GiveBetter x SF

Progress Foundation

Avenues two-week acute diversion episode

Research time: ~15 min on GPT-5.6 Sol Medium
  • Research — organization and evidence review.
  • Modeling — cost-effectiveness analysis.
  • Historical estimate for research done before time tracking.

Published: 7 September 2026.

Funding limitations

Summary

What do they do? Avenues is a 24-hour acute diversion residence for San Francisco adults, offering an alternative to psychiatric hospitalization with an average stay of about two weeks. This comparison assumes hospital care would otherwise be available. More

Why this approach interests us

  • A small, supportive residential setting can provide acute stabilization outside a hospital for suitable voluntary patients.

Our main reservations

  • The closest randomized studies do not establish superior health outcomes. Payer savings are not automatically donor savings or QALYs.

What do you get for your dollar?

Our subjective positive scenario is $391M per better life (10 QALYs). No added health benefit versus hospital care is at least as credible. This is not a finding that residential care is ineffective; it may deliver similar outcomes at lower public cost. Inspect the model →

positive central scenario
$391Mper 10 QALYs; null at least as credible
positive stress cases
$69.5M–$13.0BNot confidence bounds; harm possible
episode allowance
$15,000Assumed current two-week budget, not quote
funding room
UnverifiedExisting public contract and contingency must be reconciled

1. What do they do?

Avenues is a 24-hour acute diversion residence for San Francisco adults, offering an alternative to psychiatric hospitalization with an average stay of about two weeks. This comparison assumes hospital care would otherwise be available.

Assess clinical suitability

Residential care is not a substitute for hospital treatment for every crisis. Maintain transfer and safety backup.

Provide short-term stabilization

Model one two-week episode, keeping it distinct from Dore’s urgent-care clinic and longer-term residential rehabilitation.

Compare real alternatives

Estimate health and cost differences versus hospital treatment, not the whole improvement from crisis to recovery.

Scope of this review. Benefits of treatment itself occur in both arms. Do not count all recovery as caused by choosing residential care.

2. Monitoring and information sharing

Four SF ADUs, with Avenues as this report’s target

Local administrative budget and program description. FY2022–23 budget $6,431,386 and 584 estimated annual clients; approximately $11,013 per estimated client.

Our assessment. Old pooled budget/client ratio is not an observed episode cost, current price or marginal funding quote.

Voluntary female admissions; 19 assigned crisis house and 22 ward

Small UK randomized crisis-house pilot. No significant adjusted health/QALY difference; incomplete EQ-5D follow-up.

Our assessment. Do not treat lack of significance as equivalence. An inconsistent displayed adjusted QALY coefficient is not imported; preference-arm findings are not randomized evidence.

Adults accepting voluntary acute care

US randomized residential-versus-hospital evaluation. Lower government-payer episode costs, without significant symptom or community-day differences.

Our assessment. Not QALYs. Non-entry and hospital transfers limit extrapolation; hospital backup remains part of delivery.

3. Qualitative assessment

A small, supportive residential setting can provide acute stabilization outside a hospital for suitable voluntary patients.

Key reservations

  • No demonstrated added QALYs versus hospitalization.
  • A positive utility advantage is a subjective hypothesis.
  • Public funding displacement may erase donor additionality.
  • Safety, hospital transfers and actual marginal capacity need a current plan.

Benefits not included in our estimate

  • Government-payer hospital savings deducted from donor costs
  • Health benefits versus no treatment when hospital care is the comparator
  • Deaths prevented or benefits after the two-week episode
  • Dore urgent-care clinic, long-term residential care and housing

4. What do you get for your dollar?

Positive scenario: $391M per 10 QALYs; a credible null remains.

We assume $15,000 per episode, a small .02 utility advantage during fourteen days and 50% added funding capacity. This is a transparent hypothesis, not an estimated trial effect. No lasting health effect is added.

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 BETTER LIFE: 10 × $15,000 ÷ (0.02 × 14/365 years × 50%)
$391,071,429 per 10 QALYs

Model inputs and assumptions
Current episode allowance
15000 (range: 10000 / 15000 / 25000). Assumed $15,000, not a quote or inflation-adjusted audited cost. FY2022–23 pooled ADU budget $6,431,386 / 584 estimated annual clients is about $11,013, but estimated clients may differ from episodes, includes four programs and is not marginal Avenues pricing. Analyst assumption; external evidence does not establish benefit.
Short-term health advantage over hospital
0.02 (range: 0.05 / 0.02 / 0.002). Subjective .02, not a trial estimate. Closest small randomized QALY comparison found no significant adjusted health difference. Null is at least as credible as this positive scenario; satisfaction is not utility. Analyst assumption; external evidence does not establish benefit.
Benefit window
0.038356164383561646 (range: 0.038356164383561646 / 0.038356164383561646 / 0.038356164383561646). Fourteen days divided by 365. No persistent recovery, mortality or hospital-access gain is assumed. Analyst assumption; external evidence does not establish benefit.
Separate completion multiplier
1 (range: 1 / 1 / 1). One: utility already averages across episodes including transfer/failure. Hospital backup remains necessary. Analyst assumption; external evidence does not establish benefit.
New donor-funded capacity
0.5 (range: 0.75 / 0.5 / 0.25). 50% judgment. Public contract funding and contingency may replace donor spending; actual unfunded safe capacity unverified. Analyst assumption; external evidence does not establish benefit.

Illustrative $100,000: 6.67 gross episode-equivalents; 3.33 incremental after additionality

  • Favorable positive: $69.52M per 10 QALYs. 0.0144 QALYs per $100,000; capacity unverified
  • Subjective positive central: $391.07M per 10 QALYs. 0.00256 QALYs per $100,000; capacity unverified
  • Unfavorable positive: $13.04B per 10 QALYs. 0.0000767 QALYs per $100,000; capacity unverified

Uncertainty. The evidence-led comparator is no demonstrated incremental QALY gain versus hospital care. Null is at least as credible as the positive central scenario. No finite positive-benefit price at zero effect; negative utility preserves harm. These positive cases are not confidence limits.

Similar health at lower cost is different from buying extra QALYs.

With this donor-cost and hospital-comparator perspective, equivalent health yields no finite positive-benefit price. A public-payer allocation model could reach a different conclusion, but requires actual released resources and their alternative use.

FOURTEEN-DAY BOUNDARY: 10 × $15,000 ÷ (1.0 utility × 14/365 years)
$3.91M per 10 QALYs even at full utility advantage

QALY conversion assumptions

    This model-bound sensitivity caps the utility difference at 1.0; it is not a universal utility maximum, since some valued health states are below zero. It applies only to this cost and short pathway, not all Progress Foundation services. It is not a sub-$100K lead on the modeled pathway.

    5. Funding and previous grants

    No current marginal Avenues quote, verified funding gap or demonstrated health advantage over available hospital care.

    The 2023 SF analysis pools four ADUs and estimated annual clients, not current marginal Avenues episodes. Public funding and expansion contingency may displace philanthropy. Government-payer hospital savings are not deducted from the donor numerator.

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

    We have not verified a suitable donation route for this reviewed activity. Confirm the legal recipient and intended allocation before donating.

    6. Sources

    1. Programs — Avenues. Progress Foundation. Local program identity and average two-week stay. Published: Undated current page; retrieved: 2026-09-07.
    2. January 25, 2023 budget analysis. SF Budget and Legislative Analyst. Printed pp15–18 pooled ADU budget and estimated clients, not marginal price. Published: 2023-01-25; retrieved: 2026-09-07.
    3. Women's crisis houses compared with psychiatric wards. Howard et al.. Small randomized pilot; incomplete follow-up and an inconsistent displayed adjusted QALY coefficient, not imported. Published: 2010-08; retrieved: 2026-09-07.
    4. Cost and cost-effectiveness of hospital vs residential crisis care. Fenton et al.. Government-payer cost comparison; symptom/community-day outcomes, not QALYs. Published: 2002-04; retrieved: 2026-09-07.
    5. Randomized trial of general hospital and residential alternative care. Fenton et al.. Transfers and non-entry into assigned treatment constrain extrapolation. Published: 1998; retrieved: 2026-09-07.

    Annual expenses: years and sources

    Average annual expenses (three consecutive fiscal years): $31,196,547. Organization size is separate from the modeled cost-effectiveness of a donation.

    Progress Foundation

    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.