Summary
What do they do? Rebuilding Together San Francisco describes home repairs, mobility and safety modifications. The proposed course evaluated here adds skilled OT-equivalent assessment, tailored installation/removal, behavior training and a six-month booster for one high-risk adult aged65+ per enrolled household. Current RTSF delivery of that exact course is unverified. The local affiliate—not national Rebuilding Together—is the organization assessed; its official giving instructions list EIN94-3107808. More
Why this approach interests us
- Removing actionable hazards may prevent falls when tailored installation is paired with assessment, training and follow-up.
Our main reservations
- A larger related trial did not show health gains. RTSF has not demonstrated the exact trial protocol or a causal local effect.
What do you get for your dollar?
Our provisional best estimate is $5M per better life—10 incremental QALYs for an additional complete home-modification course. RTSF provides real SF repair services; we model a proposed skilled fall-prevention expansion, not its entire portfolio. The positive health estimate relies on a secondary trial result and a judgmental QALY bridge despite null primary outcomes. Inspect the model and sources →
- cost per better life
- $5M — per 10 QALYs; conditional all-direct-input donor scenario
- positive core scenarios
- $267K–$1.5B — Within one year; joint judgments, not confidence bounds
- central course cost
- $1,000 — All direct inputs valued; participant time shown separately
- funding room
- Unknown — 100 people is an existing funded target—not an unfunded queue
1. What do they do?
Rebuilding Together San Francisco describes home repairs, mobility and safety modifications. The proposed course evaluated here adds skilled OT-equivalent assessment, tailored installation/removal, behavior training and a six-month booster for one high-risk adult aged65+ per enrolled household. Current RTSF delivery of that exact course is unverified. The local affiliate—not national Rebuilding Together—is the organization assessed; its official giving instructions list EIN94-3107808.
Confirm a new eligible course
Prior falls or fear of falling plus actionable hazards; reconcile usual care, existing funding and landlord obligations.
Assess and install
Skilled assessment and tailored modifications, with permissions, safe installation and behavior training.
Follow up and measure
Track all enrolled households, failed visits, falls, injuries and preference-based utility through the year.
Scope of this review. One high-risk adult per household. Fixture lifespan is not health-effect duration; do not automatically add caregiver, household or mortality benefits.
2. Monitoring and information sharing
310 high-risk community-dwelling older adults in St Louis;12 months
Randomized home-hazard intervention trial. Rounded all-fall rates were2.3 versus1.5 per person-year. This secondary result favored intervention; primary time-to-first-fall and HRQoL results were not significant.
Our assessment. The0.8 difference is not an adjusted absolute effect or serious-injury rate. Every fall is not a fracture; the QALY mapping averages noninjurious events and truncated consequences.
1,331 adults aged65+ in the UK
Larger randomized occupational-therapy home assessment trial with economic evaluation. No benefit established; the economic analysis estimated −0.0042 QALYs and £18.78 additional payer cost over12 months.
Our assessment. Protocol differs from HARP but supplies important counterevidence. Net UK payer cost is not an SF delivery quote; no medical savings are credited.
San Francisco repair recipients
Provider service, application and giving disclosures. Actual local modifications and existing funded delivery are documented; exact trial fidelity and marginal availability are not.
Our assessment. Funded targets are baseline services, not additional outcomes attributable to a new gift.
3. Qualitative assessment
Removing actionable hazards may prevent falls when tailored installation is paired with assessment, training and follow-up.
Key reservations
- OT-equivalent fidelity not verified
- Permission and contractor capacity
- Public/sponsor substitution
- Exact marginal course price and cohort
- The .01 central QALY loss per all-fall event is a judgment, not a calibrated injury-mix estimate; .05 favorable is also speculative. Repeated-event overlap could inflate the result.
- No evidence warrants treating donated labor as costless or automatically translating years of fixture durability into years of utility benefit.
Benefits not included in our estimate
- Count one adult per household; no automatic household multiplier
- No extra mortality, fear, fracture, housing, caregiver or institutionalization bonus
- Do not infer health durability from physical fixture lifespan
- No extra adherence multiplier without redefining relative engagement
- Do not assign fracture QALYs to every fall
- Identical replacement makes both benefit and intervention harm zero
- Nonpositive QALYs have no attractive negative price
- No double counting shared household modifications or overlapping exercise effects
4. What do you get for your dollar?
$5M per 10 QALYs
The central budget is $300 skilled assessment/booster/travel/records + $400 installation + $200 materials + $100 coordination. These are planning judgments; the skilled allowance may be too lean. We take 0.8 fewer all-falls per person-year from rounded HARP rates, retain half for local transfer, assign 0.01 QALY per prevented all-fall event, and assume half the delivery is additional. This produces 0.002 incremental QALY per course. The utility mapping is not measured by either trial.
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
DOLLARS PER 10 INCREMENTAL QALYS: 10 × $1,000 ÷ [0.5 × (0.8 × 0.5 × 1 × 0.01 − 0)]
$5M
Model inputs and assumptions
- Conditional donor cash, USD
- 1000 (range: 1000 / 800 / 3000). Equal to direct resources only in core assumption that donor finances all inputs. Actual marginal gift cash price remains unknown Analyst judgment.
- Direct delivery resources, USD
- 1000 (range: 1000 / 800 / 3000). Complete direct resources including paid or replacement-valued donated labor/materials; not verified cash price Analyst judgment.
- Participant/caregiver time, USD
- 100 (range: 100 / 100 / 100). See derivation and source limitations; not a local measured coefficient. Analyst judgment.
- Rounded external all-falls rate difference
- 0.8 (range: 0.8 / 0.8 / 0.8). See derivation and source limitations; not a local measured coefficient. Derived secondary trial rate.
- Local clinical transfer
- 0.5 (range: 0.5 / 1 / 0.1). Protocol/population/local engagement transport and conflicting trial evidence; not funding Analyst judgment.
- Effective years within the first year
- 1 (range: 1 / 1 / 0.5). Integrated retained exposure within core one-year horizon; pessimistic 0.5 is attenuation, not a six-month measurement Analyst judgment.
- Assumed QALYs lost per all-fall event
- 0.01 (range: 0.01 / 0.05 / 0.002). Averages harmless and injurious prevented events and consequences within episode, including late-fall timing/truncation; no lifetime tail. Repeat-event overlap may overstate linear benefit Analyst judgment.
- Additional delivery share
- 0.5 (range: 0.5 / 0.75 / 0.25). Share of net health created beyond already-funded delivery/substitutes; not public revenue share Analyst judgment.
- Shared intervention harm, QALY
- 0 (range: 0 / 0 / 0). Subtract before multiplying funding. OTIS magnitude in isolated harm stress is external illustration only Analyst judgment.
One-year scenarios—not verified courses available to donors
- central: $5M. 0.002 incremental QALY; expanded resource price including time $5.5M
- favorable_one_year: $266.67K. 0.030000000000000006 incremental QALY; expanded resource price including time $300K
- pessimistic_within_one_year: $1.5B. 0.000020000000000000005 incremental QALY; expanded resource price including time $1.55B
- null_clinical: No finite positive price. 0 incremental QALY; expanded resource price including time No finite positive price
- null_funding: No finite positive price. 0 incremental QALY; expanded resource price including time No finite positive price
Uncertainty. Null and negative outcomes remain possible. The model treats funding replacement as identical replacement of both benefits and intervention harms. It does not cover separate donor-driven disruption; that would require an additional harm term outside the funding factor.
Persistence and harmful-outcome stresses are separate
A favorable second year lowers the price to about $133K, but requires unobserved persistence with no extra maintenance cost. It is not in the core range or ranking. The OTIS-magnitude harm stress is an external illustration, not an RTSF finding.
SEPARATE TWO-YEAR STRESS: 10 × $800 ÷ (0.8 × 1 × 2 × 0.05 × 0.75)
≈ $133K per 10 QALYs
QALY conversion assumptions
- harm_otis_magnitude_only: No finite positive price. -0.0021 incremental QALY; not a core ranking input
- two_year_optimistic_persistence: $133.33K. 0.06000000000000001 incremental QALY; not a core ranking input
- null_funding_even_with_harm: No finite positive price. 0 incremental QALY; not a core ranking input
At central health assumptions, direct course cost must fall below $20 to beat $100K per 10 QALYs. No such complete offer is verified.
5. Funding and previous grants
No quoted additional course or verified unfunded cohort beyond existing repair commitments.
The application page describes funding to serve100 people by December31,2026—not100 unfunded or remaining places. A separate service page says Senior Home Repair intake is closed; this does not establish universal closure. Donor cash equals direct resources only under our no-subsidy assumption. Paid and donated labor remain resources; subtracting cofunding does not prove extra delivery. The expanded central price is $5.5M including $100 participant/caregiver time; this is still not a complete societal-cost model.
This review does not establish a verified marginal funding offer or a complete history of grants.
Opens the organization’s giving page. A general donation may not fund the specific activity modeled here; confirm allocation with the recipient.
6. Sources
- harp. JAMA Network Open. primary-randomized-trial. Published: 2021-08-31; retrieved: 2026-09-08.
- otisClinical. NIHR / NCBI Bookshelf. primary-randomized-trial-summary. Published: 2021-07; retrieved: 2026-09-08.
- otisEconomic. NIHR / NCBI Bookshelf. primary-trial-economic-analysis. Published: 2021-07; retrieved: 2026-09-08.
- rtsfServices. Rebuilding Together San Francisco. primary-provider-page. Published: Not stated; retrieved: 2026-09-08.
- rtsfApplications. Rebuilding Together San Francisco. primary-provider-page. Published: Not stated; retrieved: 2026-09-08.
- recipient. Rebuilding Together San Francisco. primary-provider-giving-instructions. Published: Not stated; retrieved: 2026-09-08.
Annual expenses: years and sources
Average annual expenses (three consecutive fiscal years): $1,400,838. Organization size is separate from the modeled cost-effectiveness of a donation.
Rebuilding Together San Francisco
Whole-organization Form 990 Part I line 18; calendar fiscal years ending December 31.
Latest 2025 return filed June 22, 2026. Full-text headers confirm 12-month periods. Whole San Francisco legal entity, not the national Rebuilding Together organization.