The trial, conducted between July 2023 and August 2023, enrolled adults receiving active cancer treatment who screened positive for food insecurity on the two-item Hunger Vital Sign, a brief validated screening instrument that asks patients whether they have worried about food running out and whether food actually did run out before they had money to buy more. Of the 30 people found eligible, all 30 participated, an extraordinary 100% uptake rate for a social-needs intervention in oncology. The vast majority of the cohort, 27 of the 30 participants or 90%, identified as Latino/a/e/x, 24 of 30 or 80% preferred to communicate in Spanish, and 16 of 30 or 53% were living with stage 3 or stage 4 disease, meaning more than half the group faced both advanced cancer and unreliable access to adequate food simultaneously.

Every consenting participant received $40 monthly gift cards for six months, redeemable at preferred, local retailers of their own choosing, layered on top of usual care that included trained community health workers who screened for health-related social needs and helped patients address them. The design choices were deliberate and rooted in prior qualitative work by the same team. The researchers reasoned that food assistance interventions are most likely to succeed when they are culturally tailored, minimize stigma, and promote self-efficacy, the individual’s sense of control over their own nutrition. Traditional food pantry models, while valuable, have been criticized in the literature for unreliability, inconvenient hours, limited fresh food, and the social stigma that deters some patients from using them, and enrollment in government nutrition programs such as SNAP has been associated with persistent food insecurity among cancer patients even after enrollment. Allowing patients to select their own groceries at familiar neighborhood stores, in the company of their own families and according to their own cultural food traditions, was intended to sidestep each of these barriers at once.

The analytical approach paired each patient with themselves, comparing outcomes before and after the intervention rather than against a separate control group. Food insecurity, the primary outcome, was assessed at baseline enrollment and again at six-month follow-up using validated survey instruments, with McNemar’s tests applied to compare paired binary outcomes, a statistical procedure appropriate for matched before-and-after dichotomous data. Continuous measures of health-related quality of life and financial toxicity were compared with paired t-tests, and adherence to cancer treatment was verified independently through medical chart review rather than relying on patient report alone. Patient-reported outcomes were captured using validated instruments including the Functional Assessment of Cancer Therapy-General, or FACT-G, a widely used 27-item scale of physical, emotional, social, and functional well-being in cancer, and the COmprehensive Score for Financial Toxicity, known as COST, which quantifies the financial distress patients experience from the cost of their illness and its treatment.

The results were striking. Food insecurity fell from 100% of participants at baseline to 16.7% at six-month follow-up, with a 95% confidence interval of 0.48 to 0.84 and a p-value below 0.001, indicating that the reduction was extremely unlikely to be due to chance. Health-related quality of life also improved significantly over the intervention period. The mean FACT-G score rose from 71.9, with a standard deviation of 16.1, to 78.9, with a standard deviation of 10.2, a change that reached statistical significance at p = 0.025. Perhaps most consequential from a clinical standpoint, 100% of participants adhered to their recommended cancer treatment over the six months, an outcome the researchers assessed through chart review. In a population where food insecurity is known to force brutal trade-offs between eating and affording medical care, the finding that no patient fell behind on treatment suggests the intervention may protect the most fragile link in the care chain.

The scientific rationale for why food access should matter this much in oncology is well established. Prior studies cited by the team have shown that food insecurity among cancer patients is associated with forgone medical care, cost-related medication underuse, and worse health-related quality of life. The physiologic stress of cancer and its treatments raises caloric and protein requirements at precisely the moment when finances are collapsing under the weight of out-of-pocket expenses, reduced work capacity, and caregiving burdens. Malnutrition in cancer patients is linked to treatment interruptions, dose reductions, increased toxicity, and poorer survival, and the term financial toxicity, coined to describe the adverse effects of cancer cost burdens, captures the mechanistic overlap between a patient’s wallet and their treatment plan. When grocery money competes with copays and transportation to infusion appointments, something has to give, and the “treat or eat” dilemma described in the health services literature becomes a daily reality.

What distinguishes this study from earlier food insecurity trials in oncology is both its design philosophy and its population. A randomized controlled trial published in the Journal of Clinical Oncology in 2022 tested food insecurity interventions in a broader patient population and demonstrated feasibility of food assistance in cancer care, but uptake and retention in real-world community settings have remained challenging, particularly among racial and ethnic minoritized patients who report discomfort sharing social needs information with healthcare providers. The Stanford-led team built their intervention on the foundation of their own earlier qualitative work, in which low-income Latino, Hispanic, and immigrant cancer patients described precisely the barriers that conventional programs impose. By embedding screening in a community oncology clinic in Monterey, California, delivering assistance through universally acceptable gift cards rather than identity-revealing vouchers or pantry visits, and maintaining the wraparound support of bilingual community health workers, the program achieved something rare: every eligible patient said yes, and every patient stayed.

The authors are careful about the limits of what can be concluded. The study was small, with 30 participants, lacked a concurrent control group, and cannot definitively attribute the observed improvements to the vouchers alone, since participants simultaneously received community health worker support as part of usual care. Regression to the mean, seasonal effects, and the natural course of treatment cannot be ruled out in a single-arm pre-post design. Self-reported quality of life improvements of roughly seven points on the FACT-G are meaningful but modest, and the trial’s exploratory outcomes, including financial toxicity, did not headline the results. Data could not be shared outside the research team due to institutional review board protocol restrictions, which limits reanalysis by independent investigators. Still, the magnitude of the food insecurity decline, from universal at enrollment to one in six at follow-up, is difficult to dismiss as noise, and the perfect treatment adherence rate in a group with advanced disease speaks to the practical value of removing a fundamental barrier to care.

The implications reach well beyond the Monterey clinic. Food insecurity in the United States affects tens of millions of households, and cancer patients are disproportionately vulnerable, yet systematic screening for food insecurity within oncology practices remains far from routine. The new study adds to a growing body of evidence that food-based interventions delivered through healthcare settings, including medically tailored meals, produce pharmacies, and voucher programs such as the San Francisco-based Vouchers 4 Veggies model that informed this work, can shift health outcomes, not merely reported satisfaction. It also strengthens the case for culturally specific tailoring: a program designed with, rather than merely for, Latino/a/e/x patients succeeded where one-size-fits-all approaches have often faltered. As health systems and policymakers weigh expansion of food-as-medicine programs under Medicaid waivers and value-based payment models, this trial offers a concrete, inexpensive template, forty dollars a month per patient, a screening question pair that takes under a minute, and a workforce of community health workers, that could be replicated in other community oncology settings serving minoritized populations.

The study was supported by grant funding to the senior author, Manali I. Patel of Stanford University, from the Stanford Cancer Institute and the American Cancer Society, and the authors declare no competing interests. The full paper, “Food for cancer health: the impact of a culturally tailored food voucher intervention on food insecurity and quality of life among Latino/a/e/x patients with cancer,” by Maria T. Hanna, Emily H. Wood, Lesly Lopez Guzman, and colleagues, was published in Supportive Care in Cancer, Volume 34, article number 922, on 2 September 2026. For patients facing cancer on top of an empty pantry, the message of the study is as simple as the intervention itself: reliable access to familiar, culturally appropriate food may be one of the most underrated tools in cancer care.