GiveBetter x Los Angeles

Radiant Health Centers

Integrated HIV and primary care, prevention, counseling and social support

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

Updated: 2026-09-14

Donate

Summary

What do they do? Radiant Health Centers provides HIV care, prevention, counseling and social services in Orange County. A conditional estimate prices selected clinical health benefits against the full recipient's gross expense. It is not a complete portfolio return, and the treatment mix, additionality and transfer assumptions are explicitly subjective.

Why we’re interested in this organization:

  • Current services include actual medication access and treatment support, not only health education.

  • Original audits and returns permit a gross-cost reconciliation including drug and event resources.

  • HIV prevention and treatment have randomized causal evidence, although local marginal delivery remains uncertain.

Our main reservations:

  • Published client counts do not reveal ART, PrEP or therapy intensity and cannot be treated as successful treatment courses.

  • The latest audit still describes liquidity pressure despite improved results and corrected control findings.

  • Much of the conditional central health estimate depends on a cautious transfer from individual depression therapy to a different counseling mix.

What do you get for your dollar? $176.0M per better life: ten additional quality-adjusted life years in Los Angeles. ART, PrEP and counseling partial-health only; full recipient cost.

At s=1, the 1,872 clinic patients imply 748.8 assumed ART patients and 280.8 PrEP patients. Three-quarter-year exposure, .25 incremental strategy fraction and .25 funding response yield 35.1 ART strategy-follow-up years and 13.1625 additional PrEP person-years. The latter avert .1131975 infections at i=.01 and effectiveness .86. Counseling uses 322*.5*.25*.25=10.0625 transferred, donor-responsive CoBalT-equivalent offers. These fractional outputs are model expectations, not observed people.

1. What do they do?

This report covers AIDS Services Foundation Orange County, EIN 33-0126481, doing business as Radiant Health Centers. Its current site offers medical care, HIV prevention and treatment, counseling, case management, housing, food and transport. The recipient is not a single Rapid ART program. Orange County belongs inside the LA/Orange metropolitan edition; the central 98% local benefit share is a residence assumption, not a measured address distribution.

The model uses the full recipient budget while pricing three selected health branches: HIV treatment continuity, PrEP and counseling. Housing stability, non-HIV primary care, reproductive care, nutrition, youth support and broader wellbeing remain unpriced. The result must not be presented as a complete portfolio ranking.

2. Monitoring and information sharing

The latest published impact report covers FY2024 and reports 1,872 clinic patients and 322 unduplicated counseling clients. Its broader supportive-service totals are not added to the clinical cohort because overlap and exposure are unclear. The FY2025 return describes integrated care but does not supply updated unique clinical counts. The central model holds the older counts constant rather than converting spending growth into proven patient growth; a half-to-one-and-a-half capacity factor tests this mismatch.

Current Rapid ART delivery includes same-day starter medication and follow-up for newly diagnosed or disengaged patients. PrEP is also offered, but neither page establishes annual treatment person-years or the share who would otherwise lack medication. The model's assumed 40% ART and 15% PrEP clinic shares are not reported facts. These groups are mutually exclusive for the modeled year; they are not all 1,872 patients counted twice.

3. Qualitative assessment

The SMART randomized trial found approximately two fewer opportunistic-disease-or-death events per 100 randomized-strategy follow-up years with continuous rather than episodic ART. This is not an effect per pill or per extra ART year, nor a same-day versus five-day initiation effect. The model applies it only to a judged subset whose alternative involves meaningful interruption, reduces transfer by half, and values a prevented composite event with a finite one-year health area. It assigns no full remaining lifetime to every medication start.

The PROUD trial found 86% lower HIV incidence with immediate PrEP in a high-risk British population. The model uses a subjective 1% annual untreated risk, below that trial's high-risk setting, and only additional PrEP exposure. Prevented infections receive an explicitly assumed 0.1 annual survival-adjusted health difference for 20 years, discounted and then stopped. No secondary transmission or separate lifetime mortality term is added.

The CoBalT trial found 0.057 additional QALYs over one year from CBT added to usual care for treatment-resistant depression. Radiant's current counseling page primarily describes group counseling over Zoom, with limited in-person groups and individual care case by case. That is not CoBalT's clinical package. Only half of counseling clients are assumed clinically compatible, and a 25% transfer factor jointly accounts for treatment content, intensity and alternative counseling. These are judgments, not measured eligibility or response rates. A further 10% overlap adjustment reduces the sum of the three branches.

4. What do you get for your dollar?

Gross accrual expenses were FY2023 $11,405,310, FY2024 $13,167,555 and FY2025 $13,825,994, averaging $12,799,619.67. The gross figures restore event costs of $112,402, $262,104 and $240,092 respectively. For FY2025, the audited $13,585,902 expense plus the original return's $240,092 netted event cost gives the model denominator. No inventory-cost addback is shown separately.

The latest audit includes $3,903,594 for the 340B program, $2,650,906 case management, $1,819,615 clinical services, $535,478 Rapid ART and $630,206 mental health. Drug-program revenue of $5,072,290 is not subtracted from resource cost. Donated materials of $58,439 are already recognized; the audit describes gift cards, event tickets, trips and other goods, not a verified donated ART stockpile. Unrecognized volunteer time is not valued in the reviewed current statements, so a full social-resource denominator remains incomplete.

The conditional clinical component yields 0.78564 LA QALYs at $13,825,994 of recipient gross cost, or about $175.98 million per 10. Positive joint stress cases run from about $2.77 million to $48.88 billion per 10; these are not confidence limits. The FY2024 public annual report has internally inconsistent program-detail and net-asset presentations, so original audits and returns control the financial model.

Model, assumptions and sensitivity

Full recipient gross expense, not selected clinical-program budget; unrecognized volunteer and outside care resources unknown.

g=.98 judgment for Orange County-centered resident outcomes. No measured LA/Orange address fraction; sensitivity .9–1.

C=13585902+240092=13825994. A(T)=sum(y=1..T)(1.03^-y). QA=N*s*h*f*a*b*r*ta*qe/1.03; QP=N*s*p*f*a*b*i*ep*v*A(T)/1.03; QM=M*s*em*tm*b*qm/1.03. QLA=(QA+QP+QM)*z*g. Central QA=.1703883495, QP=.1635041709, QM=.5568567961 before overlap/geography; QLA=.7856408972. Price10=10*C/QLA. Qportfolio=QLA+Qother-H.

C
13825994 USD annual gross recipient expense (observed). Audit total plus original-return event costs; drugs and recognized donated materials retained. [audit25] [99025] [d25]
N
1872 FY2024 clinic patients (observed). Not diagnosis-specific or treatment person-years. [impact]
M
322 FY2024 unique counseling clients (observed). Not completed courses or responders. [impact]
s
1 current capacity relative to FY2024 counts (judgment). No FY2025 unique counts; test .5–1.5 without assuming cost growth proves volume growth. [impact] [99025]
h
0.4 clinic share receiving modeled ART continuity support (judgment). Actual medication mix unavailable; test .2–.6. [art] [audit25]
p
0.15 clinic share receiving modeled PrEP (judgment). Actual medication mix unavailable; test .05–.3; h+p never exceeds one. [prep]
f
0.75 strategy or PrEP follow-up years per annual patient (judgment). Exposure and retention unavailable; cap at one year, test .25–1. [impact]
a
0.25 medication-strategy exposure additional to alternatives (judgment). Meaningful interruption or absence of PrEP, not simply a few days faster; test .05–.6. [art] [billing]
b
0.25 proportional marginal funding response (judgment). Payer, collections and donor substitution; test .05–.75. [audit25] [billing]
r
0.02 composite events per randomized ART-strategy follow-up year (observed). SMART 3.3 minus 1.3 per100; not effect per extra ART year. [smart]
ta
0.5 ART strategy-effect transfer (judgment). Modern care gaps differ from protocol interruption; test .1–1. [smart]
qe
0.5 finite one-year QALY area per composite event prevented (judgment). Event mix unobserved; no remaining-lifetime multiplier; test .1–1. [smart]
i
0.01 annual HIV incidence absent PrEP in modeled group (judgment). Not local measured risk; test .002–.03. [proud]
ep
0.86 PrEP relative incidence reduction (observed). PROUD trial estimate, conditional on transferable delivery. [proud]
v
0.1 annual survival-adjusted utility difference after infection prevented (judgment). Finite health-area assumption, includes survival once; test .03–.15.
T
20 years after infection prevented (judgment). No health area after year20; low scenario10.
em
0.5 counseling share clinically compatible with depression benefit (judgment). Not measured treatment-resistant depression; test .1–.7. [mental] [cobalt]
tm
0.25 counseling effect transfer including alternatives (judgment). Current mainly-group format is not CoBalT individual CBT; test .1–1. [mental] [cobalt]
qm
0.057 QALY per randomized CoBalT offer over one year (observed). External package versus usual care; not Radiant's observed effect. [cobalt]
z
0.9 cross-component non-overlap factor (judgment). No extra housing/nutrition benefit added; residual shared health areas discounted; test .7–1.
g
0.98 LA/Orange resident benefit share (judgment). Orange County-focused services; exact residence unknown. [current]
d
0.03 annual discount (judgment). Infection health area discounted by year and one initial event year; others one year.
Qother
null non-overlapping remaining portfolio health (unknown). Primary/reproductive care, housing, food and other effects unpriced. [current] [audit25]

Conditional selected clinical health: Cost: $13.8M; Los Angeles QALYs: 0.7856408972015024; all-population QALYs: 0.8016743848994923. s=1;h=0.4;p=0.15;f=0.75;a=0.25;b=0.25;ART transfer=0.5;Q/event=0.5;T=20;v=0.1;mental eligibility=0.5;mental transfer=0.25;z=0.9;g=0.98;incidence=0.01. Joint subjective stress test, not confidence bounds.

Weak delivery and transfer: Cost: $13.8M; Los Angeles QALYs: 0.0028287456506205576; all-population QALYs: 0.0031430507229117306. s=0.5;h=0.2;p=0.05;f=0.25;a=0.05;b=0.05;ART transfer=0.1;Q/event=0.1;T=10;v=0.03;mental eligibility=0.1;mental transfer=0.1;z=0.7;g=0.9;incidence=0.002. Joint subjective stress test, not confidence bounds.

Favorable delivery and transfer: Cost: $13.8M; Los Angeles QALYs: 49.94448521193465; all-population QALYs: 49.94448521193465. s=1.5;h=0.6;p=0.3;f=1;a=0.6;b=0.75;ART transfer=1;Q/event=1;T=20;v=0.15;mental eligibility=0.7;mental transfer=1;z=1;g=1;incidence=0.03. Joint subjective stress test, not confidence bounds.

No additional clinical health: Cost: $13.8M; Los Angeles QALYs: 0; all-population QALYs: 0. b=0 or relevant causal effects are zero.

Illustrative net harm: Cost: $13.8M; Los Angeles QALYs: -1; all-population QALYs: -1. One LA QALY net loss from adverse effects, disruption or displaced care; not observed.

Whole-portfolio return unestimated: Cost: $13.8M; Los Angeles QALYs: unknown; all-population QALYs: unknown. Other pathways cannot be assumed zero for ranking.

Counterfactual: Other clinics, insurance, Ryan White services, 340B financing and usual counseling continue. a limits ART/PrEP exposure to meaningful changes from alternatives. b applies separately to the ordinary gift's resource response. Mental transfer tm includes both modality/intensity and alternative counseling.

Attribution: ART and PrEP shares are mutually exclusive in one modeled year, with h+p<=1. Annual counts are not prescription-years. SMART r is a randomized-strategy follow-up-year rate, not a biological effect per extra medication year. Counseling effect is additional to usual care; z further reduces potential cross-branch overlap. No test, contact, housing count or secondary infection is credited again.

Explicit conditional clinical estimate, not whole-portfolio return. Treatment shares and marginal responsiveness are subjective; broad scenarios, zero and harm are retained. Public fiscal-year counts lag current costs.

Sensitivity

  • Central $175,983,633.86 per10 LA QALYs; favorable $2,768,272.40; weak $48,876,766,269.06.
  • No counseling transfer removes the largest central branch; this is a material assumption, not a robust HIV-only price.
  • The scenario stops ART-event and counseling benefits after one year and post-infection utility after20; ongoing future treatment cost is not borrowed to claim longer intervention exposure.
  • Halving b doubles the positive price; improvement in billing could either expand care or substitute for philanthropy.
  • Adding unrecognized volunteer and outside treatment resources increases social cost; recognized drugs and event goods are already retained.
  • FY2024 count versus FY2025 expense mismatch is explicit, tested through s, not hidden in a synthetic mean.

Unresolved inputs

  • Current unique ART/PrEP patients, duration, adherence, viral suppression and counterfactual care gaps.
  • Local PrEP risk and persistence.
  • Counseling diagnoses, modality, completion and comparative utilities.
  • Unrestricted marginal allocation and payer/collection substitution.
  • Client-address geography and deduplication across clinical/social services.
  • Unrecognized volunteer resources and non-overlapping remaining health pathways.

5. Funding and previous grants

FY2025 public grants and contracts were $6,176,734. The audit describes 340B margins supporting other programs and FQHC Look-Alike status awarded in September 2024. The current payment policy accepts public and private insurance and offers sliding fees. These existing financing routes and alternative clinics belong in the counterfactual; private giving does not automatically create the entire medication benefit.

The latest audit, dated January 31, 2026, reports unrestricted net assets of $306,241, cash of $313,185 and current liabilities exceeding current assets. It discusses management's plans to address negative operating cash flow through billing, reimbursement rates and systems. Two short-term loans totaling $300,000 were paid off in July 2025. The audit opinion is unmodified, current reportable control findings are absent, and prior listed findings are marked corrected. The older going-concern/control flags should not be repeated as unchanged current findings, but liquidity risk remains relevant.

An ordinary gift might preserve service access during a reimbursement gap, fund care beyond payer rules or substitute for later collections and other support. The central 25% proportional funding response is a judgment, not a demonstrated uncovered gap. Full-portfolio thresholds are 138.25994 net LA QALYs for $1 million per 10 or 1,382.5994 for $100,000 per 10 at the annual denominator. These cannot be assessed from the selected clinical branches alone. No outreach or donation was made.

Annual expenses

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

  • FY 2023: $11.4M; AIDS Services Foundation Orange County dba Radiant Health Centers, EIN 33-0126481, 12-month period, February-ending gross accrual expense, restoring netted direct event costs; recognized donated materials included, unrecognized volunteer value unknown.. Source
  • FY 2024: $13.2M; AIDS Services Foundation Orange County dba Radiant Health Centers, EIN 33-0126481, 12-month period, February-ending gross accrual expense, restoring netted direct event costs; recognized donated materials included, unrecognized volunteer value unknown.. Source
  • FY 2025: $13.8M; AIDS Services Foundation Orange County dba Radiant Health Centers, EIN 33-0126481, 12-month period, February-ending gross accrual expense, restoring netted direct event costs; recognized donated materials included, unrecognized volunteer value unknown.. Source

6. Sources

  1. FY2025 original audited statements; auditor date January 31, 2026. Radiant / Stephens, Reidinger & Beller LLP. Published: not stated; retrieved: 2026-09-14.
  2. FY2024 original audited statements. Radiant / independent auditor. Published: not stated; retrieved: 2026-09-14.
  3. FY2023 original audited statements and direct event costs. Radiant / independent auditor. Published: not stated; retrieved: 2026-09-14.
  4. FY2025 original return and netted event cost. Radiant / IRS. Published: 2026-03-25; retrieved: 2026-09-14.
  5. FY2024 original return and netted event cost. Radiant / IRS. Published: 2025-01-15; retrieved: 2026-09-14.
  6. FY2025 audit-to-return expense reconciliation. Radiant / IRS. Published: 2026-03-25; retrieved: 2026-09-14.
  7. FY2024 audit-to-return expense reconciliation. Radiant / IRS. Published: 2025-01-15; retrieved: 2026-09-14.
  8. FY2024 annual report: clinical counts; financial table not used as authority. Radiant. Published: not stated; retrieved: 2026-09-14.
  9. Current integrated Orange County services. Radiant. Published: not stated; retrieved: 2026-09-14.
  10. Current Rapid ART delivery. Radiant. Published: not stated; retrieved: 2026-09-14.
  11. Current PrEP services. Radiant. Published: not stated; retrieved: 2026-09-14.
  12. Current counseling format and scope. Radiant. Published: not stated; retrieved: 2026-09-14.
  13. Insurance, FQHC status and sliding fees. Radiant. Published: not stated; retrieved: 2026-09-14.
  14. Ordinary recipient giving page. Radiant. Published: not stated; retrieved: 2026-09-14.
  15. SMART randomized continuous versus interrupted ART strategies. SMART Study Group / NEJM. Published: not stated; retrieved: 2026-09-14.
  16. PROUD randomized PrEP trial. McCormack and colleagues / Lancet. Published: 2015-09-09; retrieved: 2026-09-14.
  17. CoBalT randomized clinical and economic evaluation. Wiles and colleagues / NIHR. Published: not stated; retrieved: 2026-09-14.