An Indigenous Food Is Medicine Intervention: The MUTTON-HF Randomized Clinical Trial

Culturally tailored traditional meals reduced acute care use among high-risk Navajo patients with heart failure.
*Pragmatic randomized clinical trial; Level 2 (OCEBM).

Citation

Eberly LA, George C, Sandman S, et al. An Indigenous Food Is Medicine Intervention: The MUTTON-HF Randomized Clinical Trial. JAMA Internal Medicine. Published online July 27, 2026. doi:10.1001/jamainternmed.2026.2879

Background

Food insecurity worsens heart failure outcomes and is common in many Indigenous communities. This study tested whether medically tailored meals using traditional Navajo foods could improve outcomes in a rural tribal setting.

Patients

206 adults with heart failure receiving care at 2 Indian Health Service sites in rural Navajo Nation, with a hospitalization or emergency department visit in the prior year. Exclusions included hospice care, skilled nursing or rehabilitation facility residence, or inability to consent.

Intervention

Eight weeks of 14 frozen, low-salt, culturally and medically tailored traditional meals weekly, with delivery or pickup support and needed appliances.

Control

Usual dietary advice; control patients were offered meals after outcome assessment.

Outcome

Primary: all-cause hospitalization or emergency department visit within 90 days. Secondary: separate acute care outcomes, food security, weight, blood pressure, and Kansas City Cardiomyopathy Questionnaire score (0-100; higher is better).

Follow-up Period

90 days for acute care outcomes; 60 days for surveys and clinical measures.

Results

NNT: number needed to treat. CI: confidence interval. OCEBM: Oxford Centre for Evidence-Based Medicine.

Analyses were intention-to-treat. The primary composite outcome was mainly driven by fewer hospitalizations. Usual dietary advice was the available current treatment. A 5-point Kansas City Cardiomyopathy Questionnaire change is often considered meaningful; total-score improvement was near this threshold but statistically uncertain.

Limitations

Open-label design, small sample, short follow-up, and selected high-risk rural Navajo population limit generalizability. Missing survey and laboratory data may bias secondary outcomes. The trial cannot separate the effect of traditional cultural tailoring from healthy meal provision alone.

Funding

American Heart Association, U.S. government grants; meal producer had payments.

Clinical Application

Consider culturally tailored meal support for high-risk, food-insecure heart failure patients; evidence is promising but setting-specific, not yet universal best practice.