Summary
What do they do? DOPE is a National Harm Reduction Coalition program. This review models a hypothetical additional year of tenant outreach, training, refills and coordination in supportive housing. It excludes electronic detection systems and the wider national organization. More
Why this approach interests us
- Trusted residents may shorten the gap between an overdose and effective help, especially when formal supplies are available but not reached in time.
Our main reservations
- A staffed site-year is not cheap relative to its baseline fatal-event count. Current prevention is already substantial, and the additional mortality reduction is unknown.
What do you get for your dollar?
Our best guess is $2,425,273 per better life: 10 incremental QALYs for a hypothetical added tenant-responder site-year. We would investigate a documented high-risk, under-covered site before prioritizing this over our current leads. The model is not a measured DOPE result or a verified funding offer. Inspect the versioned model →
- our best guess
- $2,425,273 — per 10 QALYs; subjective additional site-year model
- positive scenarios
- $16.9 K–$689 M — Joint assumptions, not a confidence interval; null/harm possible
- evidence
- Very low for price — Local implementation evidence, no measured causal mortality effect
- funding room
- Unverified — General national donations are not a local expansion quote
1. What do they do?
DOPE is a National Harm Reduction Coalition program. This review models a hypothetical additional year of tenant outreach, training, refills and coordination in supportive housing. It excludes electronic detection systems and the wider national organization.
Identify an actual coverage gap
Choose occupied resident-person-time and shifts or networks not already served by contracts, staff, peers or another provider.
Support tenant responders
Pay and train peers, replenish supplies, coordinate with building staff and provide supervision and debriefing. Four specialists do not imply continuous coverage.
Count only added health
Estimate fewer otherwise-fatal events under existing usual care, not the number of trainings, kits or reported reversals.
Scope of this review. The pilot studied 50- and 160-unit buildings and paid trained specialists $50 weekly in gift cards. Those historical implementation details do not establish today’s cost, occupancy or efficacy. Our exposure is 100 occupied resident-person-years, not 100 distinct people or licensed beds.
2. Monitoring and information sharing
Two supportive SRO buildings; 35 observation days and interviews with staff and trained tenants
Qualitative local evaluation. Improved access and coordination were described, alongside turnover, overnight constraints and responder burden.
Our assessment. No controlled mortality reduction or incremental-response rate. Existing responders and front-desk supplies were part of the baseline.
San Francisco SRO residents, 2010–2017
Historical local mortality study. Table 1 reports 278.7 all-overdose and 157.8 opioid-related deaths per 100,000 annually, using SRO units as a population proxy.
Our assessment. Our 0.3% all-overdose risk and 80% responsive share are current-site judgments, not values measured by that study. Do not add substance-specific categories, which overlap.
HSH-funded permanent supportive housing
Public implementation reporting. The January 2025 SFDPH update reports 24/7 naloxone stations at all 156 sites.
Our assessment. Supply availability is not timely access or rescue, but it rules out assuming an untreated baseline. Actual current site coverage must be checked.
NHRC Naloxone Distribution program, contract 1000032403
Public contract inventory and objectives. April 2026 inventory lists$439,086; FY 25–26 targets include 50 training events and two new community sites.
Our assessment. These are a program funding line and targets—not paid marginal cost, achieved outcomes, or open funding room. The objectives are not specific proof of an unfunded SRO tranche.
3. Qualitative assessment
Trusted residents may shorten the gap between an overdose and effective help, especially when formal supplies are available but not reached in time.
Key reservations
- DOPE historically supplies partner networks including SFAF. Separate contracts do not establish independent health gains. Do not add full DOPE and SFAF rescue benefits for the same event/person. The model is an alternative site-level hypothesis, not an additive portfolio tranche.
- Current site mortality and incremental response are the decisive missing observations. A high-risk optimistic scenario is not evidence that such a site is available.
- Locked rooms, staff turnover and peer availability can prevent timely help despite nearby medication. Duty, consent, supervision and worker safety need a funded plan.
- SFDPH and DOPE obtain free intranasal naloxone through California’s NDP. Additional delivery depends on continued supply; donated resources are not zero societal costs.
- National donations may support work outside San Francisco. No specific local gift restriction or current unfunded expansion has been verified.
Benefits not included in our estimate
- Nonfatal overdose morbidity
- Peer income, dignity and social connection
- Electronic detection and button systems
- Later healthcare and other public resource costs
- Any separately counted SFAF benefit for the same event or person
4. What do you get for your dollar?
Our central estimate: $2,425,273 per 10 QALYs.
Annual donor cash is modeled as 4 × $50 × 52 for peers, plus 5 coordinator hours/week × $60 × 52, plus $4,000 for other delivery support: $30,000. The historical peer payment is a calibration only; staffing, pay adequacy and overhead require a current quote. The health model assumes 100 resident-years, 0.3% all-overdose fatal risk under usual care, 80% potentially opioid-responsive deaths, 20% additional risk reduction and 50% funding additionality. Each prevented acute death receives about 5.154 discounted future QALYs using the same survival assumptions as SFAF.
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
USD PER BETTER LIFE: 10 × $30,000 ÷ (100 × 0.003 × 0.80 × 0.20 × 0.50 × 5.1541)
≈ $2,425,273 per 10 QALYs
Model inputs and assumptions
- Annual donor cash
- 30000 (range: 15000 / 30000 / 60000). Central $30K is a transparent staffing construction, not $439086 divided by contract training targets. Optimistic/pessimistic are alternative work-package assumptions. Analyst judgment unless explicitly sourced.
- Occupied resident-years
- 100 (range: 160 / 100 / 50). 100 is a hypothetical full-year exposure. The pilot involved 50- and160-unit buildings; units are not a measured occupied population. Analyst judgment unless explicitly sourced.
- Annual all-overdose mortality probability
- 0.003 (range: 0.02 / 0.003 / 0.001). 0.3% is a judgment near the historical2010–17 SRO rate of278.7/100000. Neither historical rate nor our scenario is a current target-site observation; 0.1%–2% tests selection and time changes. Analyst judgment unless explicitly sourced.
- Potentially opioid-responsive death share
- 0.8 (range: 0.9 / 0.8 / 0.5). 80% is an explicit current-site assumption, not historical opioid involvement157.8/278.7 or proof that every opioid death is reachable. Timely reach is handled in risk reduction. Analyst judgment unless explicitly sourced.
- Added reduction versus existing response
- 0.2 (range: 0.5 / 0.2 / 0.05). 20% is a subjective hypothesis covering access, reach in locked rooms, responder availability, use and successful rescue relative to stations, staff, peers and EMS. No controlled local mortality effect was found. Do not apply a second engagement discount. Analyst judgment unless explicitly sourced.
- Share not replacing funded activity
- 0.5 (range: 0.8 / 0.5 / 0.25). 50% discounts funding substitution separately from response effectiveness. New work must exceed existing contractual commitments; current expansion is unverified. Analyst judgment unless explicitly sourced.
- Survivor health utility
- 0.7 (range: 0.8 / 0.7 / 0.6). 0.70 reuses the SFAF survival scenario for comparability; not a measured DOPE survivor value. Analyst judgment unless explicitly sourced.
- Later annual mortality hazard
- 0.08 (range: 0.055 / 0.08 / 0.15). 8% is an external-survival-informed analyst choice shared with SFAF. It is a continuous hazard, not the site's baseline overdose probability. Analyst judgment unless explicitly sourced.
- Future survival credit
- 15 (range: 20 / 15 / 10). 15-year cap; no normal-lifespan assumption. Annual discount3% is applied within the survival integral. Analyst judgment unless explicitly sourced.
- Additional responder burden
- 0 (range: 0 / 0 / 0). Zero in displayed positive cases means unquantified, not absent. Stress tests subtract0.05 and0.15 QALYs/site-year; these are illustrative, not measured burdens. Analyst judgment unless explicitly sourced.
Illustrative $100,000: 0.412 QALYs—not available funding room
- Optimistic positive: $16,872 per 10 QALYs. 1.15 net acute deaths prevented and 8.89 QALYs per site-year. Favorable assumptions are not observed site performance.
- Central analyst estimate: $2,425,273 per 10 QALYs. 0.024 net acute deaths prevented and 0.124 QALYs per site-year. Favorable assumptions are not observed site performance.
- Pessimistic positive: $688,979,250 per 10 QALYs. 0.000313 net acute deaths prevented and 0.000871 QALYs per site-year. Favorable assumptions are not observed site performance.
Uncertainty. No added response or full funding substitution gives zero benefit. Harmful effects or responder burden can make net QALYs negative; no finite positive cost-effectiveness ratio is displayed.
What would it take to beat $100,000 per better life?
At central cost and survival, the model needs more than 0.582 net deaths prevented per site-year. Yet its entire baseline is only 0.3 all-overdose deaths. Even perfect prevention with no funding substitution cannot cross the threshold at that exposure and risk. A cheaper supplemental tranche or genuinely higher-risk site is a different hypothesis to test—not a reason to assign it the whole site’s benefit.
THRESHOLD: $30,000 ÷ $10,000 per QALY ÷ 5.1541 QALYs per prevented death
More than 0.582 net deaths prevented/site-year
QALY conversion assumptions
- 0 responder QALYs lost/site-year: $2,425,273 per 10 QALYs. 0.124 net QALYs. Illustrative burden stress test, not measured worker harm.
- 0.05 responder QALYs lost/site-year: $4,070,700 per 10 QALYs. 0.0737 net QALYs. Illustrative burden stress test, not measured worker harm.
- 0.15 responder QALYs lost/site-year: No finite positive price. -0.0263 net QALYs. Illustrative burden stress test, not measured worker harm.
Responder psychological burden is unquantified, not presumed absent. The positive cases display zero subtraction; the stress tests show how small site-level losses can alter the result. Nonfatal overdose morbidity and peer income benefits are also excluded.
5. Funding and previous grants
No current DOPE-restricted marginal work plan, site-specific cost or causal mortality estimate has been verified.
Marginal donor cash; publicly supplied naloxone, existing building services and later healthcare are excluded from the cash numerator and must be reconciled separately. The funding discount addresses substitution; the risk-reduction assumption already includes reach, availability and rescue relative to usual care. Do not discount engagement a second time. No benefits from repeated annual tranches may be summed without accounting for the same survivors.
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
- Saving lives in our homes. Olding et al., International Journal of Drug Policy. Qualitative implementation study; pilot size and historical peer compensation, not mortality efficacy. Published: 2023-06-08; retrieved: 2026-09-07.
- SRO overdose mortality2010–2017. Rowe et al., Drug and Alcohol Dependence. Historical local mortality; table1 uses SRO units as population denominator proxy. Published: 2019; retrieved: 2026-09-07.
- FY25–26 performance objectives. SFDPH. Contract1000032403 targets, not achieved outputs or funding room. Published: 2026-06-22 revision; retrieved: 2026-09-07.
- Contracted programs inventory. SFDPH. $439086 listed Naloxone Distribution program line; not current incremental cost. Published: 2026-04-17; retrieved: 2026-09-07.
- Naloxone guideline. SFDPH. Free NDP intranasal supply and clearinghouse/DOPE division; later-page pending approval footers retained in source. Published: 2025-03-06 approval on cover; retrieved: 2026-09-07.
- Health Commission BHS Directors Update. SFDPH. Slide15 reports all156 HSH-funded PSH sites have24/7 naloxone stations; availability not timely response. Published: 2025-01-06; retrieved: 2026-09-07.
- San Francisco response and partner network. National Harm Reduction Coalition. Partner overlap including SFAF; not current independent marginal reach. Published: Undated historical page; retrieved: 2026-09-07.
- Donation page. National Harm Reduction Coalition. General national donation route; DOPE-restricted marginal allocation not verified. Published: Undated; retrieved: 2026-09-07.
Annual expenses: years and sources
Average annual expenses (three consecutive fiscal years): $10,211,151. Organization size is separate from the modeled cost-effectiveness of a donation.
National Harm Reduction Coalition
Whole-organization Form 990 total expenses; calendar fiscal years ending December 31; consistent original-current-year reporting, with 2022 verified from IRS-derived extraction and organization filing summary.
Includes national operations, not only San Francisco DOPE. Original 2022 IRS extraction reports $13,022,356; the later 2023 return reports a 2022 prior-year comparative of $11,648,825, an unreconciled difference. Original 2022 value is retained rather than silently restated. Official 2024 audited GAAP expense is $6,599,528 versus Form 990 $6,583,528, a $16,000 donated-services difference. 2024 is the latest filing found. Original 2022 PDF is linked by the API but was not independently readable in this pass.