[Original post: Kathleen Noonan. Disappointing Randomized Controlled Trial Results Show A Way Forward On Complex Care In Camden And Beyond , Health Affairs Blog, January 9, 2020. https://www.healthaffairs.org/do/10.1377/hblog20200102.864819/full/: Copyright ©2020 Health Affairs by Project HOPE – The People-to-People Health Foundation, Inc.]

Two years ago, when I joined the Camden Coalition of Healthcare Providers, I met with a client we’d been serving for nearly 18 months. When we first met Sarah, she was in the early days of recovery from heroin addiction, just starting to self-manage her medical conditions. She was living with her infant in a shelter. Following intensive work with the Camden Coalition and our partners, we were able to support Sarah’s successful transition to stable housing—and along with it, reliable food, peer support, and behavioral health care. 

Today, she’s thriving. 

Across the country, leaders, clinicians, think tanks, and payers are racing to find better ways to treat people like Sarah. These patients are what many call the “5 percent”—the small portion of the population that accounts for more than half of all health care spending.  

The Camden Coalition has been at the forefront of this work since our founding in 2002. We’ve made tremendous progress and seen significant impact on many patients’ health and well-being. Yet, new evidence from a recent randomized controlled trial (RCT) shows that there is much more work to be done if we truly want to improve care, reduce costs, and advance overall health. 

The Camden Coalition Model 

At the Camden Coalition, we identify individuals who return to the hospital again and again, receive care, but never get well. Like Sarah, our patients have multiple chronic conditions exacerbated by and layered atop lifetime burdens of housing instability, food insecurity, low literacy, and social vulnerability. Add to that societal challenges such as poverty, racism, and disinvestment, and these patients are left with few resources to self-advocate and heal. 

Our model originated after examining claims data from three local hospitals in the Camden, New Jersey, region—one of the nation’s poorest cities. Through our analysis, we discovered that just 1 percent of all patients accounted for 30 percent of the area’s health care spending. Indeed, many of the patients we identified had more than 100 visits to the emergency department annually. We suspected that their outlier status as high utilizers would mean that even modest additional intervention could yield measurable return; we wanted to know over what period of time, how sustainable the changes would be, and in what areas could we have the most impact. 

Alongside leaders and clinicians from across Camden, we made underserved populations our focus—to increase their use of preventative care and social services and to reduce health care costs. Together, we built a regional health information exchange, linking real-time data from local hospitals, primary care providers, medical providers in the county jail, and federally qualified health centers; drove practice transformation in local primary care offices so Medicaid patients could be seen more quickly upon discharge from the hospital; passed state legislation to create a regional Medicaid accountable care organization model; engaged houses of worship in community-organizing efforts; launched a housing first program for chronically homeless patients; partnered with the criminal justice sector to ensure better continuity of care for people entering and released from jail; and built a National Center for Complex Health and Social Needs to engage stakeholders and identify other promising and evidence-based complex care models from around the country. 

A Randomized Control Trial 

Nearly five years ago, we welcomed the rigor of an RCT into assessing our model for managing care for this population. While we’d always incorporated evaluation of our work into the work itself to iterate and improve on our practices, we felt an RCT could add to the growing body of research on effective interventions for the most underserved. 

Although there is no doubt our efforts have led to immense transformation in the local community and among the populations we served, the recently released results from our RCT—an undertaking with researchers affiliated with J-PAL North America, a research center based at the Massachusetts Institute of Technology—reveal that the data we were originally tracking didn’t bear this out. Even with increased use of services such as Supplemental Nutrition Assistance Program benefits and primary care engagement, patients in the intervention groups were readmitted to the hospital at similar rates to those in the control group. 

To be sure, we were disappointed. We’re also now motivated by two important lessons from the RCT that have powerful implications for the field of complex care. 

First, we learned that all of us focused on health need to widen the aperture on our metrics for success. 

While the RCT used the 180-day hospital readmission rates as a proxy for improved health, systems-level interventions cannot be effectively appraised using a single quantitative metric. That approach may work well in biomedical research and scientific discovery—but our work is the very human, iterative, and long-term process of building personal relationships and supporting a community as it recreates its ecosystem of care. Telling the full story of an initiative’s impact on people’s lives requires a broad array of both quantitative and qualitative measures—which many health systems are beginning to adopt. 

The second lesson: Health services interventions in complex care are only as effective as the resources available in the communities in which those interventions are applied. 

Our original hypothesis was that targeted, short-term care management, care coordination, and hospital care transitions would be sufficient to reduce costs. We focused on dismantling silos between care systems as we helped patients meet the goals they’d set for themselves. We met patients at the bedside, accompanied them to their primary care appointments, navigated them to specialty and behavioral health services, and helped them apply for benefits. 

While the theoretical approach was sound, the practical realities were challenging. Our teams struggled to find community-based services that could meet the needs of patients with high rates of mental health diagnoses, addiction, homelessness, and incarceration. Most of our patients have lifetimes of trauma in addition to their chronic health conditions, making both physical and emotional healing an essential element of their care—a process that takes years, not months. Improvement in outcomes therefore stalled because primary care, behavioral health, addiction treatment, and housing services weren’t built to provide services to people with such extreme experiences and needs. What’s more, the systems themselves weren’t built for collaboration. 

We know that the mechanisms by which patients in the 5 percent drive up health care spending are not limited to health care needs alone. The amount spent on their care is high because of root-cause social needs that influence health. And until we address the issues outside the clinic walls by building capacity and forming coalitions among diverse partners, costs will continue to rise as health outcomes decline. 

Going Forward 

We’ve learned a lot since 2014, when the RCT began. Chief among the takeaways is that no single intervention or institution can independently provide the spectrum of support required to sustainably, meaningfully improve the likelihood of better health and well-being for the 5 percent. As we seek to augment care in the clinic with services outside the clinic walls—such as stable housing, legal resources, access to primary care, and medication-assisted treatment for substance use disorders, we must continue to bring together all community resources to focus on social determinants of health. 

At times, we’ll fall short on the path to progress. 

In Complications: A Surgeon’s Notes on an Imperfect Science, Atul Gawande notes, “We look for medicine to be an orderly field of knowledge and procedure. But it is not. It is an imperfect science, an enterprise of constantly changing knowledge, uncertain information, fallible individuals, and at the same time lives on the line. There is science in what we do, yes, but…the gap between what we know and what we aim for persists. And this gap complicates everything we do.” 

Like medicine itself, the work of caring for our nation’s most vulnerable people is neither simple to do nor easy to measure. Sometimes, we can’t make change. And sometimes, the successes just take longer to observe. But let’s welcome the setbacks, iterations, and partnerships that clear the path for better health for all.