Who funds the advocates?

The reason we can show you any of this

A public charity's Form 990 has its Schedule B — the schedule of contributors — redacted by law in the copy the public can read. The 990 will tell you NAMI raised $35,296,208 in contributions in FY2023. It cannot tell you that a single dollar of it came from a drug company [3].

So the only reason this page exists is that NAMI volunteers the information, quarterly, with named companies and exact amounts. That cuts in NAMI's favour and the page says so plainly: an organisation that discloses is legible, and organisations that disclose nothing look cleaner only because they are opaque. Do not read the detail here as evidence that NAMI is the worst actor. It is evidence that NAMI is the visible one.

What the registry shows

Reconciliation status first: 22 of the 25 published quarters reconcile to the penny against NAMI's own stated totals; three do not, and the discrepancies are itemised below rather than papered over. With that on the table — all 25 quarterly registries, Q1 2020 through April 2026: 861 line items totalling $74,820,902.97, of which $20,826,948 across 276 lines came from drug manufacturers [1].

Why 2026 is not on either chart. Two disclosure-regime changes break the series, and both must be labelled rather than smoothed over. First, the 2026 registry is the only one that states a reporting floor, and it is $10,000; earlier registries state no threshold and contain rows as small as $5,000. Second, the purpose column was dropped — every registry from 2020 to 2025 published what the money was for, and the 2026 one does not. 2026 is also a four-month stub, not a year. Its manufacturer share of 44.8% is an artifact of a short window and a raised floor, and presenting it as a trend point would be dishonest.

Three quarters do not add up, and we are not papering over it. Of the 25 registries, 22 reconcile to the penny against NAMI's own stated totals. Three do not: Q4 2022 by −$87,000, Q4 2025 by −$32,025, and Q2 2023 by −$5,000. These were checked with a second, independently written summation before any conclusion was drawn. They are arithmetic discrepancies in NAMI's published PDFs, not parsing errors on our side [1].

A classification judgment you should be able to overrule. Corporate foundations are counted by their legal donor identity, so the Sozosei Foundation — Otsuka America Pharmaceutical's corporate foundation, EIN 84-3915935 — contributes $825,000 across the window and is counted as other, not manufacturer. Counting it as manufacturer money would raise the manufacturer share by roughly one to three points in the years it appears. The same question applies to the AbbVie Foundation and to Tides Foundation (lululemon). We picked a rule and are telling you what it is rather than choosing silently.

Who the manufacturers are

Basis: the eight largest manufacturers by total disclosed contributions in the registry. Together they account for $15,553,587 of the $20,826,948 manufacturer total (276 line items); the remaining ~$5.3 million is spread across smaller manufacturer entries not shown in this table.

The most quotable rows are the purpose strings

Because NAMI published what the money was for through 2025, the registry contains manufacturer money explicitly earmarked for policy and advocacy — 15 rows, $1,075,000. This is the strongest table on the page because every cell is the funder's own description, not our characterisation [1].

Two further details, both disclosures rather than inferences [1]:

• Agency-routed money is disclosed as such. Q3 2023, “High Lantern Group (Otsuka)”, $50,000. Q3 2024, “High Lantern Group (on behalf of Otsuka)”, $100,000, for “PTSD Awareness & Education”.
• Q4 2025, Sozosei Foundation, $200,000, for “Unbranded Educational LAI grants”. LAI is long-acting injectable antipsychotics. A drug company's foundation funding “unbranded education” about a drug class the company sells is precisely the disclosed-but-notable fact this page exists to surface — and note that it is disclosed.

The government record: 2005–2008

There is no public pre-2020 registry. NAMI's own page says anything earlier is available only by emailing donor services. The 2005–2008 numbers exist because a Senate committee compelled them [2]:

“Senator Grassley received a response from NAMI National, which reported that it received $28,659,300 from pharmaceutical companies from 2005 to 2008.”— U.S. Senate Committee on Finance, S. Rept. 112-11, 111th Congress, issued 2011-03-31 [2]

And from the same section [2]:

“All but three (Alabama, Arizona, Connecticut) of the NAMI affiliates responded. All stated they accept unrestricted contributions from pharmaceutical companies.”

And the passage that fairness requires, from the same page of the same report — because the conduct changed after the disclosure, and that is part of the honest story rather than an inconvenience to it [2]:

“NAMI National responded that NAMI's chartered state organizations are independently incorporated and that NAMI National does not have direct oversight on the operations. However, for the last 2 years NAMI has undertaken a Standards of Excellence process that will include expectations for maintenance of independence, conflict of interest procedures, and transparency.”

The public quarterly registry that makes the rest of this page possible is downstream of that. A reader who takes away only “NAMI took $28 million” has taken away half the document.

The ratio, which is a better headline than the dollar figure

NAMI's FY2023 Form 990 reports $35,296,208 in contributions [3]. The 2023 registry reports $3,505,550 from manufacturers [1]. So manufacturer money is roughly 10% of NAMI's contributed revenue.

Both numbers are documented and the ratio is honest. It is also less alarming than a raw multi-million-dollar total sounds, which is exactly why it belongs here rather than in a footnote. Ten per cent is neither nothing nor capture; it is a fact you can weigh.

The finding that cuts against the obvious theory

The intuitive model is that drug companies fund advocacy through their charitable foundations, where it would show up in tax filings. We assembled those filings — 49,847 itemised grant lines from the Sozosei (Otsuka), AbbVie, Pfizer, Eli Lilly and Bristol-Myers Squibb foundations for FY2022–FY2025 [12].

Only 215 of 49,847 lines touch mental health at all. And the pattern inside those 215 refutes the lazy version of the thesis in two ways:

• Sozosei — Otsuka's foundation — is the one making substantial, purpose-labelled policy grants: $300,000 to NAMI (FY2022, “Decriminalizing Mental Illness”), $300,000 to the Kennedy Forum, $255,000 to Inseparable, $140,000 to the Bazelon Center, $115,000 twice to the American Psychiatric Association Foundation, and $100,000 to NAMI National in FY2024 for “Enforcement of the Mental Health Parity & Addiction Act” [12].
• Pfizer's and Bristol-Myers Squibb's foundation money to these organisations is employee matching gifts of a few hundred dollars — $725 to $1,013 in Pfizer's case, $500 in BMS's. That is employee-directed charitable giving, not corporate influence spending. Presenting a $782 matching gift alongside a $675,000 corporate payment as the same category of thing would be straightforwardly dishonest, so we are labelling it rather than quietly totalling it [12].

The real conclusion is about visibility, not volume. NAMI's registry shows roughly $3.5 million a year from manufacturers. The five foundation tax filings show a few hundred thousand across four years. Most industry money to advocacy organisations flows directly from the operating company — and direct corporate giving appears in no tax filing at all. It is visible only where the recipient chooses to publish it. Which is the whole reason the first section of this page says what it says.

The sector, for scale

NAMI is one organisation. For the sector, the anchor is a survey of 439 patient advocacy organisation leaders drawn from 7,865 US PAOs, with 289 responding (65.8%), published in JAMA Internal Medicine [7]:

• 165 of 245 (67.3%) reported receiving industry funding.
• 19 of 160 (11.9%) received more than half their funding from industry.
• Among those receiving industry funding, the median amount was $50,000 (IQR $15,000–$200,000).
• 220 of 269 (81.8%) said conflicts of interest are very or moderately relevant to PAOs; 94 of 171 (55.0%) believed their own organisation's COI policies were very good (the survey's own label, quoted verbatim).

Label it with its date. That survey was fielded September 2013 to June 2014. It is a snapshot, not a current measurement. Note also the second bullet: two-thirds take industry money, and about one in eight are majority-funded by it. Those are different claims and they are routinely merged.

The other sector figure worth having is about money rather than survey answers, and it comes from the companies' own disclosures, counted in a peer-reviewed paper. Of the 10 largest pharmaceutical companies in 2016, only 6 disclosed their payments to patient advocacy organisations in the United States — against all 10 in France, Germany and the UK, where disclosure rules exist. Those 6 reported $120 million of patient advocacy funding across 8 countries, and 74% of it, $88 million, went to US organisations [8].

Read it as a floor, and note what is missing. It is one year (2016), six companies, and only what they chose to publish; four of the ten largest published nothing about their US advocacy funding at all. That absence is the same point this page makes about NAMI: in the US, this money is visible only where someone volunteers it.

Corrected 2026-09-29. An earlier version of this section quoted a 2018 Kaiser Health News investigation: drug companies gave “at least $116 million to patient advocacy groups in a single year” (data year 2015), from a database of 12,000 donations covering 594 of more than 1,200 groups. That was a news organisation's count, its database no longer exists, and this page cites only journals, government records, courts and the organisations' own filings. We have removed those figures and replaced them with the peer-reviewed count above, which measures something different (six companies' own disclosures, 2016) and should not be read as a revision of the old number [8].

The manual's panels — and the rebuttal that has to sit beside them

A naming precision, because this section will be attacked on its weakest citation. The DSM is published by the American Psychiatric Association. The American Psychological Association is a different organisation and has no role in the DSM.

Published counts of financial ties among DSM panel members [4][5]:

• DSM-IV: of 170 panel members examined, 95 (56%) had one or more financial associations with pharmaceutical companies. 100% of the Mood Disorders and the Schizophrenia and Other Psychotic Disorders panels did. Leading categories: research funding 42%, consultancies 22%, speakers bureau 16% [4].
• DSM-5: “Currently, 69% of the DSM-5 task force members report having ties to the pharmaceutical industry,” against 57% of DSM-IV task force members [5].

The American Psychiatric Association published a formal response, and it is required reading beside those numbers [6]:

“In 2012, 72 percent of the 153 members report no relationships with the pharmaceutical industry during the previous year. The scope of the relationships reported by the other 28 percent of member varies: 12 percent reported grant support only, including funding or receipt of medications for clinical trial research; 10 percent reported consultations…; and 7 percent reported receiving honoraria.”

“since there were no disclosure requirements for journals, symposia or the DSM-IV Task Force at the time of the 1994 release of DSM-IV, Cosgrove and Krimsky's comparison of DSM-IV and DSM-5 Task Force and Work Group members is not valid.”— APA, statement of President John M. Oldham MD, March 15, 2012 [6]

These two sets of numbers are not contradictory, and nobody here is lying. Cosgrove's 69% is the task force — a small leadership body. APA's 72%-with-no-ties is all 153 task force and work group members. Different denominators, different years, both accurate. If a page shows you one of these without the other, it is arguing rather than informing. This is the single highest-risk fairness trap on the subject and we would rather bore you than fall into it.

Two limits on the disclosure data itself, both from the critical paper: APA's disclosures are categorical self-reports with no dollar amounts, so no industry-dollar figure can be computed from them; and Cosgrove found them materially incomplete — “None of the DSM panel members identified participation on a speakers bureau. When we did an internet search of the 141 panel members, we found that 15% had disclosed elsewhere that they were members of drug companies' speakers bureaus or advisory boards” [5].

The association's own revenue — and the number we refuse to give you

For 2006, APA's own annual report gives operating revenues and grants of $62.4 million and does not break out industry money at all — journal advertising sits inside Publishing ($28.6 million), and industry-supported symposia and exhibit fees sit inside Meetings ($14.8 million) [9].

On 10 July 2008 Senator Grassley, as ranking member of the Senate Finance Committee, wrote to APA's medical director asking for “an accounting of industry funding” from January 2003 onward, by year, company, amount and reason [16]. The letter asks; it contains no figure. APA answered on 2 September 2008, and its then-president described what the answer covered: “payments for advertising in APA periodicals; payments for exhibits at our meetings; support for minority and resident fellowships in leadership and research; funding for anti-stigma public education campaigns; and support for public and continuing medical education” [15]. We have not found APA's answer itself in any published Committee record.

What APA did next is the primary record. It began publishing its drug-company revenue itself, company by company, split into advertising, annual meeting and research: $13,939,051 in 2008, $9,214,322 in 2009 and $7,773,366 in 2010 [14]. Set against the $65.3 million of 2008 operating revenue in APA's 2008 annual report [13], the 2008 figure is about 21%. The disclosure does not say whether its scope matches the annual report's revenue total exactly, so treat that share as approximate.

Corrected 2026-09-29. An earlier version of this section led with a 2008 New York Times report that “the drug industry accounted for about 30 percent of the association's $62.5 million in financing” in 2006, and the box below repeated it. That figure came from a newspaper, and no primary record we could find contains it: Grassley's July 2008 letter asks for the data and gives none, APA's 2006 annual report does not break out industry money, and APA's own published totals begin with 2008. We have removed it. The nearest primary figure is APA's own: $13.9 million from drug companies in 2008, about a fifth of that year's operating revenue [13][14].

There is no current figure, and we are not going to invent one. The last year for which we can find APA publishing its own pharmaceutical revenue total is 2010 [14]. The FY2024 Form 990 reports total revenue of $58,199,477 across categories — Publications $14,845,018, Annual Meeting $13,909,681, Membership dues $4,335,076, Royalties $4,528,572 — and none of them isolates industry money [17]. Annual Meeting bundles exhibit fees with attendee registration; Publications bundles journal advertising with subscriptions; Royalties includes DSM licensing. An “industry share of APA revenue” for any recent year cannot be computed from public records. Any percentage you see quoted for a recent year is not from a public record.

And the conduct change, which belongs here. APA's 2008 annual report states in its own words that “in early 2008, the APA established a work group to develop guidelines for interactions between the APA, individual psychiatrists and medical industries” [13]. Its Industry Supported Symposia page exists in the Internet Archive in 2009 and the equivalent 2010 URL returns a Page Not Found [18] — consistent with the phase-out of industry-supported symposia. We flag that as suggestive rather than proof: an archived 404 is a URL observation, not an announcement.

How the industry talked about advocacy, in its own words

Read the provenance before the quotes. These are 2002 documents. They are 24 years old, they describe one company's plan for one drug, and they are litigation exhibits selected by plaintiffs to be adverse. One correction, made visibly: an earlier version of this box said the documents “say nothing about NAMI or any other organisation named on this page” — that was wrong. The Key Player Playbook explicitly names NAMI, NMHA and NDMDA as national campaigns to support. What the documents do NOT show is any payment conditioned on advocacy positions — support listed in a marketing plan is not a quid pro quo, and no document here evidences one. They are kept in this separate section precisely so that no juxtaposition implies a transaction the record does not support.

From Eli Lilly's “Key Player Playbook”, 9 August 2002, Bates ZY100174816, Plaintiffs' Exhibit 5843 in MDL No. 1596 [10]. The document lists “Advocacy” as a Key Player segment alongside physicians, payers and pharmacists, and treats advocacy funding as a budget line to be traded off:

“We have eliminated all ‘Reintegration’ awards, Center for Reintegration grants and underfunded advocacy by $700K to insure that our exposures in CMHCs, our most valuable customers, are sufficient.”

“Primary Care will be a critical component of growing the bipolar market. We have not invested to create this market or aid in the detection of bipolar disorder. Gaps include: – $1M journal spend from $0 – $800K for CME to diagnose bipolar – $500K to understand DTC advertising for bipolar disorder”

From the “Zyprexa Global Marketing Plan 2003-2004”, 16 July 2002, Bates ZY100072220, Plaintiffs' Exhibit 1155 [11] — diagnosis discussed explicitly as market growth:

“Bipolar Disorder is often misdiagnosed as unipolar depression by psychiatrists and primary care physicians alike. It can take years before the manic aspects are detected. Development of a diagnostic tool to help clinicians differentiate bipolar from other disorders could significantly grow the market.”

And advocacy appearing as an environmental factor in the same plan [11]:

“Social/Cultural: The advocacy community (especially OUS) will become better organized, educated and developed and make strides to lessen the stigma of mental illness.”

We display the flows and the quotes. Influence remains your inference, and we are not going to make it for you — partly because that is the honest posture, and partly because a page that asserts it can be dismissed, while a page that shows the record cannot.

What this page does not claim

• That any organisation here is a front. They run helplines, support groups and education programmes people rely on.
• That any position was bought. No document on this page shows a payment conditioned on a position, and we have not implied one.
• That NAMI is the worst actor. It is the most transparent one, which is why it is the most documented one here.
• That the 2002 Lilly documents show a payment for a position. The Key Player Playbook names NAMI, NMHA (now Mental Health America) and NDMDA (now DBSA) as national campaigns to support; it records no payment conditioned on any organisation's positions. (Corrected 2026-09-29: this line earlier said the documents name no recipient organisation, which contradicted the correction above.)
• That any of the historical material describes 2026. Grassley covers 2005–2008. The company-disclosure count is 2016. The APA revenue figures are 2006 to 2010. The Zyprexa documents are 2002. Cosgrove's counts are 2006 and 2012. The Rose survey was fielded 2013–14. Only the NAMI registry, the Form 990s and the foundation filings are current.