Summary
What do they do? Sanctuary provides counseling and crisis support to survivors of gender violence and their children. Its lawyers help clients with protection, immigration, housing, public benefits and family matters. It also provides shelter, career training, job placement and practical financial support.
Why we’re interested in this organization:
An integrated service portfolio addresses safety, mental health and financial barriers.
Original financial statements distinguish donated legal services from paid resources.
Career support includes childcare access, transportation and other practical assistance.
Our main reservations:
The $207.1M reference per DALY depends on unmeasured health and financial effects.
Service reach, placements and salaries do not establish additional benefit beyond other providers.
Marginal expansion, survivor-controlled resources, harms and client overlap remain unknown.
What do you get for your dollar? $20.7M per DALY in New York City. DALY is our one-year comparison unit. QALY-based health gains and income-welfare equivalents retain their original assumptions; they are not measured clinical DALYs. Conditional marginal unrestricted-gift judgment reference; 10.00 healthy-year equivalents combining health and signed net-consumption welfare.
Reported annual people, additional clinical responders and distinct adults with retained consumption changes; gross wages, visits, jobs and case wins are not automatic QALYs.
1. What do they do?
Sanctuary for Families, Inc., EIN 13-3193119, operates with New Beginnings Housing Development Fund Corporation in the consolidated portfolio. We assess unrestricted support across its services. Training, internships and placement may improve economic resources; legal work may secure benefits, housing and support. EEP Legal services
2. Monitoring and information sharing
FY24–25 reports 9,230 adults and children served, more than 2,600 clinical/crisis clients, more than 7,000 legal clients, 428 shelter residents and more than 450 broad economic-support clients. These are people, not sessions; cross-program uniqueness and therapy completion are undisclosed, so do not sum them. The impact page says 65% find work within one year; EEP reports $41K average salaries without a current annual cohort or comparison group. Neither establishes causal earnings or survivor-controlled net consumption. Feedback and progress forms support monitoring. Needed: deduplicated adults/households, all-entrant follow-up, exposure, retention, baseline/comparison consumption, taxes, benefit withdrawal, work costs and resource control. Annual report Impact Evaluation
3. Qualitative assessment
PATH randomized normal advocacy against advocacy plus eight psychological sessions and two follow-ups in 260 UK women; 64% had twelve-month follow-up. Depression and distress improved, but physical health and further abuse did not differ; missing-data imputation reduced effects. This supports a mental-health mechanism, not Sanctuary's utility, response or injury prevention. The new reference removes uncalibrated shelter-injury credit and halves the prior response (.10 to .05), utility (.08 to .04) and duration (one year to half a year). These choices test mixed crisis/referral/therapy exposure, transfer and persistence uncertainty; they are judgments, not fitted trial coefficients. Shelter can still provide safety and housing value. Retaliation, stress and retraumatization are possible harms; adverse-event imbalance does not prove a causal harm rate. PATH
4. What do you get for your dollar?
Conditional reference: $207,106,666 per 10 NYC-MSA healthy-year equivalents, separating health QALYs from signed independent consumption welfare. This is not an empirical expected value. Annual average non-donated spending is about $3,677 per reported client, or 0.000272 reported client-years of reach per budget dollar; additional reach from an actual gift is unknown. The assumed half-capacity expansion produces 0.000001915 additional clinical responders per dollar. Financial pathways use distinct adult resource recipients, not visits, gross wages or household multipliers. Judgment priors: 67.5 workforce adults (15% of a 450-person broad-support anchor) and 200 disjoint legal/benefits adults; $20,000 baseline consumption; net gains $1,500/$1,000; .25 additional-resource shares; one finite year and one/.5-year delays. A 3% time discount plus .005 ordinary mortality attrition applies; no survival benefit or ordinary extra survivor income is credited. Half positive financial welfare remains after health overlap; negative costs receive full weight across all 450 exposed workforce adults and 200 legal adults without a success multiplier. Positive, negative, clinical-null, neutral, zero-expansion and unknown cases are explicit. The old $34,644,592.51237719 alpha health-only price rises because clinical assumptions shrink and injury credit is removed, partly offset by conditional financial welfare; the full original model remains diagnostic. Full donated-resource burden gives $62.6M per DALY with unchanged annual outcomes and capacity; donated labor/goods are an economic resource cost, not the gift's marginal cash requirement.
Model, assumptions and sensitivity
Full consolidated non-donated gross accrual cost C=$33,936,024. D=1 is arithmetic normalization, not a gift suggestion. b maps annual-recipient output to conditional marginal output; actual gift expansion unknown. Existing pro bono labor remains baseline. The reference is cash-like non-donated accrual, not cash flow. Full-resource sensitivity uses $102,625,589 including donated-resource valuation with identical annual outcomes and capacity assumptions.
.95 NYC MSA share based on local operations is a judgment, not measured residence. Half-share scenario tests leakage; policy benefits unpriced.
Health=D/C*b*N*r*u*exp(-k*lag)*(1-exp(-k*T))/k; k=ln(1+d)-ln(1-m). Income=.5*people*discounted receipt-years*ln(1+net gain/baseline consumption)*causalShare*editionShare*independentShare; people=D/C*b*N. Income effective discount=(1+.03)/(1-.005)-1 includes common-alive attrition. .5 is the repository $50K/$100K welfare conversion. Health fields are health-only. Price10=10*D/(editionHealth+signedIncomeEquivalent) if total positive and known.
- C
- 33936024 USD/year (observed). Consolidated full non-donated gross accrual expense. [sff-audit]
- D
- 1 USD (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides.
- b
- 0.5 model parameter (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides.
- g
- 0.95 model parameter (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides.
- d
- 0.03 model parameter (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides.
- clinicalN
- 2600 people (observed). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides. [sff-annual]
- clinicalR
- 0.05 model parameter (judgment). Half alpha response; additional exposure/completion/causal response unresolved.
- clinicalU
- 0.04 model parameter (judgment). Half alpha utility; PATH symptoms do not map directly to utility.
- clinicalT
- 0.5 model parameter (judgment). Half-year finite benefit; persistence uncalibrated.
- clinicalLag
- 0.25 model parameter (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides.
- clinicalMortality
- 0.005 model parameter (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides.
- financialMortality
- 0.005 model parameter (judgment). Common-alive receipt attrition, no extra survival/income benefit.
- workforceN
- 67.5 people (judgment). 450 broad-service people *15% judged adults with resource improvement; not placement rate. [sff-eep] [sff-impact]
- workforceConsumption
- 20000 USD/adult/year (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides. [sff-eep] [sff-impact]
- workforceNetGain
- 1500 USD/adult/year (judgment). 9000 resources -1000 taxes -3500 benefit withdrawal -3000 costs =1500 hypothetical net. [sff-eep] [sff-impact]
- workforceCausal
- 0.25 model parameter (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides. [sff-eep] [sff-impact]
- workforceYears
- 1 model parameter (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides. [sff-eep] [sff-impact]
- workforceDelay
- 1 model parameter (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides. [sff-eep] [sff-impact]
- legalN
- 200 people (judgment). 200 judged distinct adult resource recipients excluding workforce group. [sff-legal]
- legalConsumption
- 20000 USD/adult/year (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides. [sff-legal]
- legalNetGain
- 1000 USD/adult/year (judgment). 2000 recovered resources -250 access costs -750 lost partner support =1000 hypothetical net. [sff-legal]
- legalCausal
- 0.25 model parameter (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides. [sff-legal]
- legalYears
- 1 model parameter (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides. [sff-legal]
- legalDelay
- 0.5 model parameter (judgment). Explicit unmeasured reference assumption; inspect executable parameters and scenario overrides. [sff-legal]
- independentShare
- 0.5 model parameter (judgment). Half positive welfare retained beyond health; negative losses receive full weight.
Conditional judgment reference: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. Unmeasured reference: half annual output expands; 5% additional clinical response, .04 utility for half a year; 67.5 workforce adults and 200 disjoint legal/benefits adults, net consumption changes +1500/+1000, one year, .25 causal share and .5 welfare-overlap residual. Not empirical expected value.
Hypothetical neutral income, explicitly unmeasured: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. Income is assumed net zero for illustration only; not evidence that financial effects are absent.
Clinical null with positive financial welfare: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. No additional clinical effect; signed financial components remain positive under reference assumptions.
Clinical null with financial burden: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. Benefit withdrawal, costs or lost resources outweigh gains; financial burdens persist when clinical effect is zero.
Stronger positive conditional expansion: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. All stronger parameters are judgments, not measured forecasts.
Low expansion and short benefits: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. Small effects and capacity with fixed whole-recipient cost.
No donor additionality: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. Gift entirely substitutes for other funding; no new services or financial effect.
Adverse health and financial effects: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. Stress/retaliation/retraumatization modeled as 0.00 QALY per annual clinical client plus net financial burdens; magnitude entirely hypothetical.
Health and income remain unknown: Cost: $1; New York City health QALYs: unknown; all-population health QALYs: unknown. No calibrated recipient causal-health utility, net resources or marginal expansion data: combined price is not identified.
Reference health, unknown signed income: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Conditional health retained; unknown financial welfare blocks combined price.
Half benefits within NYC MSA: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. Residence allocation is not measured.
Two-year financial delay: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. Reference effects arrive later.
All financial welfare overlaps health: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. No independent income-equivalent credit; health remains.
Positive health, negative net consumption: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. Independent signed financial burdens reduce total despite positive clinical benefit.
Original alpha health-only diagnostic: Cost: $1K; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. All original serialized inputs retained exactly; not current welfare reference.
Full donated-resource burden: Cost: $1; New York City health QALYs: 0.00; all-population health QALYs: 0.00. Income-equivalent healthy years: 0.00. Same annual outcomes, causal/resource assumptions and capacity b as reference; cost denominator includes donated legal labor and goods, plus event benefits. Total-resource burden sensitivity, not actual marginal cash requirement.
Counterfactual: Existing public services, other workforce/clinical providers, legal aid, benefits and self-found jobs continue. Positive causalShare is an additional-resource fraction, not placement success. b is donor expansion separately. Negative scenarios charge exposed cohort costs with causalShare1.
Attribution: Legal/benefits adults exclude workforce adults; no child or household multiplier. Positive financial independentShare.5 removes possible mental-health overlap; negative burdens retain1. No new shelter-injury or legal-to-health credit.
Explicit conditional judgment reference, not empirical expected value. Effects, consumption baseline, resource control, donor capacity and geography lack calibrated data; signed adverse, unknown and null cases remain.
Sensitivity
- Hypothetical neutral income, explicitly unmeasured: $28.0M per DALY. Income is assumed net zero for illustration only; not evidence that financial effects are absent.
- Clinical null with positive financial welfare: $79.9M per DALY. No additional clinical effect; signed financial components remain positive under reference assumptions.
- Clinical null with financial burden: no finite positive estimate. Benefit withdrawal, costs or lost resources outweigh gains; financial burdens persist when clinical effect is zero.
- Stronger positive conditional expansion: $1.3M per DALY. All stronger parameters are judgments, not measured forecasts.
- Low expansion and short benefits: $700.6M per DALY. Small effects and capacity with fixed whole-recipient cost.
- No donor additionality: no finite positive estimate. Gift entirely substitutes for other funding; no new services or financial effect.
- Adverse health and financial effects: no finite positive estimate. Stress/retaliation/retraumatization modeled as 0.00 QALY per annual clinical client plus net financial burdens; magnitude entirely hypothetical.
- Health and income remain unknown: no finite positive estimate. No calibrated recipient causal-health utility, net resources or marginal expansion data: combined price is not identified.
- Reference health, unknown signed income: no finite positive estimate. Conditional health retained; unknown financial welfare blocks combined price.
- Half benefits within NYC MSA: $39.4M per DALY. Residence allocation is not measured.
- Two-year financial delay: $21.0M per DALY. Reference effects arrive later.
- All financial welfare overlaps health: $28.0M per DALY. No independent income-equivalent credit; health remains.
- Positive health, negative net consumption: no finite positive estimate. Independent signed financial burdens reduce total despite positive clinical benefit.
- Full donated-resource burden: $62.6M per DALY. Same annual outcomes, causal/resource assumptions and capacity b as reference; cost denominator includes donated legal labor and goods, plus event benefits. Total-resource burden sensitivity, not actual marginal cash requirement.
Unresolved inputs
- Actual marginal expansion from unrestricted gifts
- Deduplicated clinical exposure/completion, causal utility/duration and adverse events
- Current workforce cohort net resources including taxes/benefits/work costs/control
- Distinct legal/benefits adults, controlled consumption baseline and alternative-provider effects
- Overlap, household allocation, residence and finite retention
Health and income breakdown
For the modeled $1: 0.00 health QALYs + 0.00 income-equivalent healthy years in New York City: $20.7M per DALY. Income equivalents are welfare comparisons, not observed QALYs or DALYs.
5. Funding and previous grants
FY25 operating expense $102,237,655 plus $387,934 donor event benefits gives $102,625,589 gross; removing $68,528,009 donated legal services and $161,556 goods gives $33,936,024 non-donated accrual cost. FY24 comparable cost is $31,917,251. All programs, overhead and event costs remain in scope; this is not cash outflow. The 990 reports $33,654,637: Schedule D subtracts pro bono services and $55,009 indirect fundraising from operating expense. Annual-summary $33,539,090 is $9,000 below comparable net non-donated operating expense; this residual remains unreconciled. Only two comparable original years are verified, so no three-year average is claimed. Public, fee and volunteer support stay baseline; no gift-funded expansion bottleneck is documented. The official donation route is live. Audit 990 Donate
Annual expenses
Organization-level spending, including programs, administration and fundraising. The research list averages three comparable, consecutive full fiscal years when available.
- FY 2024: $31.9M; Sanctuary for Families, Inc. and New Beginnings Housing Development Fund Corporation, consolidated, 12-month period, audited gross accrual expense less donated legal services and goods. Source
- FY 2025: $33.9M; Sanctuary for Families, Inc. and New Beginnings Housing Development Fund Corporation, consolidated, 12-month period, audited gross accrual expense less donated legal services and goods. Source
6. Sources
- FY25/24 consolidated original audit. Sanctuary for Families. Published: not stated; retrieved: 2026-10-05.
- FY25 original return Part I and Schedule D. Sanctuary for Families. Published: not stated; retrieved: 2026-10-05.
- FY24-25 annual report. Sanctuary for Families. Published: not stated; retrieved: 2026-10-05.
- Evaluation. Sanctuary for Families. Published: not stated; retrieved: 2026-10-05.
- PATH randomized trial. Ferrari et al. / PLOS One. Published: 2018-11-27; retrieved: 2026-10-05.
- Original financial index. Sanctuary for Families. Published: not stated; retrieved: 2026-10-05.
- Economic Empowerment Programs. Sanctuary for Families. Published: not stated; retrieved: 2026-10-05.
- Impact and Results. Sanctuary for Families. Published: not stated; retrieved: 2026-10-05.
- Legal Services. Sanctuary for Families. Published: not stated; retrieved: 2026-10-05.
- Ways to Donate. Sanctuary for Families. Published: not stated; retrieved: 2026-10-05.
- EEP 2011–2015 original report. Sanctuary for Families. Published: not stated; retrieved: 2026-10-05.