When Natalia sat on the exam table, her posture conveyed a sense of exhaustion that seemed to emanate from her core. Her dark hair obscured her face, a veil behind which she could retreat from the world. There was weariness in her, a resignation that came from a life marked by hardship.

Natalia had come to me for treatment of hepatitis C, a disease that had taken a heavy toll on her liver. But as I reviewed her chart, I noted the additional challenge she faced: a history of fentanyl addiction. As we began to talk, I couldn’t help but feel a sense of disconnect. Natalia answered my questions with short, guarded responses, and I found myself struggling to break through the invisible barrier between us. It was clear that her fentanyl addiction was a major obstacle in her life. She was homeless, unable to hold down a job, and her health was deteriorating rapidly.

Over the next few weeks, we worked on treating Natalia’s hepatitis C, but progress was slow. She would miss appointments or show up under the influence, and I began to feel a sense of frustration. I couldn’t understand why she wouldn’t just quit using drugs and focus on her health.

It wasn’t until we started addressing Natalia’s depression that things began to shift. As she opened up about her past, I realized that her addiction was rooted in a much deeper trauma. Natalia had been born in Ukraine and had dreamed of a better life in America. However, when she arrived, she found herself in the clutches of human traffickers who operated in the Capp Street district of San Francisco and expertly exploited her vulnerability.

For years, Natalia was forced to work in the sex trade, with her captors using drugs to keep her compliant. She described the horror of being injected with fentanyl against her will, the way it would make her numb and disconnected from the repeated rapes she endured. Even more heartbreaking was the story of her pregnancy. Natalia had held her baby girl for only a few precious moments before her captors took the child away, never to be seen again.

As I listened to Natalia’s story, I felt a deep sense of shame for my initial judgment. I had assumed that her addiction was a choice, amoral failing on her part.

Natalia’s story is not unique. In my years of working at a federally qualified health center (FQHC) in San Francisco’s South of Market district, I’ve seen countless patients whose lives have been ravaged by the fentanyl epidemic. Many of them have histories of trauma, whether it’s childhood abuse, domestic violence, or the horrors of war. For some, drug use starts as a way to self-medicate untreated mental health issues. For others, it’s a means of survival on the streets.

One patient, a young gay man named Ethan, started using fentanyl after witnessing the horrific mass shooting at a Colorado nightclub. He had escaped the carnage physically unscathed, but the emotional trauma was overwhelming. At first, he found solace in the leftover oxycodone from his grandmother’s knee surgery. But when the prescription ran out, he turned to the cheap, potent fentanyl that flooded the streets.

Another patient, Javier, had grown up in a household where any sign of weakness was met with brutal beatings from his father. When he started hearing voices and experiencing panic attacks, he was too ashamed to seek help. Instead, he started using fentanyl, first in the form of patches sold to him by a classmate, then as a powder he would smoke to numb his pain.

Despite their different backgrounds and circumstances, all of my fentanyl-addicted patients share one thing in common: the crushing weight of stigma. They are acutely aware of how society perceives them, and many go to great lengths to distance themselves from the label of “addict.” They insist that they are not like “those people,” that their use is under control. But beneath the surface, they all carry the same shame and self-loathing, the feeling that they have somehow failed at life.

The reality is that addiction is not a moral failing, but a complex interplay of social, economic, and psychological factors. For many of my patients, their drug use is a symptom of the trauma they have endured, a desperate attempt to cope with the pain of their past and the harsh realities of their present. Yet, society continues to view addiction as a choice, a personal weakness to be overcome through sheer willpower.

This stigma is only compounded by the lack of accessible and effective treatment options. Medicaid, which covers many of my patients, has historically placed strict limits on addiction treatment, making it difficult for those who need help to get it. Even when treatment is available, it often fails to address the underlying social and emotional issues that drive addiction in the first place.

This is where the role of social determinants comes into play. Addiction, like so many other health issues, is deeply intertwined with factors like poverty, trauma, and lack of social support. For many of my patients, their drug use is not an isolated problem, but a symptom of a much larger web of challenges.

Take Natalia, for example. Her addiction to fentanyl was not simply a matter of personal choice, but a direct result of the trauma she endured as a victim of human trafficking. To truly address her addiction, we would need to provide not only medical treatment, but also mental health support, safe housing, and a path to legal status and employment.

Similarly, for patients like Ethan and Javier, addressing their addiction would require a holistic approach that takes into account their histories of trauma and untreated mental health issues. It would mean providing access to therapy, support groups, and other resources to help them process their pain and build healthy coping mechanisms.

In recent years, there has been a growing recognition of the importance of addressing social determinants of health, particularly in the context of Medicaid. Medicaid beneficiaries are disproportionately affected by social needs such as food insecurity, housing instability, and lack of transportation compared to those with commercial insurance or Medicare. A majority of Medicaid beneficiaries report at least one unmet social need, whereas such needs are far rarer among people with commercial insurance. This is not surprising given that Medicaid primarily serves low-income individuals and families who are more likely to experience poverty and its associated challenges.

In response, many state Medicaid agencies have begun mandating the use of rapid screening and referral programs to identify and address beneficiaries’ social needs. These programs typically involve using standardized questionnaires to assess patients’ needs, followed by referrals to community organizations or social service agencies. To facilitate this, many states have contracted with expensive electronic referral systems. However, these mandates have placed a significant burden on already stretched-thin primary care providers and emergency room staff who may lack the time and resources to adequately address patients’ social needs.

One major limitation of rapid screening and referral programs is that they often rely on a superficial understanding of social needs. A checklist can only capture so much information and may not fully reflect the complexity of a patient’s situation. For example, a patient reporting food insecurity may be struggling with the stigma of admitting they steal food when budgets are most constrained. The act of screening itself can also be problematic. Many patients feel uncomfortable or stigmatized by the questions, particularly if they are already dealing with the shame associated with poverty. This can lead to under-reporting of needs or even healthcare avoidance.

Even when social needs are accurately identified, the referral process can be fraught with challenges. Many community organizations and social service agencies are chronically underfunded and understaffed. They may lack the capacity to handle a sudden influx of referrals from healthcare providers, leading to long wait times, limited services, or outright denial of care. Navigating the complex web of social services can also be overwhelming for patients, particularly those already dealing with significant health and social challenges. Without adequate support and follow-up, many fall through the cracks and never receive the services they need.

The proliferation of electronic referral systems has exacerbated these issues. While promising to streamline the screening and referral process, these systems are often designed by tech companies with little understanding of the complex realities of poverty and social determinants of health. They frequently use algorithms that oversimplify patient needs or generate referrals to organizations not actually equipped to provide the necessary services. By generating a high volume of referrals without adequate follow-up, these tools can create a “revolving door” effect where patients are repeatedly referred to services that don’t meet their needs. The result is a system more focused on generating referrals and checking boxes than actually improving patient outcomes.

Healthcare agencies are often poorly positioned to effectively and efficiently tackle the root causes of social needs. They are primarily designed to deliver medical care, not social services, and may lack the necessary expertise, resources, and community connections. When healthcare agencies do attempt to address social needs, they often do so in a piecemeal, uncoordinated fashion—screening patients and making referrals but lacking the capacity to ensure patients actually receive the services they need. This leads to a fragmented system where patients are bounced around without getting comprehensive support.

Healthcare agencies also frequently assume what patients need without allowing providers the time to listen to their perspectives and experiences. This top-down approach focuses more on meeting the healthcare system’s needs than addressing patients’ real-world challenges. Furthermore, when healthcare agencies invest in social interventions, they often expect a quick return on investment and fund programs designed to reduce short-term costs rather than addressing long-term needs. There is a risk of medicalizing social needs, viewing them as problems to be solved through clinical interventions rather than broader social and policy changes.

The current approach of relying on healthcare agencies to address complex social issues is deeply flawed. A prime example is the proliferation of “food as medicine” initiatives in healthcare, which have had mostly null results compared to the positive outcomes from deferring to the national Supplemental Nutrition Assistance Program (SNAP), formerly known as the Food Stamp Program—which is run entirely outside of healthcare entities. SNAP has less administrative overhead but interestingly better health outcome measures across multiple studies.

For example, a randomized controlled trial tested the effectiveness of an intensive “food as medicine” program providing medically tailored meals, nutrition education, and clinical support to patients with uncontrolled type 2 diabetes. Despite the program’s significant cost and comprehensive nature, there was no difference in HbA1c levels (a key diabetes biomarker) between the intervention and control groups after 6 months. In contrast, a study analyzing national datasets found that SNAP participation was associated with lower healthcare expenditures, likely by alleviating food insecurity and freeing up resources for medical care and medications.

These findings underscore the importance of deferring to established, efficient programs like SNAP to address social needs rather than recreating the wheel through resource-intensive healthcare interventions. SNAP’s success lies in its relatively low administrative burden, allowing more funds to directly reach beneficiaries as a debit card that can be used at most stores, and in its ability to improve dietary quality and health outcomes by increasing access to nutritious foods with greater flexibility and convenience than a food program run exclusively out of a hospital or clinic. Indeed SNAP appears to have more benefits ironically because it’s not medically tailored; many recipients report that the extra income to help offset other expenses like rent or heating that impact health in indirect ways, and offset the unpredictable nature of debts and expenses among people who face poverty.

Rather than trying to address social needs in a vacuum, the healthcare system needs to partner with existing organizations that have expertise in these areas. Medicaid agencies should focus on building relationships and capacity with community organizations to ensure they can handle increased referrals. Investing in social services infra-structure is key to avoiding “bridge to nowhere” situations where patients are referred to overburdened agencies.

Researchers often refer to the “wrong pocket problem” — a situation where the benefits of an intervention accrue to a different sector (e.g., increasing funds to a hospital) than the one that bears the costs of implementing it (e.g., the local food bank). This misalignment of incentives can lead to underinvestment in interventions that could improve overall health and well-being.

In the context of addressing social determinants of health, the wrong pocket problem often arises when healthcare organizations refer patients to social services but the resulting cost savings and benefits accrue to another sector, such as insurance companies. This point can be illustrated using the example of food insecurity interventions. If a private human services organization provides a food insecurity intervention that improves health outcomes, the benefits may primarily accrue to the healthcare system in the form of reduced healthcare costs. This creates a wrong pocket problem, as the organization investing in the intervention does not directly reap the financial rewards.

As primary care practices increasingly screen for social needs, it is crucial that we develop a more comprehensive understanding of the resources required to effectively intervene. Recent studies provide a valuable starting point for estimating these costs and identifying gaps in current funding mechanisms. By investing in evidence-based interventions and strengthening community-based capacity, we can begin to build a more equitable and effective system for addressing social determinants of health.

In a recent study published in JAMA Internal Medicine, my colleagues and I sought to estimate the cost of implementing evidence-based interventions to address social needs identified in primary care practices. We studied nationally representative data from the National Center for Health Statistics to estimate needs among patients seen in primary care practices and their self-reported social needs in the domains of food, housing, transportation, and community-based care coordination. We looked at how much funding was already available, and how much new funding would be needed, to provide evidence based solutions to each patient’s social needs.

Our analysis revealed that among people with food and housing needs, most were eligible for federally funded programs, but had low enrollment due to factors such as inadequate program capacity. For example, 78.0% of people with housing needs were program eligible, but only 24.0% were enrolled. By contrast, among those with transportation insecurity and care coordination needs, eligibility criteria were the primary limiting factor, with only 26.3% of those in need being program eligible for transportation programs and 5.7% being program eligible for care coordination programs.

We estimated that the cost of providing evidence-based interventions for these four domains averaged $60 per member per month (PMPM) at a population level, including approximately $5 for screen-ing and referral management in clinics. Importantly, only 45.8% of these costs ($27 PMPM) were financed by existing mechanisms using federal funds, while the remaining 54.2% ($33 PMPM) would require other funding sources to address those not eligible or eligible but not enrolled.

The proportion of social needs costs borne by existing federal payers varied by domain, with 61.6% for food insecurity costs, 45.6% for housing insecurity costs, 27.8% for transportation insecurity costs, and 6.4% for community-based care coordination costs. Subgroup analyses revealed that populations attributed to FQHCs required interventions costing a mean of $93 PMPM, compared to $77 PMPM for non-FQHC practices in high-poverty urban areas, $77 PMPM for non-FQHC rural practices in high-poverty rural areas, and $24 PMPM for non-FQHC practices in lower-poverty areas.

Notably, the intervention costs not borne by existing federal payers were highest among non-FQHCs in low-income rural areas ($45 PMPM), followed by non-FQHCs in low-income urban areas($40 PMPM), and FQHCs ($36 PMPM). These findings suggest that while disproportionate funding was available to populations seen at FQHCs, the populations seen at non-FQHC practices in high-poverty areas had larger funding gaps in terms of the intervention costs not covered by existing federal funding mechanisms.

Our study highlights the substantial new funding and resources that would be needed to implement a comprehensive approach to addressing social needs identified in primary care settings, much of which falls outside of existing federal financing mechanisms. The findings underscore the “bridge to nowhere” problem, where inadequate community-based capacity limits the ability to address social need seven when they are identified through healthcare-based screening.

Importantly, our analysis found that screening and referral management in primary care was a relatively small expenditure compared to the cost of interventions to address social needs. This suggests that while healthcare organizations may face a lower financial barrier to screening and referring patients to social services, community-based or public-sector agencies may have greater barriers to financing the necessary interventions upon referral.

So, what is the solution? Addressing social determinants of health requires a more comprehensive and collaborative approach. It means bringing together healthcare providers, social service agencies, community-based organizations, and policymakers to develop coordinated, long-term strategies for improving health and well-being.

This approach requires a significant shift in the way we think about healthcare. Rather than viewing it as a standalone system, we need to recognize that health is deeply interconnected with other social and economic factors. We need to break down the silos between healthcare and other sectors, and work together to create more integrated and effective systems of care.

To overcome the wrong pocket problem, focusing on state or government programs to address food insecurity may be effective. By investing in these programs, the government can take a more holistic view and recognize the benefits of improved population health across multiple sectors. This approach aligns the incentives and ensures that the entity making the investment also realizes the value generated by the intervention.

Rather than trying to reinvent the wheel, the healthcare system needs to partner with existing organizations that have expertise in addressing social needs. There are already highly effective and efficient systems in place for providing food assistance, affordable housing, and transportation services. Medicaid agencies should focus on building relationships with these organizations and ensuring that they have the capacity to serve Medicaid beneficiaries. This may involve providing funding for staff, training, and other resources to help these organizations scale up their services.

However, simply increasing funding may not be enough. Medicaid agencies also need to work closely with community-based organizations to understand their unique needs and challenges, and to develop targeted strategies for building capacity. One promising example of this approach is the work of Bon Secours Health System in Baltimore, Maryland. As described in a recent Health Affairs article by my former post-doctoral fellow Dr. Emmanuel Drabo, Bon Secours made a direct investment in affordable housing as part of its community health improvement strategy.

The health system recognized that many of its patients were struggling with housing instability, which was contributing to poor health outcomes and high healthcare costs. To address this issue, Bon Secours partnered with a local affordable housing developer to build apartments in West Baltimore. The health system provided funding for the project, which helped leverage additional public and private investment. The apartments were designed to be affordable for low-income families, with rents set at a narrow fraction of household income.

The results of Bon Secours’ investment in affordable housing were indeed impressive. Residents of the apartment complex had improvements in their health and well-being. These positive health outcomes likely stemmed from several factors. By providing stable, affordable housing, Bon Secours helped alleviate the chronic stress and instability that often accompanies housing insecurity. This, in turn, may have led to improvements in mental health and overall well-being. The housing also likely reduced exposure to environmental hazards like mold, pests, and extreme temperatures, which can exacerbate respiratory conditions like asthma.

Importantly, Bon Secours’ investment also generated a positive return for the health system itself. By reducing emergency department visits and hospitalizations among residents, the program helped decrease the burden of uncompensated care. Emergency department overuse by uninsured or underinsured patients is a significant cost driver for many hospitals, particularly those serving low-income communities. Patients experiencing homelessness or housing instability often turn to the emergency department for issues that could be better addressed in primary care, or for basic needs like shelter and food.

The Bon Secours experience was an application of research on “Housing First” strategies for addressing homelessness and its associated health impacts. Housing First is an approach that prioritizes providing permanent, affordable housing to individuals experiencing homelessness, without preconditions such as sobriety or participation in treatment programs. The idea is that having a stable place to live makes it easier for people to address other challenges, such as substance use disorders or mental health issues. By contrast, making housing conditional on sobriety or treatment participation can actually be counterproductive, as it creates additional barriers and stress for people who are already struggling.

Systematic reviews of Housing First programs have found that they successfully reduced homelessness and improved housing stability compared to other approaches. Importantly, the Housing First participants experienced fewer emergency department visits and hospitalizations than control groups who were offered housing only if they first quit alcohol or drug use. The evidence suggests that providing stable housing can help reduce the stress and instability associated with homelessness, which in turn may lead to better health outcomes and lower healthcare utilization.

The cost savings associated with Housing First programs has been notable. Across all U.S. studies, the median economic benefit of Housing First was $18,247 per person per year, with a benefit-to-cost ratio of 1.8:1 (e.g., for every $1 spent on Housing First, there was an economic benefit of $1.8). Among the highest-quality studies, the median benefit was even higher at $33,637 per person per year. The cost savings stemmed from a combination of factors. By reducing emergency department visits and hospitalizations, Housing First programs help lower healthcare costs for high-need individuals. Stable housing may also help people avoid interactions with the criminal justice system and reduce their need for welfare assistance.

By providing housing first and then offering voluntary supportive services, Housing First programs aim to create a foundation of stability from which participants can begin to address other challenges in their lives. This approach recognizes that recovery is a process and that having a safe, stable place to live is essential for making progress.

While supporting effective housing, food and transportation infrastructure is critical to address social determinants of health, the experience of people entering into these programs is critical to generate participation. Many people receiving coverage for healthcare via Medicaid have had negative healthcare and social services experiences. This means investing in programs that build trust with patients is often critical to actually gaining participation from patients.

A promising model for this approach is the training and deployment of community health workers (CHWs). CHWs are trusted members of the community who can serve as a bridge between patients and the healthcare system. They can help patients navigate complex social service systems, provide emotional support, and advocate for their needs.

A recent series of studies provide compelling evidence for the effectiveness of CHW interventions in reducing healthcare utilization and costs. The studies using a protocol called IMPaCT (Individualized Management for Patient-Centered Targets), involved structured and standardized CHW intervention that was tested in three randomized controlled trials across multiple clinical settings.

The results were striking. Patients who received the IMPaCT intervention had significantly fewer hospitalizations, shorter hospital stays, and fewer repeat hospitalizations compared to those who received usual care. These effects persisted for up to a year after the intervention ended, suggesting that the benefits were long-lasting. Importantly, the program was cost-saving to Medicaid health plans within a single fiscal year: due to higher use of primary care and lower hospitalizations and illness among patients receiving CHW support, every dollar invested in the intervention returned $2.47 in healthcare savings.

The IMPaCT studies highlight several key features of effective CHW interventions. First, the intervention was highly structured and standardized, with clear protocols for hiring, training, and supervising CHWs. This ensured that patients received consistent, high-quality care across different clinical settings.

Second, the intervention was tailored to the unique needs and preferences of each patient. CHWs worked closely with patients to develop personalized action plans and provide hands-on support for achieving their goals. This patient-centered approach was critical for building trust and engagement.

Third, the intervention was integrated with the healthcare system, with CHWs working closely with primary care providers and other members of the care team. This allowed for seamless coordination of care and ensured that patients received the full range of services they needed.

One of the key benefits of CHW interventions is their ability to connect with patients across race and class divides. CHWs often come from the same communities as the patients they serve, which can help build trust and rapport. They may also have shared life experiences, such as a history of incarceration or substance use, which can make them more relatable and approachable than traditional healthcare providers.

Despite the clear benefits of CHW interventions, many Medicaid agencies have been slow to adopt this approach. One reason is that CHWs are often not reimbursed by Medicaid, which can make it difficult for healthcare organizations to sustain these programs over the longterm. Additionally, many CHWs are volunteers or are paid low wages, which can limit their ability to provide consistent, high-quality care.

To address these challenges, Medicaid agencies need to prioritize funding for CHW programs and ensure that these workers are fairly compensated for their important role in the healthcare system. They also need to provide training and support for CHWs to help them build the skills and knowledge needed to effectively serve Medicaid beneficiaries.