GiveBetter x SF

Easy Does It Emergency Services

Emergency attendant continuity and wheelchair repair within a broader disability-support organization

Research time: ~7 min on GPT-6 Astra Light
  • Research — reviewed programs, finances and impact evidence.

Published: 11 September 2026.

Funding limitations

Summary

What do they do? EIN94-3227346 is the Berkeley disability-service nonprofit, not the similarly named recovery-housing organization. Emergency attendants assist with transfers, food, hygiene and medication self-administration when regular help fails; this is not replacement full-time care. Repair and transport complete the wider portfolio. More

Why this approach interests us

  • Prompt help with essential daily activities or broken mobility equipment can shorten a genuine gap in care. Readiness has value even when emergency usage is low.

Our main reservations

  • Only finite direct functional relief is quantified. The favorable5% world supplies 92.7% of signed expected health, yet itself costs $9,633,872 per10. Without it the estimate is $2,321,904,339. None of the assigned worlds beats $1M per10.

What do you get for your dollar?

The research list uses the central scenario: $518,235,450 per 10 Bay QALYs. The signed prior-weighted result is $178,597,966 per10 Bay QALYs (0.0000559916792539315 QALYs per illustrative $1,000 gift). The separate central world is $518,235,450. This prices two monitored Berkeley pathways at the full FY2025 organization expense, not a complete portfolio return. Inspect the model and assumptions.

whole financial denominator
$2,166,918 FY2025 expense — Government-contract baseline retained; not the mixed-year website expense or marginal copayment.
health scope
Partial Berkeley health at whole cost — Emergency attendant relief and wheelchair repair only. Matching, accessible transport, outside-Berkeley work and severe-event prevention remain unquantified.
funding status
No verified marginal offer — An ordinary gift is not automatically a new shift or prevented emergency. Current source describes tax-supported services; prior monitoring reported staffed capacity.

1. What do they do?

EIN94-3227346 is the Berkeley disability-service nonprofit, not the similarly named recovery-housing organization. Emergency attendants assist with transfers, food, hygiene and medication self-administration when regular help fails; this is not replacement full-time care. Repair and transport complete the wider portfolio.

Fund ordinary operations

Whole operating expense includes readiness, dispatch, administration and other programs. No denominator reduction for public revenue.

Resolve a nonoverlapping impairment episode sooner

Convert service hours into episode-equivalents, then account for overlapping periods and care alternatives. Hours, visits and new clients are not interchangeable.

Credit finite health only

Multiply additional impaired days avoided by a modest utility difference, subtract separate nonfatal harm, and apply gift-response and Bay share.

Scope of this review. Modeled Berkeley services have Bay share1, not an inferred SF share. All organization costs remain, but unobserved non-Berkeley throughput receives no invented scale multiplier. Donated parts, unpaid work, household assistance, public-system and clinical resources are unpriced complements; this is not a complete societal-resource calculation or a guaranteed lower bound.

2. Monitoring and information sharing

Entire legal organization

Original FY2025 Form990 and ScheduleO, reused verified finance screen. Whole expense$2,166,918; periodJuly1,2024–June30,2025. Dashboard lower expense belongs to older period.

Our assessment. Gross operating cost, not complete societal resources or present marginal budget.

Berkeley contract, not whole-organization cohort

FY25 agency monitoring in March2026 city packet. Tables list878 attendant/473 repair units; quarter service-hour sums872/476.75/172 new service clients and252 new Berkeley clients are not annual unique totals. Fully staffed; lower emergency use attributed partly to matching/BRSD. Survey only five replies.

Our assessment. Small internal discrepancies and YTD display errors retained. No causal independent-living effect or no-alternative rate is identified; funded readiness is not new-gift additionality.

Medicaid beneficiaries in Arkansas, Florida and New Jersey; consumer-directed assistance vs usual arrangements

Cash and Counseling randomized demonstration. Greater paid care, satisfaction and fewer unmet needs in most groups; no greater adverse health/injury susceptibility.

Our assessment. Different sustained intervention and active comparator. Does not identify EDI utility, extra hours, prevented admissions or mortality; functional priors remain judgments.

Full-time wheelchair users with spinal cord injury

Multicenter observational SCI wheelchair repair study. 310 reported repairs;127 had consequences, median5days (IQR2–17.3).

Our assessment. Not randomized repair access. Backup-chair use and other consequences are not necessarily severe health loss; removable delay, utility and EDI case mix are unknown.

3. Qualitative assessment

Prompt help with essential daily activities or broken mobility equipment can shorten a genuine gap in care. Readiness has value even when emergency usage is low.

Key reservations

  • Whole-cost, partial-health scope can omit important benefits; it is not a complete-return estimate.
  • Low measured emergency usage can coexist with valuable readiness; neither proves additional donor impact.
  • Severe clinical outcomes and lasting caregiver placements could change value, but no causal local counts were established.
  • The favorable tail remains far above the target without any output tuning.
  • Older audit deficiencies are not asserted to persist today; current service-table inconsistencies are the narrow data issue.

Benefits not included in our estimate

  • Unmeasured caregiver-matching duration and caregiver wellbeing
  • Unmeasured outside-Berkeley repair/rescue and transportation health
  • Avoided hospital admissions, pressure injuries, institutionalization and mortality without causal evidence
  • Dollar savings treated as QALYs
  • Lifetime mobility benefit or multiple simultaneous impairment periods

4. What do you get for your dollar?

$178,597,966 per10 Bay QALYs

Six signed worlds are probability-weighted in QALYs for a fixed gift before inversion. Prior weights were recorded before first execution: funding-null25%, clinical-null15%, harm10%, cautious20%, central25%, favorable5%. These are judgments, not fitted or calibrated probabilities. No coefficient changes followed the unfavorable first result.

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

Finite two-pathway model: gift / whole expense × cash response × nonoverlap × [(care hours / hours per episode × utility × incremental days /365) + (repair hours / hours per episode × utility × incremental days /365) − episode equivalents × independent harm] × Bay share
0.0000559916792539315 Bay QALYs per$1,000; modeled all-geography QALYs identical because only Berkeley activity is quantified.

Model inputs and assumptions
Whole annual expense
$2,166,918 (range: FY2025 actual; double-resource stress separately). Original return; no old-year expense substitution. Revenue$2,224,118, government contracts$2,074,301, copays$23,073, contributions$121,158. Net assets$482,310 are not verified available cash. High: original annual return.
Monitored service-hour scale
872 attendant;476 repair (range: Adjacent unit table878/473 sensitivity). Quarter sums, not the displayed YTD column that repeatsQ4. Does not establish all-organization activity or unique annual recipients. Low or very low; conversion and health values are analyst priors.
Hours per episode
1.5 attendant;2 repair (range: 1–2 attendant;1–3 repair). Analyst conversion. Hour-level denominators do not establish distinct cases; no assumption one hour equals one life or unique person. Low or very low; conversion and health values are analyst priors.
Residual nonoverlapping benefit fraction
.8 (range: .6–.9). Covers repeat/overlapping impairment periods across repair and attendant pathways; same person can have distinct episodes, but periods must not be stacked. Low or very low; conversion and health values are analyst priors.
Attendant utility and additional time
.15 utility ×.5day (range: .05×.125day to.3×2days;null0). Finite judgment of average impaired time removed before best alternative help/need resolution, including short or zero-delay cases. Not an EDI trial effect or duration of chronic disability. Low or very low; conversion and health values are analyst priors.
Repair utility and additional time
.05 utility ×3days (range: .02×1day to.15×14days;null0). Observational repair consequences motivate days, not six months. Utility, removable delay and high tail unmeasured locally. Low or very low; conversion and health values are analyst priors.
Marginal response to whole gift
.25 (range: 0–.6; full-response diagnostic). Separate from delay before alternative care: budget-level substitution, capacity and ordinary portfolio allocation. No observed $1,000 tranche. Low or very low; conversion and health values are analyst priors.
Independent harm per episode
.00001Q (range: 0–.0001Q). Nonfatal adverse-care/repair stress, not measured event frequency. No mortality credited or double-counted. Low or very low; conversion and health values are analyst priors.
Modeled residence share
1 (range: Berkeley monitored program only). No Bay attribution inferred for the unmodeled whole portfolio. Low or very low; conversion and health values are analyst priors.

Illustrative $1,000; bounded calculator cap$10,000, no linear scale assurance

  • funding_null: No finite positive price per10; weight0.25. Signed gift Bay Q0; annual care/repair episode-equivalents581.3333333333334/238; annual gross pathway Q0.11945205479452055/0.0978082191780822; harm0.008193333333333334, before overlap and gift response.
  • clinical_null: No finite positive price per10; weight0.15. Signed gift Bay Q0; annual care/repair episode-equivalents581.3333333333334/238; annual gross pathway Q0/0; harm0, before overlap and gift response.
  • harm: No finite positive price per10; weight0.1. Signed gift Bay Q-0.000007562199707910807; annual care/repair episode-equivalents581.3333333333334/238; annual gross pathway Q0/0; harm0.08193333333333334, before overlap and gift response.
  • cautious: $84,648,858,042 per10; weight0.2. Signed gift Bay Q1.1813508452829799e-7; annual care/repair episode-equivalents436/158.66666666666666; annual gross pathway Q0.007465753424657534/0.008694063926940639; harm0.011893333333333334, before overlap and gift response.
  • central: $518,235,450 per10; weight0.25. Signed gift Bay Q0.0000192962484634185; annual care/repair episode-equivalents581.3333333333334/238; annual gross pathway Q0.11945205479452055/0.0978082191780822; harm0.008193333333333334, before overlap and gift response.
  • favorable: $9,633,872 per10; weight0.05. Signed gift Bay Q0.001038004201839246; annual care/repair episode-equivalents872/476; annual gross pathway Q1.4334246575342464/2.738630136986301; harm0.00674, before overlap and gift response.
  • unit_table: $178,887,874 per10 Bay QALYs. Signed expected gift Bay Q0.00005590093832571691; unweighted diagnostic, full matrix saved.
  • full_response: $97,079,849 per10 Bay QALYs. Signed expected gift Bay Q0.00010300798890258131; unweighted diagnostic, full matrix saved.
  • double_days: $88,434,158 per10 Bay QALYs. Signed expected gift Bay Q0.00011307847762789123; unweighted diagnostic, full matrix saved.
  • double_resource_cost: $357,195,931 per10 Bay QALYs. Signed expected gift Bay Q0.00002799583962696575; unweighted diagnostic, full matrix saved.
  • no_harm: $175,171,849 per10 Bay QALYs. Signed expected gift Bay Q0.00005708679837395973; unweighted diagnostic, full matrix saved.

Uncertainty. No hospitalization, pressure-injury or death reduction is inferred from an emergency call or independent-living outcome. Repair study median duration is not mean removable delay. Caregiver matching might generate longer benefits than emergency visits, but no verified completed placement/counterfactual duration was found; it is omitted rather than silently inserted. No within-year discount is applied to the short modeled benefit windows.

5. Funding and previous grants

HOLD giving recommendation. Need current additional service capacity per unrestricted gift, reconciled service-hours/episode data, and evidence of consequential unmet care versus available alternatives. Negative partial-health findings are retained, not tuned.

MeasureE primarily funds emergency attendant/matching services, ACTC supports roadside assistance, and accessible rides have public funding. Public money is already in the baseline. Ask: (1) What would an additional unrestricted$1,000 change over the next year that existing contracts, reserves and staffing would not? (2) Reconcile hours, units, unique recipients and truly unserved requests, including outside-Berkeley work. (3) How long do clients otherwise wait for equivalent care, and what completed matching or serious health outcomes are additional? General fundraising and a premises constraint are not a dated, priced additional-capacity commitment. No donation URL was verified in this bounded phase.

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. FY2025 Form990, EIN94-3227346. Easy Does It, original IRS return hosted by ProPublica. Primary. Published: Fiscal year ended2025-06-30; signed2025-11-19; retrieved: 2026-09-11.
  2. FY2025 ScheduleO. Easy Does It, original IRS return hosted by ProPublica. Primary. Published: FY2025; retrieved: 2026-09-11.
  3. HWCAC March18,2026 packet; FY25 monitoring pp12–20 and expense schedule pp22–28. City of Berkeley / Easy Does It. Primary. Published: Meeting2026-03-18; monitoring submitted2025-07-23, approved2025-08-05; retrieved: 2026-09-11.
  4. Current service geography and public-funding descriptions. Easy Does It. Primary. Published: Undated; retrieved: 2026-09-11.
  5. Emergency caregiving: activities and limits. Easy Does It. Primary. Published: Undated; retrieved: 2026-09-11.
  6. Financial dashboard; mixed-year headline, not denominator. Easy Does It. Primary. Published: Undated; retrieved: 2026-09-11.
  7. December premises update. Easy Does It. Primary. Published: December2025; precise day not established; retrieved: 2026-09-11.
  8. Factors Influencing Incidence of Wheelchair Repairs and Consequences Among Individuals with Spinal Cord Injury. Worobey et al.. Primary. Published: Online2021-04-09; issueApril2022; retrieved: 2026-09-11.
  9. Effects of Cash and Counseling on Personal Care and Well-Being. Carlson et al.. Primary. Published: 2007; retrieved: 2026-09-11.

Annual expenses: years and sources

Average annual expenses (three consecutive fiscal years): $2,040,759. Organization size is separate from the modeled cost-effectiveness of a donation.

Easy Does It Emergency Services Program

Form 990/990-EZ reported whole-entity expenses; includes program, administration and fundraising costs, but excludes any costs netted against revenue.

Latest original return and prior-year comparative combined with earlier filing data.