Project Open Hand medically tailored meals — charity research

Summary

What do they do? The modeled pathway begins with an adult recently discharged after a hospitalization who has known heart failure. A health system or community partner identifies eligibility, Project Open Hand prepares and delivers condition-matched meals, and the outcome is a heart-failure hospitalization within 90 days. The evidence intervention delivered one large meal per day for up to 10 weeks; the California budget anchor covered three meals per day for 12 weeks plus nutrition and case-management components. More

Why this approach interests us

After discharge, a condition-matched meal can make a heart-failure diet feasible when illness, low income, mobility, or cooking constraints would otherwise undermine adherence.

Our main reservations

The primary outcome was null, the favorable heart-failure result was exploratory, two meal providers were pooled, and the public cost anchor describes a different meal intensity.

What do you get for your dollar?

Project Open Hand helped deliver meals in a 1,977-person randomized trial. The intervention did not reduce the primary outcome of all-cause hospitalization: 27.1% with meals versus 24.6% with usual care. A narrower exploratory analysis among participants with heart failure found 7.9% versus 13.2% heart-failure hospitalization. We discount that subgroup result and estimate about $212,500 per additional 90-day heart-failure hospitalization averted. Bridging only that admission’s short-term quality-of-life loss yields about $132.8M per 10 QALYs—one better life. Both figures are conditional on the exploratory effect being real and transferable; a null remains plausible.

primary endpoint: No benefit shown — all-cause hospitalization was 27.1% with meals versus 24.6% usual care
cost per better life: ≈ $133M — per 10 QALYs, counting only short-term morbidity from an avoided heart-failure admission
positive-effect sensitivity: $32M–$7.8B — conditional on a positive heart-failure effect; a null effect has no finite impact price
funding room: Not published — the $100,000 gift is a scenario, not a current marginal offer

1. What do they do?

The modeled pathway begins with an adult recently discharged after a hospitalization who has known heart failure. A health system or community partner identifies eligibility, Project Open Hand prepares and delivers condition-matched meals, and the outcome is a heart-failure hospitalization within 90 days. The evidence intervention delivered one large meal per day for up to 10 weeks; the California budget anchor covered three meals per day for 12 weeks plus nutrition and case-management components.

Identify a high-risk discharge

A participating health system identifies an adult with known heart failure who is returning home.

Match meals to the condition

Meals prioritize heart-failure nutrition standards and account for co-occurring diabetes or kidney disease.

Deliver the course

Project Open Hand or a partner delivers meals; some programs add dietitian counseling or case management.

Measure the 90-day outcome

A recommendation-grade version would link every eligible participant to complete all-cause and heart-failure hospitalization data.

Scope of this review. The model covers one targeted post-discharge heart-failure pathway. It excludes Project Open Hand’s HIV services, groceries, senior meals, disability programs, produce prescriptions, community nutrition, and the organization-wide meal count. It also does not turn the exploratory mortality result into lives saved.

2. Monitoring and information sharing

1,977 recently hospitalized high-risk adults in Northern California; Project Open Hand and Ceres delivered meals

Pragmatic randomized trial. The primary all-cause 90-day hospitalization outcome was not improved: 27.1% with meals versus 24.6% with usual care, a 2.6-point adverse risk difference (95% CI -1.3 to 6.5) and adjusted hazard ratio 1.02 (95% CI 0.86-1.21). Among 641 participants with heart failure, an exploratory analysis found 7.9% versus 13.2% heart-failure hospitalization, a -5.3-point difference (95% CI -10.0 to -0.6) and adjusted hazard ratio 0.52 (95% CI 0.32-0.86).

Our assessment. This is the strongest evidence and directly involved Project Open Hand, but it combines two providers and five medical centers. The heart-failure result was exploratory and must not replace the null primary result. Our model discounts the 5.3-point subgroup difference and keeps a null effect plausible.

191 Project Open Hand clients living with HIV; intensive food-plus-nutrition support versus less-intensive Project Open Hand standard care

Pragmatic randomized trial. At six months, the intensive arm had lower odds of hospitalization, food insecurity, depressive symptoms, poor antiretroviral adherence, and unprotected sex. The trial found no difference in its primary viral-nonsuppression or health-related-quality-of-life outcomes.

Our assessment. This is organization-specific and locally relevant, but small, population-specific, and a comparison of two Project Open Hand service intensities rather than food support versus no service. It does not identify the heart-failure effect or cost used in our model.

Medi-Cal beneficiaries with congestive heart failure across eight California counties; Project Open Hand served San Francisco, Alameda, and San Mateo

State multi-provider pilot evaluation. The authorizing program set a $6 million four-year budget to serve 1,413 beneficiaries with three medically tailored meals per day for 12 weeks. That implies $4,246 per targeted participant before any distinction between budget, contracted price, actual spending, provider cost, or current price. The final evaluation pooled multiple providers and did not publish a Project Open Hand-only causal estimate.

Our assessment. We round the public program budget ratio to a $4,250 best-guess course cost and use a wide $3,000-$7,000 range. It is the closest program-specific public anchor, but the intensity differs from the randomized trial's one large meal per day for up to 10 weeks and is not current philanthropic funding room.

3. Qualitative assessment

After discharge, a condition-matched meal can make a heart-failure diet feasible when illness, low income, mobility, or cooking constraints would otherwise undermine adherence.

Key reservations

The largest randomized trial missed its primary all-cause hospitalization outcome; the favorable heart-failure result was exploratory.
The heart-failure subgroup result combined Project Open Hand and Ceres and does not identify a provider-specific effect.
The randomized intervention delivered one large meal per day for up to 10 weeks, while the state pilot cost anchor covered three meals per day for 12 weeks plus nutrition and case-management components.
Only 53.5% of randomized meal recipients received all 10 weeks, and those receiving more than five weeks differed from those receiving fewer.
The smaller HIV trial found no effect on its primary viral or health-related-quality-of-life outcomes and compared intensive with standard Project Open Hand service.
FY2024 scale combines several programs and three counties, so dividing organization expenses by meals or clients would be misleading.
Public financing, health-plan reimbursement, and philanthropy may all support medically tailored nutrition; the current private-gift counterfactual is not published.
No current organization-supplied marginal plan for a post-discharge heart-failure cohort was found in the reviewed public sources.

Benefits not included in our estimate

All-cause hospitalization reduction, because the randomized primary endpoint did not improve
Exploratory mortality reduction, because it was not the primary endpoint and was not explained by lower all-cause utilization
Food security, depression, adherence, sexual-health, and HIV outcomes from the smaller intensive-versus-standard-care trial
Meal enjoyment, caregiver relief, nutrition knowledge, social contact, and avoided food spending
Hospital cost savings, because price and value are different and the modeled outcome is an admission rather than dollars saved
Mortality, downstream readmissions, caregiver effects, medical savings, and non-hospital meal benefits from the separate acute-morbidity-only QALY bridge

4. What do you get for your dollar?

Primary result: no demonstrated all-cause benefit. Conditional heart-failure model: roughly $213,000 per admission averted.

California authorized $6 million to serve 1,413 medically tailored meal participants, implying $4,246 per targeted participant. We round that to a $4,250 best-guess course cost, with a $3,000–$7,000 range. We then discount the randomized exploratory 5.3-point heart-failure hospitalization difference to a 2.0-point best guess because the primary endpoint was null, multiple outcomes were tested, two providers were pooled, and the current Project Open Hand pathway is unpublished.

A better life is our comparison unit of 10 additional quality-adjusted life years (QALYs), potentially spread across people. These are uncertain estimates, not measured returns or verified donation offers.

How we calculate the estimate

CONDITIONAL COST PER ADDITIONAL HEART-FAILURE HOSPITALIZATION AVERTED: $4,250 ÷ 2%
= $212,500

Model inputs and assumptions

Modeled marginal cost per meal course: $4,250 (range: $3,000–$7,000). The California pilot authorized $6 million to serve 1,413 beneficiaries with three meals daily for 12 weeks, or $4,246 per targeted participant. MFI rounds to $4,250 and widens the range for provider mix, budget-versus-spend, intensity, inflation, delivery, dietitian, and shared-cost uncertainty. This is not a current Project Open Hand quote. low.
Randomized all-cause hospitalization difference: 2.6% (range: -1.3%–6.5%). At 90 days, all-cause hospitalization was 27.1% with meals and 24.6% with usual care. The positive sign is adverse. This primary endpoint is not converted into a favorable cost-effectiveness result. high for the trial; provider-pooled.
Exploratory randomized heart-failure hospitalization reduction: 5.3% (range: 0.6%–10%). Among 641 participants with known heart failure, 90-day heart-failure hospitalization was 7.9% with meals and 13.2% with usual care: a 5.3-point reduction with a 95% confidence interval from 0.6 to 10.0 points. This was exploratory and combined two meal providers. moderate for the subgroup; indirect for Project Open Hand.
MFI estimate of absolute 90-day heart-failure hospitalization reduction: 2% (range: 0.5%–5.3%). Judgmental transfer discounted below the exploratory 5.3-point trial result because the primary endpoint was null, multiple outcomes were tested, providers were pooled, course intensity differs, and current Project Open Hand targeting is unpublished. The low bound is a small positive effect, not a guarantee; a null effect remains plausible. very low.

What would $100,000 buy in the conditional model?

Observed exploratory effect: 1.77 heart-failure hospitalizations averted. 33.33 modeled courses · $56.6K each
MFI best guess: 0.47 heart-failure hospitalizations averted. 23.53 modeled courses · $212.5K each
Small positive effect: 0.07 heart-failure hospitalizations averted. 14.29 modeled courses · $1.4M each

Uncertainty. The numeric range is conditional on a positive heart-failure effect. The trial's primary all-cause hospitalization endpoint was null, the favorable subgroup result was exploratory, and Project Open Hand's provider-specific effect is unknown. If the subgroup finding does not replicate, Project Open Hand reaches a different population, or private money substitutes for reimbursed care, additional hospitalizations averted approach zero and cost per outcome has no finite upper bound.

About $133 million per 10 QALYs—conditional on the exploratory heart-failure effect.

The bridge deliberately counts only the short-term morbidity of an avoided acute heart-failure admission. NICE’s model implies 0.016 QALY per admission from a six-week utility dip. The sensitivity spans an acute-stay-only floor from ASCEND-HF to a 90-day recovery profile reported in the heart-failure utility review. It excludes mortality, downstream readmissions, caregiver effects, medical savings, and the direct quality-of-life effects of meals.

CONDITIONAL COST PER 10 QALYS (ONE BETTER LIFE): $212,500 ÷ 0.016 QALY × 10
= $132,812,500

QALY conversion assumptions

Conditional cost per heart-failure hospitalization averted: $212,500 (range: $56,604–$1,400,000). Imported from the Project Open Hand native-outcome model. The range is conditional on a positive effect and combines course-cost and causal-effect sensitivities. very low.
Short-term QALYs per heart-failure hospitalization averted: 0.016 QALY (range: 0.0018 QALY–0.0175 QALY). Best: NICE modeled a 6-week utility dip from a worsening-heart-failure admission as a 0.064 decrement over a 3-month cycle, equal to 0.016 QALY. Low: acute-stay-only floor using ASCEND-HF's 0.11 gap between 24-hour inpatient and discharge utility over a six-day median stay (0.11 x 6 / 365 = 0.0018 QALY). High: McMurray et al.'s reported decrements of 0.105 in days 0-30 and 0.054 in days 31-90, integrated over 90 days (0.0175 QALY). low; external heart-failure populations.
QALYs per one better life: 10 QALY (range: 10 QALY–10 QALY). Market for Impact's shared comparison denominator. It is a reporting convention, not a clinical threshold or a claim that ten QALYs accrue to one identifiable person. definition.

Optimistic positive effect: $32.3M. per 10 QALYs · $56,604 per admission averted ÷ 0.0175 QALY × 10
MFI best guess: $132.8M. $212,500 per admission averted ÷ 0.016 QALY × 10
Small positive effect: $7.8B. $1.4 million per admission averted ÷ 0.0018 QALY × 10

This bridge does not repair the causal uncertainty in the native model. The randomized primary all-cause hospitalization endpoint did not improve; the favorable heart-failure subgroup result was exploratory and provider-pooled. If Project Open Hand's true additional effect is zero, cost per 10 QALYs has no finite upper bound.

5. Funding and previous grants

Project Open Hand has not published a current heart-failure cohort, provider-only outcome, course price, payer-denial funnel, or marginal plan showing that a private gift creates additional courses rather than replacing reimbursed care.

The $100,000 scenario is illustrative. Project Open Hand has not published a current post-discharge heart-failure waitlist, payer-denial funnel, restricted balance, course price, provider capacity, reimbursement counterfactual, or dated plan showing that a private gift would create additional courses.

This review does not establish a verified marginal funding offer or a complete history of grants.

Donate

Opens the organization’s giving page. A general donation may not fund the specific activity modeled here; confirm allocation with the recipient.

6. Sources

Effect of Medically Tailored Meals on Clinical Outcomes in Recently Hospitalized High-Risk Adults. Medical Care. peer-reviewed randomized trial. Published: 2022; retrieved: 2026-08-31.
Medically Tailored Meals Pilot Program. California Department of Health Care Services. official program budget and target. Published: Program page; publication date not stated; retrieved: 2026-08-31.
Evaluation of the Medi-Cal Medically Tailored Meals Pilot Program. California Department of Health Care Services / Mathematica. government evaluation. Published: 2023-12-14; posted by DHCS June 2024; retrieved: 2026-08-31.
Acute heart failure: appendices — cost-effectiveness analysis. National Institute for Health and Care Excellence. Official health-technology-assessment model. Published: 2014; retrieved: 2026-08-31.
Health State Utilities of Patients with Heart Failure: A Systematic Literature Review. PharmacoEconomics. Peer-reviewed systematic literature review. Published: 2021; retrieved: 2026-08-31.
Model-based cost-effectiveness analysis of B-type natriuretic peptide-guided care in patients with heart failure. BMJ Open. Peer-reviewed cost-effectiveness model. Published: 2016; retrieved: 2026-08-31.
Hospitalization for Recently Diagnosed Versus Worsening Chronic Heart Failure: From the ASCEND-HF Trial. Journal of the American College of Cardiology. Peer-reviewed trial analysis. Published: 2017; retrieved: 2026-08-31.

Annual expenses: years and sources

Average annual expenses (three consecutive fiscal years): $18,690,387. Organization size is separate from the modeled cost-effectiveness of a donation.

Project Open Hand

Whole legal entity Form 990 Part IX total functional expenses; latest original reconstructed filings checked, fiscal-end year from printed reporting dates. Older fallback rows from IRS extracted API.

Latest original filings checked; ProPublica extracted API lags these original returns.