterman: Just to give a little bit of background, we were looking at the Supplemental Nutrition Assistance Program, or SNAP—which, as you mentioned, is the primary safety net program for food insecurity in the United States. There’s very good evidence that this program reduces food insecurity and poverty in the United States.
SNAP is a federal program that provides money to people who qualify for it to spend on food. And, in the U.S., there are currently more than 36 million Americans who receive this benefit. They receive an average of $1.35 per person, per meal.
The program itself is a federal program, but there are some ways that states can affect who is eligible for it and who has access to SNAP. One mechanism that the federal government—through a policy called Broad-Based Categorical Eligibility—allows for is that the states can alter both the income eligibility and the asset eligibility for people to enroll in this program.
What that means is that at baseline, people qualify for SNAP if their income is at or below 130% of the federal poverty level, and if they have assets that are below $2,250 in household assets. If they meet those eligibility criteria, then they’re able to qualify for SNAP.
With Broad-Based Categorical Eligibility, what this means is that a person who qualifies for the Temporary Assistance for Needy Families program—which is a state-administered program—if they qualify for that, they automatically qualify for SNAP. This both reduces the administrative burden to applying for two programs, and it allows states to increase the income limit or change the asset limit. They can increase the income limit from 130% of the federal poverty level to up to 200% of the federal poverty level. They can also reduce the asset limit, or they can eliminate it.
We see that there’s a broad range of income limits and asset limits that are put into place by states to make people eligible for SNAP. So, what we hypothesized was that—based on the connection between HIV and food insecurity, and given that we know that SNAP reduces food insecurity—there may be a relationship between state policies that make it easier or harder for people to get SNAP and HIV outcomes.
In the current paradigm of treatment-as-prevention, by finding people living with HIV, helping them remain in care, helping them get on antiretroviral therapy, and keeping the virus suppressed, that’s one of the best ways to prevent transmission of HIV within a community. Based on that, the thought was that perhaps in states where it’s easier for people living with food insecurity to get access to SNAP, they may be able to better control their HIV and they’d be less likely to then transmit HIV to others.
What we looked at is all states between 2010 and 2014 and looked at these policies—the income eligibility policy and the asset policy. We looked at changes in those policies over that time period, and then the relationship between that and the rate of new HIV infections in those states. We controlled for a number of different variables that may also affect new HIV diagnoses. We looked at things like the cost of food, the amount of health care spending, the housing pricing index, the levels of unemployment, and then other SNAP policies, as well, and spending on other social programs.
Study Results: SNAP Income Limits Affect HIV Incidence Rates
Richterman: What we found is that, after controlling for all of these other things, that the state income limits for SNAP eligibility were very closely associated with the number of new HIV diagnoses, both in the year that eligibility changes were made and in the several years after that. But we didn’t find any association between the asset limits and new HIV diagnoses.
To put this into some numbers, what we basically found was that if you think about 130% of the federal poverty level as the lowest eligibility level that people can have, and 200% as the highest, increasing the income eligibility level about halfway—from 130% to 165%—was associated with a 6% decrease in the rate of new HIV diagnoses. A hypothetical state that increased from 130% to 200% for the income limit would then be associated with a 12% decrease in the rate of new HIV diagnoses.
And just to put this in a little bit of context, the rate of new HIV diagnoses in the United States has been pretty consistent. It went down quite a lot leading up to the years about a decade ago, but over the course of the last decade, it’s plateaued at around 39,000 or so new infections per year.
There are efforts on the federal level to try to reduce this. There was a plan announced in 2019 under the Department of Health and Human Services that was called Ending the HIV Epidemic: A Plan for America. The goal of this plan was to reduce HIV infections to less than 3,000 new infections per year by 2030.
At the same time, the Department of Agriculture under the Trump administration has proposed making a variety of changes to SNAP, one of which would be to get rid of Broad-Based Categorical Eligibility, which would essentially eliminate the ability of states to increase the income limit.
Part of the implications of these findings is that by making a change in policy like that—where we would go from states being able to increase the income eligibility limit to the income eligibility limit being brought down to 130% across the board—that would be associated with, give or take, approximately about a 10% increase in new HIV infections. Really, this would be undermining our efforts in this country to end the HIV epidemic and to bring down the rate of new infections to the goal that we’ve set for 2030.
Wilder: For the states where you saw more severe policy changes that would impact people, did you find a pattern? Were the states impacted most likely to be in a certain region of the country?
Richterman: I think a couple things. First of all, the broader pattern of policy changes in this regard has been one of increasing eligibility for the program. Some states are more, some states are less, and some states haven’t increased really much at all. But over the last 20 years, in general, the income eligibility limits have increased. That’s one thing to keep in mind.
The second piece is that this planned change of getting rid of Broad-Based Categorical Eligibility, which would really reduce the income limit for all states down to 130%; that has not yet been implemented. That was in the final stages of approval early this year, when the pandemic hit. Since then, we haven’t heard anything about it. It’s still listed on the website as being in the planning stages.
What I would say—thinking about a couple of examples here—when you’re thinking about states in different regions and different areas in the country that are hot spots for HIV transmission, one example is Georgia, where there were about 2,500 new cases of HIV identified in 2018, according to the latest data. Their income eligibility limit for SNAP is at the baseline: 130%.
So, what our findings suggest is that if they were to increase their income limit up to the maximum currently allowed of 200%, that would be associated with a reduction in new infections of over 10%. That would potentially prevent around 300 infections per year. That is one way of thinking about one place that has a lot to gain.
In Pennsylvania, where I live, the income limit for eligibility is 160%. We’re not at the top, and we’re not at the bottom. If we were to see our income limit decrease to 130%, we would go from about 1,000 new cases per year to a predicted [reduction of] 70 or so.
On the flip side, you can think about places where there is quite a lot to lose. For example, in New York State, the income limit for eligibility is currently 200%. New York State had close to 2,500 new infections diagnosed in 2018. If Broad-Based Categorical Eligibility was removed, that would be associated with about an increase of 300 cases per year.
I think that because states have all made slightly different changes to their eligibility limits, there’s both room to really improve things in places where we were able to increase the income eligibility limit, and there’s a lot of damage that can be done by decreasing it in places that have already increased it.
In general, parts of the country that have not done much as far as increasing the income eligibility limit tend to be in the South, which is also an area of a lot of HIV transmission and hot spots for this infection in the United States.
I gave you the example of Georgia, which has not increased the income eligibility limit. The same is true for Alabama, where they had over 600 cases in 2018. Kentucky, Louisiana, Kansas, Tennessee, and Virginia—none of these states have increased the income eligibility limit.
Whereas the more imminent threat is the loss of this policy for the places that have made changes, there is also opportunity in places where we haven’t seen increases that could really make inroads into reducing new infections.
You Can’t End HIV Without Social Support and HIV-Specific Food Programs
Wilder: New York State was the first state in the country to have a governor’s plan to end HIV. People in Georgia are working on their plan as well.
What are your thoughts on advocacy around policies to either stop these programs from going away completely or to increase funding?
Richterman: That’s a really good question because anyone who is working in HIV, or who has spent any amount of time learning even just a little about it, understands the very important role of social determinants in both individual outcomes and community outcomes. Unfortunately, there is a disconnect between that understanding and many of the policies that have been put in place.
There is emphasis—rightly so—on finding people with HIV, getting the diagnosis, getting people connected to care, and getting people on antiretroviral therapy. There’s a lot of emphasis on the biomedical part of this.
There’s relatively much less effort, in my opinion, spent on the social support aspect of this, on the poverty reduction part. A really important message is that social programs should be seen as part and parcel to the treatment of HIV, and part and parcel to the prevention of HIV. Right now, they’re not thought about at the same level.
Part of this advocacy—and part of what these findings, and other studies like this, can really point to—is that there is a measurable effect from safety net programs and from social support, and that those have direct community outcomes for us. We can have all the access to antiretroviral therapy in the world, have all the access to preexposure prophylaxis, all the biomedical pieces in place; but for this disease and in our context, I truly believe—and I think this data would support—that we’re not going to be able to fully control this epidemic until we really do focus on poverty reduction and social support, and on having that more holistic approach.
I hope that, in terms of advocacy, looking at studies like this and data like this can put an actual number on the effects of social policies—on both the benefit that we can get by expanding access and also the harm that we can propagate by getting rid of access. And I hope that the studies and data really point out the contradiction between a plan that we’re putting out—saying that we’re going to reduce new infections by 90% by 2030—at the same time as reducing access to a program that, even as it stands right now, is able to reduce new infections by over 10% per year.
Highlighting that contradiction is important as we think about expanding our toolbox for ways to end the HIV epidemic in the U.S.
Wilder: Ryan White Part A dollars can go to food and nutrition services, and that includes services like home delivered meals, congregate meals, pantry bags, and vouchers. And I know that the Centers for Disease Control and Prevention is charged with HIV prevention funding. Is there a role here for the CDC to use HIV prevention dollars for food insecurity?
Richterman: There are two ways you can think about interventions that target food insecurity in the context of HIV. There are programs that are HIV-sensitive, meaning that they aren’t specific to people living with HIV, but by their nature, they end up helping people with HIV. SNAP would be a policy like that, where SNAP is really targeting people with food insecurity, people living in poverty.
By the nature of the disease and the conditions of our country, many people living with HIV are living in poverty and living with food insecurity. So by expanding this safety net program, we end up helping people with HIV, in addition to many other people.
What you’re talking about is more of an HIV-specific type of program, where we look at the population of people with HIV as one population that is especially high risk for impoverishment and food insecurity. They’re also especially high risk for the poor outcomes associated with those conditions. And so we think about specific programs for that population.
These kinds of programs can be, in the way that you outlined, implemented through HIV clinics or city health departments and programs like that—rather than looking at this big program for the whole population, looking specifically at the HIV population. I would argue that there is a role for that.
In a country like the United States, the safety net programs should, at the end of the day, be the big players here, because that’s really helping everyone in those conditions. But the reality is that, many times, they don’t do enough. I think there is a role here for CDC dollars, Ryan White dollars, HIV clinics, and public health departments to use programs targeting food insecurity to help improve specifically these outcomes.
There is potentially a very high benefit to be gotten by undertaking these programs, specifically targeting these populations, with the goal of alleviating hunger—which is a great goal in and of itself. The goal is very much in line with the goals of these institutions, which is to get people to control HIV better and reduce community incidence of the disease.
So I’m a big advocate of incorporating thinking about this type of issue within HIV programs. Because, outside of it, the safety net is just not strong enough at this point to tackle the issues for people living with these problems.
When you look at places outside the United States, where there is no safety net—for example, I spent a lot of time working in Haiti—in those places especially, that’s where these types of HIV-specific programs can be hugely beneficial to improving outcomes.
The Responsibility of Clinicians and Researchers in Pushing for Policy Changes
Wilder: What if you and your colleagues reached out to these state health departments or government officials and you told this story? What impact would that have?
Richterman: First of all, it is incumbent on us, as researchers and clinicians and advocates, to do that kind of work and to educate government officials about these issues. And so, I think that’s really important wherever we are.
What is complex in the United States is that we have 50 states, and they all have slightly different approaches and interests to these sorts of things. Some of them are more friendly to the idea of social programs and social support, and some are less friendly. The fight for these programs is going to look a little different, depending on the environment that we exist in.
I’ll give you an example. I did my training in Massachusetts, which is probably one of the more generous states when it comes to social programs. That’s a place where I have seen a lot of really amazing on-the-ground work being done to target food security in the context of health outcomes. Not to pick on Georgia again but, in Georgia, where there are lots of new infections every year, the government is less amenable to thinking about these sorts of social programs.
We have to keep communicating the message and putting the numbers to it and saying that this is undermining all of our ability to live in a healthy society, and to make the argument for it. I think the success will be different in different places.
The message that needs to be communicated to people in power, and it needs to be communicated loud and frequently, is, “Look. Here’s the actual human cost of not enacting these kinds of policies. Here’s the potential benefit.” It will vary based on where we live.
Wilder: What’s next for you, in terms of studying this topic? Is there anything to further analyze from this data?
Richterman: There are a number of approaches that I’m thinking about that will be important as we work through this further. One is to take a look at some of the other policies around SNAP.
One of the other big SNAP policies that is currently being litigated in court is an attempt by the Department of Agriculture—again, under the Trump administration—to basically get rid of certain ways for states to waive a work requirement for certain non-disabled adults who do not have dependents. As we’ve seen in other sorts of situations, work requirements generally do not fulfill the stated goal of getting more people to work; of improving employment and improving self-sufficiency. All that work requirements really do in other conditions is to exclude people from social programs.
And so, I suspect that, like with these other policies, implementation of work requirements may potentially have similar impacts on HIV outcomes. That’s another policy that I’m really interested in taking a deeper look at.
Then outside of SNAP, we do a lot of work in very low-income settings, settings of extreme poverty, as well. I’m very interested in the role of food insecurity in settings of extreme poverty with HIV and, in particular, in rural Haiti. We’re working on a number of projects there to look at the relationship between food insecurity and health outcomes.
The goal is ultimately to move from the observation that people who live with food insecurity don’t do well, in whatever health outcome we measure, to: What can we do about it? What does an intervention look like that’s actually able to help people in these conditions? What are the nuts and bolts of that policy change or clinical intervention, or something that’s going to be incorporated that can improve this person’s condition?
Then, when you can develop that kind of intervention, you can point to it and say, “Look. We have the evidence behind something that is going to improve outcomes for this person. It’s going to improve outcomes for this community. It’s well worth our investment to do so.”
We spend all this money to get antiretroviral therapy to people. We should be thinking about this in a very similar way. That’s a long-term goal—to be able to point to specific interventions and say, “This is what we’re going to be able to do about this problem to improve people’s lives and improve the health of the community.”