Improving Behavioral Health Access in Oncology

A cancer diagnosis changes everything, and not just physically. The psychological weight of confronting a life-threatening illness is well documented, yet the systems designed to support patients through that experience have consistently failed to reach those who need them most. A new study presented at the 2026 American Society of Clinical Oncology Annual Meeting offers a meaningful step toward changing that, and its implications stretch well beyond the data it presents.

The research, led by Nina Balanchivadze, MD, FACP, a hematologist/oncologist at Sarah Cannon Research Institute at Virginia Oncology Associates, and Kyle Lavin, MD, MPH, an assistant professor of psychiatry at the University of North Carolina at Chapel Hill, asked a deceptively simple question: what happens when you stop waiting for distressed cancer patients to seek behavioral health support and instead bring that support directly into the oncology workflow? The answer, it turns out, is that patients value this intervention, and they get better.

The starting point for understanding why this study matters is the sheer size of the gap it is trying to close. Between 60% and 80% of patients with a serious cancer diagnosis experience some level of psychological distress. Yet fewer than 10% of those patients ever access behavioral health care. That is not a minor inefficiency in the system. It is a near-total failure to deliver a category of care that oncology professionals widely agree patients need.

That failure is not evenly distributed. Medicaid patients in this study presented with statistically significantly higher levels of depression, as measured by the PHQ-9, and anxiety, as measured by the GAD-7, along with meaningfully lower quality of life scores on the FACT-G7. BIPOC patients showed directionally similar patterns, though the differences did not reach statistical significance. The people carrying the heaviest psychological burden are, predictably, the same people least likely to be reached by the services that exist. That is the inequity we must all address.

The numbers behind the study were substantial. Oncology providers referred 1400 patients into the program. Of those, 850 consented and scheduled an intake appointment, and 631 completed intake and began receiving behavioral health care. Black, Indigenous, and people of color (BIPOC) patients made up 43.5% of the overall study population. Further, among the 631 patients who received behavioral health treatment, 36.5% identified as Black and 18.7% were covered by Medicaid. These figures are notably higher than the representation these groups typically achieve in behavioral health services, and that alone is a finding worth pausing on.

The headline result was not just that patients improved, rather it was that the patients who started out worst off improved as much as, or more than, everyone else. Medicaid patients and BIPOC patients, despite presenting with higher levels of distress at intake, showed equal or directionally greater reductions in depression and anxiety scores and comparable gains in quality of life. The gap at entry did not translate into a gap in outcomes. Patient satisfaction was exceptionally high, with a Net Promoter Score of 88, and a majority of patients reported improved adherence to both oncology visits and non-chemotherapy medications.

These findings matter not just as clinical data points but as a rebuttal to a quiet assumption that sometimes shapes how underserved populations are discussed in health care. The assumption has been that harder-to-reach patients are harder to help. However, it seems that when access barriers are genuinely removed, those patients engage and benefit.

Community Oncology Is an Appropriate Setting

The decision to study this model in community oncology practices rather than large academic medical centers was deliberate, and it reflects something important about where the real-world gap actually lives. Academic centers tend to have more psychosocial oncology resources on hand and tend to serve patients who are, on average, higher in socioeconomic status and more equipped to navigate complex health systems. Community practices serve a far broader cross-section of the population, including patients at different income levels, different educational backgrounds, different distances from care, and different relationships with the medical system.

For community oncologists the logistical reality is concrete. Many patients live in remote areas where driving to a single appointment is already a hardship. Asking those patients to then seek out a separate behavioral health provider, at a separate location, on a separate schedule, is asking something that many of them simply cannot do. Virtual care embedded in the oncology workflow removes that ask entirely. The referral happens within a system the patient already trusts, and the care comes to them.

Building trust shapes whether patients follow through with treatment, whether they disclose symptoms, and whether they stay engaged with their care over time. The community setting, for all its resource constraints, often carries a level of relational trust that larger institutions struggle to replicate. Building behavioral health into that existing relationship is strategically sound in a way that standalone referrals are not.

The most pressing is the link between behavioral health integration and hard clinical outcomes. Symptom scores on the PHQ-9 and GAD-7 are meaningful, but they are not what moves payers or health systems to restructure how they deliver care. What the field needs now are prospective, controlled studies that connect integrated behavioral health to treatment completion rates, hospitalization rates, and ultimately survival. Balanchivadze points specifically to oral oncolytic adherence as an underexplored frontier. With the rapid expansion of oral targeted therapies and immunotherapies, nonadherence is a growing clinical problem, one that behavioral health support may be uniquely positioned to address.

A second open question involves clinical trial participation. BIPOC and low-income patients are historically underrepresented in oncology trials, and psychological distress and mistrust are known contributors to that gap. If integrated behavioral health reduces distress and strengthens the patient-provider relationship, it may also lower barriers to trial enrollment. That hypothesis has not been tested, but it deserves to be.

The field also faces a structural challenge that no single study can solve: reimbursement. Virtual collaborative care models operate in a payment environment that has not caught up with the evidence. Demonstrating that psychosocial support reduces downstream costs, through fewer emergency visits, better adherence, and improved treatment completion is essential to building the value-based care frameworks that would make these programs financially sustainable at scale.

Finally, there is the workforce question. Scaling models like this one requires behavioral health providers who understand oncology, a specialty-informed workforce that does not currently exist in sufficient numbers. Training pipelines, supervision structures, and genuine collaboration between oncology and psychiatry need investment that the field has not yet prioritized.

For decades, the oncology field has documented the psychological burden of cancer with considerable precision and addressed it with considerably less. This study points toward a model that is feasible, equitable, and effective. Behavioral health is not an optional add-on and should be treated as a core component of cancer care. The data suggest that when access barriers are removed and support is integrated where patients already are, the patients most likely to be left behind are also the most likely to benefit. That finding should shape how the field thinks about what comes next.

REFERENCE

Balanchivadze N, Danso MA, Lavin KN. Advancing health equity in oncology: virtual collaborative behavioral health engagement and outcomes among Medicaid-insured and BIPOC patients. J Clin Oncol. 2026;44(suppl 16):1552. doi:10.1200/JCO.2026.44.16_suppl.1552