GiveBetter x California

Vision To Learn

School-based vision screening, eye examinations and eyeglasses

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

Updated: 2026-09-13

Donate

Summary

What do they do? Vision To Learn brings vision care to students at school without charging their families. California is a substantial part of a national organization, so an ordinary gift also supports activity elsewhere. Its modeled California health value depends heavily on additional delivery, sustained glasses use and a judgment-based health utility conversion.

Why we’re interested in this organization:

  • A direct, understandable pathway removes logistical barriers to correctable vision care.

  • Randomized school-program evidence supports some educational benefits, and measured refraction supports immediate acuity improvement.

  • Current original accounts and a real California public-funding mechanism make the cost pathway inspectable.

Our main reservations:

  • California's marginal gift allocation is unknown; historical reach is not the next-dollar share.

  • The child health-utility gain and sustained counterfactual correction period are judgments, not measured QALYs.

  • Public reimbursement, large restricted gifts and donated inputs can substitute for cash or change its leverage.

What do you get for your dollar? $2.0M per better life: ten additional quality-adjusted life years in California. Vision-correction health only; California share modeled.

Additional unique children receiving a needed correction course; Additional effectively corrected child-years versus alternative care; Measured visual acuity change, distinct from preference-weighted health utility.

1. What do they do?

The recipient is the national nonprofit Vision To Learn, EIN 45-3457853. Mobile clinics supply screenings, examinations and glasses in low-income school communities. The current site advertises $150 for a child's screening, exam and glasses; that is not verified marginal full-recipient cost. Current program and advertised contribution Ordinary donation route

The 2024–25 annual report's cumulative map shows 262,821 California recipients out of 601,741 nationally, about 44%. This is historical reach, not a fiscal-year or marginal allocation. The report also describes new Dallas work and a two-year grant expanding services in Maryland, DC and New Jersey. Those outside-state benefits remain outside this edition's numerator. 2024–2025 Annual Report

2. Monitoring and information sharing

In a Baltimore cross-sectional study of 4,972 children prescribed glasses, 70% improved at least two lines in their better-seeing eye with refraction. Only 46% met the study's clinically significant refractive-error definition. Immediate best-corrected acuity is not sustained wear, utility or an untreated comparison. Visual acuity and refractive findings in children prescribed glasses

The Baltimore cluster randomized school intervention improved some reading outcomes after one year, but sustained achievement effects were not observed at two years. This supports a real intervention pathway while arguing against automatically crediting lifetime educational gains. Test scores are not converted into QALYs here. Cluster randomized school vision program academic outcomes

A Chinese cluster trial tested free glasses with teacher incentives against prescriptions alone, showing that dispensing and adherence interventions are distinct. Its setting and package do not measure California wear. Free glasses and teacher-incentive cluster randomized trial

Request unique children with a correction gap, delivered courses rather than pairs, observed wear over time, replacement costs, presenting/best-corrected acuity and alternative care. Avoid counting replacements or two pairs as two independently benefited children.

3. Qualitative assessment

School delivery can reduce transportation, scheduling and caregiver barriers. Clearer vision can improve daily functioning even when an educational benefit is not captured by a health metric. Children needing specialist care also require a completed referral pathway rather than only screening.

The model intentionally omits speculative lifetime earnings and education-to-health effects. Its one-year central correction window is a finite advantage over alternative care, not a claim that glasses work for only a year. The two-year favorable case requires continued effective correction and replacement resources within its assumed costs.

For a California-focused donor, the decisive question is the next funded route, staffing capacity and beneficiary mix. A national organization's California history alone cannot establish that an unrestricted gift stays in California.

4. What do you get for your dollar?

The conditional central estimate is about $2.00 million per 10 California QALYs. It is an analyst scenario, not a measured effectiveness estimate or funding offer. A favorable joint-assumption case is about $62,300; an adverse positive case is about $894 million. These are not confidence limits, and zero or harmful net effects are possible.

The calculation keeps the entire recipient gift in the numerator. Its central assumptions allocate 80% to core cash delivery at $225 per unique-child course, discount half the financed capacity for funding substitution, and assign a 60% counterfactual correction gap. Effective wear covers 60% of one year, with utility gain 0.02 while corrected, a six-week delay and a small health-burden allowance. California receives an assumed 40% of net benefit. Every one of those marginal conversion assumptions is judgment, not an observed result.

In model normalization, $100,000 finances 355.56 courses before additionality and yields 1.24776 all-population QALYs, of which 0.499104 are assigned to California. Adding assumed donated resources of $40 per course raises gross resource cost to $114,222 and the California full-resource price to roughly $2.29 million. It does not lower the cash gift denominator.

The $1 million price requires 0.00001 California QALYs per dollar, about twice central yield. The $100,000 price requires 0.0001, about twenty times central. With other central inputs fixed, the utility gain would need to be about 0.040 rather than 0.020 for the first threshold; this is a break-even diagnostic, not evidence of such utility.

Program expense divided by reported pairs is not the model's unique-child marginal cost. Support expenses remain charged, and unmodeled other benefits receive no numeric credit rather than being declared absent.

Model, assumptions and sensitivity

Full ordinary national-recipient gift G retained; f allocates modeled delivery without shrinking denominator. Donated inputs/public leverage disclosed separately. Reuses prior data/us/vision-to-learn-v2.json structure, with fresh original accounts and independently scoped California allocation. Course cost covers core delivery; the 20% outside that allocation covers other recipient activities/support, avoiding a second charge for the same overhead.

g is the share of marginal net health benefit accruing to California residents, not office location, cumulative pairs or restricted-program scope. Central .4, favorable .6, adverse .2 are judgment scenarios. Actual g may be 0–1.

N=G*f/c. D(T)=[1-(1+r)^(-T)]/ln(1+r). Q_all=N*a*[p*w*u*D(T)-h]*(1+r)^(-d). Q_CA=g*Q_all. Price10=10*G/Q_CA for Q_CA>0. Full resources=G+N*inkind+M. If genuinely extra public M funds only CA courses: Q_CA_match=Q_CA+(M/c)*a*[p*w*u*D(T)-h]*(1+r)^(-d); do not multiply that addition by g again.

G
100000 USD (judgment). Calculation normalization, not recommended gift.
f
0.8 core delivery cash allocation fraction (judgment). Remaining 20% retained in full gift denominator for support/other activity; not a restricted gift. [99025]
c
225 cash USD per delivered unique-child care course (judgment). Not advertised $150 nor accounting expense per pair. Includes delivery/follow-up assumption; $150–400 scenarios. [home] [99025]
a
0.5 financial additionality fraction (judgment). Half of financed capacity additional after substitute donors/public funding, not measured. [mosfaq]
p
0.6 fraction with counterfactual correction gap (judgment). Absent adequate alternative correction during horizon; historical Medicaid nonuse is not this probability. [mos]
w
0.6 fraction of correction time used effectively (judgment). Wear, breakage and actual use combined; not directly measured in California. [wear]
u
0.02 utility gain during effective correction (judgment). Preference-weighted child utility is missing; 0.005–0.04 sensitivity, not clinical bounds. [acuity]
T
1 years (judgment). Finite incremental correction period before alternative care/refractive change; no lifetime tail. [rct]
r
0.03 annual discount rate (judgment). Continuous integral using ln(1+r).
d
0.125 years before dispensing (judgment). Six-week illustrative funding-to-benefit delay.
h
0.00005 QALYs lost per additional course (judgment). Small allowance for burden/inappropriate correction; could be wrong.
g
0.4 California share of marginal net health benefit (judgment). Cumulative California glasses share about44% is context, not marginal evidence. 20–60% cases are analyst scenarios, not bounds. [annual25]
inkind
40 additional donated-resource USD per course (judgment). Separate gross resource sensitivity, not deducted from gift. [annual25]
M
0 incremental public USD (judgment). No automatic match for ordinary donation; approved/claimable extra match unverified. Financing may be incremental while delivery still displaces other service: a separately discounts that delivery substitution. [mosdonor]

Conditional central judgment: Cost: $100K; California QALYs: 0.49910387587274435; all-population QALYs: 1.2477596896818608. G100000,f.8,c225,a.5,p.6,w.6,u.02,T1,r.03,d.125,h.00005,g.4. Not an empirical estimate.

Favorable joint assumptions: Cost: $100K; California QALYs: 16.062601770694304; all-population QALYs: 26.771002951157175. f.9,c150,a.8,p.9,w.8,u.04,T2,d1/12,h.00002,g.6,r.03. Sustained correction/replacement included in cost assumption; not a bound.

Adverse positive joint assumptions: Cost: $100K; California QALYs: 0.0011184499854481314; all-population QALYs: 0.005592249927240656. f.65,c400,a.2,p.3,w.3,u.005,T.5,d.25,h.00005,g.2,r.03. Not a lower bound.

Full substitution: Cost: $100K; California QALYs: 0; all-population QALYs: 0. a=0; all funded capacity would occur anyway. Harm-dominant cases can be negative.

Hypothetical approved incremental California public funding: Cost: $100K; California QALYs: 0.9982077517454887; all-population QALYs: 1.7468635655546052. M32000 purchases142.222 extra CA courses at central parameters, giving extra.499104Q. No verified match offer; gross resources151911.11 including inkind40 per all497.778courses.

Counterfactual: Other donors or reimbursement may fund the same capacity (a); children may obtain adequate correction elsewhere even if the gift adds capacity (p). The finite period ends the modeled advantage over alternative care. These are distinct funding and patient-care counterfactuals.

Attribution: Count unique additional delivered child courses, not screenings, exams, multiple pairs or lifetime educational stories. Wear enters once as effective corrected time. No automatic Medicaid matching or causally unmeasured education QALYs.

Conditional model with explicitly judgment-based joint scenarios, not empirically bounded confidence interval. Child utility, marginal California mix and funding additionality dominate. Zero/negative net benefit cannot be ruled out.

Sensitivity

  • g alone at .2 or .6 changes central price to about $4.01m or $1.34m; cumulative reach does not resolve it.
  • p, w and u multiply; sustained correction and child utility dominate alongside funding additionality.
  • Holding remaining central values, one-year break-even utility for $1m/10Q is approximately .040 and for $100k/10Q approximately .398; solve exact formula including h.
  • If two pairs per child inflate output counts, inferred unit costs must rise; the model's c is per unique child course.
  • No public leverage in central. Public payments, donated lenses and shared school resources must not be claimed as free full-resource cost.

Unresolved inputs

  • Actual next-dollar recipient allocation and California delivery mix.
  • Current core delivery cash cost per unique additional child, including replacements and referral completion, reconciled with separately charged support costs.
  • Funding substitution, restricted-grant capacity and qualified incremental MOS claims.
  • Current counterfactual correction gap and observed effective wear over finite follow-up.
  • Child preference-weighted utility by severity; not adult impairment utility or test-score conversion.
  • Latest original audit findings and remediation: attempted ProPublica audit download returned security HTML, so no audit conclusion asserted.

5. Funding and previous grants

Accrual Form 990 total expenses were $19,589,987, $24,131,355 and $25,497,887 for June-ending fiscal years 2023–2025; the comparable mean is $23,073,076.33. The 2024 filing supplies 2023's comparative. Original FY June 2024 Form 990 and 2023 comparative

The 2025 return reports $20,761,273 program expense, $3,314,355 administration and $1,422,259 fundraising. It records 93,225 pairs and 107,520 examinations, outputs rather than health gains. Revenue is $28,180,734; net assets $19,005,970 include $15,354,367 restricted and $3,651,603 unrestricted. Government grants are $3,778,592; Medicaid/Medi-Cal fees $2,576,632 and CHIP fees $751,020. This mix requires funding-substitution analysis rather than assuming each donated dollar purchases otherwise absent care. Original FY June 2025 Form 990

The annual report uses a different funding summary from the tax return; those totals are not silently mixed. Donated optical inputs support delivery but are real resources, not evidence that all additional cash is unconstrained. 2024–2025 Annual Report

DHCS lists Vision To Learn as a MOS provider. Its cited 2018 evaluation found 66% of Medicaid-eligible examined children lacked Medicaid-funded vision care in the prior four years. This is not proof that they had no glasses from any source or a current 66% treatment gap. Mobile Optometric Services program

MOS combines private donations and federal funds, without State General Fund support. Provider qualification and contractual processes matter. MOS funding and provider requirements Donors must undergo DHCS conflict review and receive approval instructions, so the central model assumes no automatic extra public match. MOS donor approval process

An explicitly hypothetical $32,000 additional public payment tied to California delivery would double central California benefit if it bought additional courses under all other central assumptions. No claim that this is currently obtainable is made. Confirm eligible spending, approved donor status, reimbursement timing and whether those payments would occur without the gift.

Annual expenses

Organization-level spending, including programs, administration and fundraising. The research list averages three comparable, consecutive full fiscal years when available.

  • FY 2023: $19.6M; Vision To Learn, EIN 45-3457853, 12-month period, Accrual Form 990 total expenses, June-ending fiscal year. Source
  • FY 2024: $24.1M; Vision To Learn, EIN 45-3457853, 12-month period, Accrual Form 990 total expenses, June-ending fiscal year. Source
  • FY 2025: $25.5M; Vision To Learn, EIN 45-3457853, 12-month period, Accrual Form 990 total expenses, June-ending fiscal year. Source

6. Sources

  1. Current program and advertised contribution. Vision To Learn. Published: not stated; retrieved: 2026-09-13.
  2. Ordinary donation route. Vision To Learn. Published: not stated; retrieved: 2026-09-13.
  3. 2024–2025 Annual Report. Vision To Learn. Published: not stated; retrieved: 2026-09-13.
  4. Original FY June 2025 Form 990. Vision To Learn / IRS via ProPublica. Published: not stated; retrieved: 2026-09-13.
  5. Original FY June 2024 Form 990 and 2023 comparative. Vision To Learn / IRS via ProPublica. Published: not stated; retrieved: 2026-09-13.
  6. Mobile Optometric Services program. California DHCS. Published: not stated; retrieved: 2026-09-13.
  7. MOS funding and provider requirements. California DHCS. Published: not stated; retrieved: 2026-09-13.
  8. MOS donor approval process. California DHCS. Published: not stated; retrieved: 2026-09-13.
  9. Visual acuity and refractive findings in children prescribed glasses. Guo et al., American Journal of Ophthalmology / Harvard. Published: not stated; retrieved: 2026-09-13.
  10. Cluster randomized school vision program academic outcomes. Neitzel et al., JAMA Ophthalmology. Published: not stated; retrieved: 2026-09-13.
  11. Free glasses and teacher-incentive cluster randomized trial. Yi et al., American Journal of Ophthalmology. Published: not stated; retrieved: 2026-09-13.