Methodology

Woolgar and Pawluch’s (1985) objection of ontological gerrymandering concerns treating some accounts as constructed while treating the analyst’s preferred conditions as unproblematic facts. The archive’s response is to apply the same evidentiary questions to institutional claims, counter-claims, and empirical studies: who produced the account, by what procedure, for which population or setting, and with what limits. A study may support a bounded inference about conditions without making its categories or measurement rules immune to examination.

Consequently, the persistence of crisis language is evidence about public definition and institutional action; it cannot show that distress is unreal or that a condition has remained constant. Conversely, a critic’s challenge to diagnosis is recorded as a claim whose supporting evidence can be assessed, rather than accepted as the analyst’s explanation merely because it contests professional authority.

Counter-claims are selected to expose disagreements with particular arguments, and their repeated place in the display does not establish equally frequent opposition across eras. Comparisons concern the content of those disputes, while claims about the historical frequency or strength of opposition would require a separate sample that searched for supportive and critical statements on the same terms.

The initial search looked for newspaper titles containing “mental health crisis” or “mental health epidemic,” sorted oldest first in ProQuest. It covered US Newsstream and the historical Los Angeles Times, New York Times, Wall Street Journal, Washington Post, USA Today, and Chicago Tribune collections, through 2022.

Because that search found only explicit crisis language, further records were selected to explain changes in definition, evidence, responsibility, and remedy, including material before the phrase became common and developments through 2026.

1. An identifiable claim and source. A record must name a speaker, institution, publication, or governing body whose statement or action can be inspected.
2. A reason for inclusion. The record must change or contest the problem’s definition, evidence, cause, ownership, measurement, or response. A mention of mental health alone is insufficient.

The journal layer follows definitions, population findings, proposed causes and the institutions that produce knowledge. It contains 37 research records. Two further journal articles, Rockmore (1957) and Rosenhan (1973), enter as historical public records, so the Timeline contains 39 journal publications in 197 records. A paper’s publication type is separate from its archival role.

The latest selection reconsiders all 41 displaced works, restores 13 for distinct contributions, and adds Hooker (1957) and Kendell and colleagues (1971). All 22 previously retained papers remain. Older measurement, classification and community research broadens the chronology; no target count or direction ratio defines quality. Twenty-eight displaced works remain in Sources. This is a selected evidence map, not a fresh systematic search or a full-text review of every paper.

Among the 37 research records, 23 papers supply 30 assessed links, each distinguishing the archived interpretation from the specific empirical claim being tested. Fourteen further scholarly connections explain historical, methodological or institutional relevance without an empirical verdict. Each restored or new paper states its contribution and source-access limits. In particular, Rennie’s full study description and Crandell’s full methods were not independently verified. An unscored connection neither affirms nor qualifies nor disputes a claim.

Direction

Affirms supports the specified empirical component. Qualifies supplies relevant empirical evidence that supports it partly or leaves a stated inference limited or unresolved. Disputes supplies empirical findings against it; missing evidence alone is not a dispute. The 30 assessed links contain 7 affirms, 22 qualifies, and 1 disputes. The 49 former Context / not tested links remain in research provenance. Restored papers receive newly reasoned connections, not their mismatched old ratings. Direction totals count links, not independent studies: Goodwin’s three links reuse one analysis, and several social-media reviews overlap.

Scope

Direct addresses the relevant setting, population, measure, or institutional question. Comparative tests a concept or mechanism in another setting. Contextual marks a bounded connection to the conditions around a claim. The assessed links have 13 direct, 11 comparative, and 6 contextual scopes. Scope is separate from direction. Scholarly connections have no empirical scope rating or direction and are excluded from these totals.

The September 19, 2026 review initially examined 78 links using retained study descriptions and targeted primary-source checks, without conducting a fresh systematic search, a full-text review of every paper, or an independent inter-rater assessment. Five targeted replacements first left 56 papers and 75 links. A subsequent selection retained 26 assessed links, removed 49 unassessed links, and added two validation studies and a Goodwin connection, leaving 22 papers and 29 assessments. The contribution-based reconsideration then restored 13 papers and added two historical studies; Bommersbach’s matched youth-utilization finding supplied the thirtieth assessment, while fourteen other admissions received unscored scholarly connections. Earlier decisions, complete displaced records, and reasons for each change remain in research provenance.

Prigerson and colleagues test the prolonged-grief criteria discussed in the DSM-5-TR entry; Miller and Ozonoff examine one DSM-IV category’s neuropsychological distinctness. Neither decides the validity of the whole manual. Prigerson’s samples limit generalization; Miller and Ozonoff’s full methods were unavailable, so the entry reports what the primary abstract establishes and identifies unverified sample details. New links have admission records rather than invented prior directions. The 988 launch and other historical events without a matched test retain their primary sources.

Of the 41 reconsidered works, 28 remain in Sources as background scholarship because a distinct contribution or matched question has not yet justified their return to the Timeline. This placement is an editorial decision rather than a finding of low quality, and a later question can reopen it. Publication and event dates establish sequence without showing that a historical claimant knew or used a study; Goodwin and Twenge also use overlapping survey data, so reuse across records does not constitute independent replication. Link counts are neither votes for a consensus nor weights in an estimate of an effect.

The reading notes cover all 39 journal publications and all 44 connections between papers and records, while the original study descriptions, titles, citations, assessments, and source-access notes remain preserved separately. Technical terms are explained without upgrading an abstract-level check into a full-text review. Connection headings identify the destination’s publication type; an assessment continues to describe what the paper shows in either direction of navigation. The all-scope view includes unscored scholarly connections, whereas selecting an empirical direction or scope excludes connections without that rating.

These notes explain how conflicting dates, source descriptions, and decisions about inclusion were resolved.

Ray, Lemkau, and Packard

Ray’s Mental Hygiene was published in 1863; the Timeline record and its later cross-reference use that date. Lemkau’s 1949 book is Mental Hygiene in Public Health. Packard’s record refers to the 1866 Hartford edition of Marital Power Exemplified, with the commitment and jury verdict treated as earlier events.

Former-patient organizing

The record retains 1969 with an explicit qualification: the movement’s conference history gives that date, while other histories give 1970. A contemporary 1973 newsletter announces the Psychosis Validation Coalition’s Festival of Creative Psychosis. The unverified 1971 coalition date has been removed.

Budget reductions and policy responses

The 1982 record no longer presents a total for broad social-program cuts as a mental-health appropriation. The Columbine record distinguishes the NRA’s December 2012 response from the White House’s January 2013 policy package. The early-psychosis record identifies 50 states and the District of Columbia. The October 2021 professional emergency declaration precedes the March 2022 State of the Union by four months.

Kaiser settlement

The February 10, 2026 release reports at least $28,323,219 in reimbursement and a $2,832,321 penalty. The signed settlement gives the penalty as $2,832,321.90. The archive retains the release’s whole-dollar figure rather than recalculating a penalty from a general statutory percentage.

Historical names and dates

The Hopkins history spells William Sweetser’s name “Sweetzer” and Wickliffe Rose’s name “Witcliffe Rose”; the entries use the published author’s name and the standard biographical form. The 1845 psychosis date comes from a 2023 historical review, not an independently verified first use. See Sweetser, psychosis, and public-health training.

Priority and approval dates

The available clinic history supports early outpatient and child-guidance activity, but does not establish the priority claim “first U.S. outpatient clinic.” Sources agree on chlorpromazine’s 1954 U.S. approval year but differ on the month; the entry therefore supplies the year only. See outpatient care and chlorpromazine.

Publication type and appraisal

The professional book review and Rosenhan’s Science article are classified as Scholarship because they are journal publications. Controversy about a study’s methods affects appraisal, not whether the document is a journal article. The Rosenhan citation is Science 179(4070), 250–258, DOI 10.1126/science.179.4070.250. See the book review and Rosenhan.

Public copies and bylines

ProQuest-only access is a source-availability state, not a historical finding. The Stangler byline is retained without expansion because sufficient authoritative metadata was not located. The letters-page and two newspaper items likewise retain the available citation rather than an invented public link. See 1913 newspaper item, Stangler, letters, and newspaper coverage.

Organizational intent

The Active Minds record supports claims-making through a standardized campus network. “Brand strategy” would add an intention the public record does not establish. PhRMA’s 2023 statement is an added context document because the original newspaper title search would not retrieve a trade-association blog. It is the selected industry statement making the societal claim in the association’s own name. See Active Minds and PhRMA.

The organizational cohort

Six organizational records enter as contextual additions rather than newspaper-search results: Fountain House, the American Association of Suicidology, the Mental Health Law Project, the Treatment Advocacy Center, The Trevor Project, and Crisis Text Line. Each is dated from the organization’s own published history, which is used for dates and self-description and not as evidence that the organization’s substantive claims are correct. Fountain House is dated to its 1948 incorporation; the We Are Not Alone group that preceded it formed at Rockland State Hospital earlier in the decade. The Mental Health Law Project’s own history names “a group of committed lawyers and professionals” without listing them, and the connection to the Alabama right-to-treatment litigation is taken from secondary accounts of that case. The Trevor Project’s record uses the organization’s anniversary account of August 11, 1998, the date of the HBO broadcast and the line’s first night; a later founder retrospective on the same site gives March 25, 1998. The archive retains the dated anniversary account and records the discrepancy here rather than reconciling it silently. Crisis Text Line is dated to its August 2013 launch, and the 2022 arrangement named in its note is the license under which conversation data trained a commercial spin-off’s customer-service software, which the organization ended at the end of January 2022 after the arrangement was reported publicly; the episode is summarized in a 2025 review of the consent and data-sharing questions it raised. The record cites the organization’s own founding history, because the archive uses organizational self-histories for dates and self-description rather than as evidence about conduct.

The international classification records

Eight World Health Organization records enter as contextual additions because the classification, diagnostic interview, and burden measures it develops are used in the archive’s account of U.S. definitions, measurement, and policy. The records run from ICD-6 in 1948 to the 2022 World Mental Health Report. Global burden estimates and member-state commitments retain their international scope; admitting them does not turn them into U.S. prevalence estimates or evidence of domestic implementation. The WHO profile belongs to a separate intergovernmental group, while the state group continues to identify U.S. federal and judicial actors.

The MAHA cohort

Nine records enter as contextual additions on September 25, 2026. The archive already held the department’s 2026 plan and the profession’s answer to it, but not the commission documents the plan grew from, the regulatory history it revised, or the movement that carried the claim outside government. A model proposed the cohort through a web search, and each record was checked against a primary or archived copy of its source before it was admitted. The commission’s assessment links the corrected version the administration now publishes, whereas archived copies of the original of May 22 preserve its citation to a study that does not exist. The 2004 warning links an archived copy of the FDA page that the assessment cites, because the live page no longer resolves. Similarly, the 1991 hearing rests on a 1997 case study that relies on contemporaneous wire reporting, since no public transcript was located. The records place the claim that psychiatric prescribing causes the crisis in the chronology as the archive places every other causal claim, by recording who made it, on what evidence, and who answered it.

Implementation and outcomes

Expansion of coordinated specialty care shows that a program was put into operation. It does not by itself establish population-level effectiveness. The FDA’s July 2026 guidance is treated as a guidance document; it does not establish whether a drug was approved later. See the early-psychosis set-aside and FDA guidance.

Repeated estimates must come from a federal statistical agency, vital-record system, or identifiable national probability study team with a traceable public result. The single exception is Stress in America, which the American Psychological Association commissions from The Harris Poll: its samples come from an online panel rather than a probability design, yet its stress question kept one wording and one 1–10 scale from 2007 to 2025, and the association titled its 2020 report a national mental health crisis. A producer-generated study snapshot is admitted when the organization reporting it also produced the study. Selected historic snapshots retain their population, finding, and limits without receiving a connected trend line.

Series

At least two comparable observations, a time axis, source ledger, and explicit breaks.

Study snapshot

A national study or first observation that does not yet support a like-for-like series. The producer, population, threshold, recall period, and fieldwork are shown without a time axis.

Historic snapshot

A selected historical measurement with its type of inference and population boundary. It explains how people were counted or a result was generalized.

Comparison example

A result used to explain a difference in measures. Inclusion does not add it to the 509-estimate total.

NSDUH’s 2021 redesign is not connected to pre-2020 estimates, and 2020 is omitted. Household Pulse’s July 2021 recall-period change breaks the line. NHANES begins a new segment when fieldwork resumes after 2021. The 1996 extension of the historic self-appraisal measure is separated from its 1957–1976 replication segment. Stress in America stays one line because its question never changed, and the changes in its panel weights in 2023 and 2024 are stated beside the line rather than drawn as breaks.

Percentages, counts, age-adjusted rates, and the 1–10 stress rating retain their published units. Unlike outcomes may inform the same question, but they are not normalized or overlaid as one mental-health index.

The charts do not decompose an observed increase into change in diagnostic definitions, change in underlying conditions, and other measurement effects. Diagnostic rules and conditions may both change, so neither a category count nor a prevalence estimate alone supplies that attribution.

The DSM-linked dashboard extract includes 13 indicators and 326 estimates, including the first 2024 NSDUH GAD-7 observations for adults and youth. Although the chapter’s library now carries the 2025 GAD-7 observations as the second point of each series, the extract has not incorporated them, so its coverage is distinct from both the library and the latest available publication.

Return to the comparison · Open the full source inventory

The horizontal axis is an author-defined description of relation to the profession, not a published scale of expertise or influence. It distinguishes no professional standing, lay petition, adjacent expertise, credentialed participation inside the field, and authority to set its diagnostic rules. These categories adapt the chapter’s ownership question to the selected profiles; they are not presented as Gusfield’s own five-part taxonomy.

No standing in the profession

The documented intervention rests on political, legal, market, or public standing without a professional role in the mental-health field.

Lay petitioner

A person or movement presses a claim through lived experience, family connection, or advocacy without professional credentials supplying the basis of the claim.

Adjacent expertise

An actor participates through research outside clinical practice, advocacy administration, service administration, or drug development and sponsorship.

Inside the profession

Clinical credentials, psychiatric research, or sustained scholarly participation in the field supplies the documented basis, including criticism from within it.

Sets the rule

The organization authors and maintains a diagnostic classification the field is bound to, whether that is the manual the profession reads or the classification its payments and its mortality statistics are coded on.

The World Health Organization enters the rule-setting column on that same definition rather than on the reach of its claims. It has authored and maintained the International Classification of Diseases since 1948, whose sixth revision carried the first section for mental disorders and supplied the base the American manual was drawn from in 1952. The United States uses ICD-10-CM, the clinical modification maintained by the National Center for Health Statistics, for diagnosis coding, while mortality statistics use ICD-10 itself. These related classifications have distinct coding rules and uses. Widening the column from the manual to the classification is what admits it, so the column now holds two bodies and states a dependency between them rather than a rank. Its placement at mixed grounds follows the archive’s treatment of every other governmental body, because the organization publishes a program budget while no disclosure in this archive ties its classification to revenue.

The pharmaceutical-industry profile moves from inside the profession to adjacent expertise: the selected records establish product development, sponsorship, and trade advocacy, rather than a company’s clinical credentials or authority to write the manual. The ordinal arrangement remains an interpretive judgment; proximity to rule-setting does not rank scientific truth, moral worth, or political power.

The vertical axis describes the documented basis of participation, resting on financial disclosure where one exists and on the ground an actor gave where none does. Its lowest row is reserved for participation resting on material interest alone, which is why it holds the insurance industry, the drug industry and the National Rifle Association, none of which takes part in the field on any other ground. The American Psychiatric Association sits one row above that floor, at chiefly material interest, because the fourteen-year publication-revenue series broadens the evidence for its material stake well beyond the single industry-financing figure. The association nevertheless convenes the profession and issues its positions across the archive, so it holds that stake alongside a professional ground rather than in place of one. Neither placement implies that every member is financially motivated, and missing financial evidence does not establish disinterest.

This is a purposive display sample of the larger claims collection. To qualify, a mark must identify a dated institutional act or public claim, a named lane, one of the five ownership operations, and an exact supporting Timeline record. A shared interest, an undated relationship, or a later interpretation without its own act does not create a mark. Where multiple operations qualify, the display retains selected institutional transitions and contests; it is not an exhaustive application of the admission rule.

Although the World Health Organization now has a profile and eight supporting Timeline records, the ownership map retains its eight institutional lanes and 41 selected operations. The classification records are available through the profile and Timeline; their omission from the map reflects its display selection, not an absence of rule-setting authority. Level-of-care criteria and commercial utilization guidelines remain outside the collected records, so the archive cannot map their authority from this evidence.

The 12 connectors summarize the selected periods for which a lane holds, claims, or disowns authority. Their endpoints follow the dated acts or the 2026 archive cutoff, while length does not establish continuous activity or influence. Connectors are separate from the count of 41 operations.

Six qualifying events shown in the earlier 47-mark version are omitted from this display to keep detailed financing and enforcement sequences in Money: the 1980 Mental Health Systems Act, 2009 NAMI funding disclosure, 2024 parity/network rule, 2025 suspended parity enforcement, 2026 Kaiser settlement, and the 2025 withdrawal of the specialized 988 service. Their absence is an editorial display decision, not evidence that ownership was unchanged. The 1980 act remains explicitly linked in the live chapter’s reading of the 1981 withdrawal.

The chapter contains 42 profiles, 51 documented relationships, and 41 ownership operations. Each relationship must return to a dated supporting record.

Relationships

A line joins two profiles only where a dated archive record supports the connection. An absent line therefore records a gap in the archive, not proof that no relationship existed. Line styles identify authority, advocacy, industry, or contest; distance, proximity, and motion do not measure influence.

Position

Professional standing and basis of participation are ordered interpretive categories rather than scores. They describe how a claims-maker entered the field, not its reach or power. Position within a cell has no additional meaning, and the map measures neither influence nor the spread of opinion.

Ownership

Time runs downward through eight institutional lanes. Marks identify assertion, transfer, contest, disowning, or ratification of authority. The 41 operations are selected dated acts rather than a census; the selection protocol explains the admission rule and the six events retained in Money. A new claim to ownership does not necessarily remove an earlier holder.

Persistence

A bar follows the profile’s declared span on the 1843–2026 scale, with open ends closed at 2026. Solid ticks fall within that span; hollow ticks identify later references or connections outside it. Neither establishes continuous activity, visibility, or power.

Grouped profiles

A grouped profile may join several people in a documented tradition, so its duration cannot be read as one person’s career.

Assigned dates

Where dates had to be assigned from the supporting records rather than declared, they remain an editorial choice open to revision.

Claims-only boundary

The 37 evidence studies do not create 37 author profiles. An evidence author enters this directory only when a separate, attributable public claim or institutional act meets the claims-maker inclusion rule.

The twelve institution examples are a close-reading sample of institution-owned pages, not a site census or a frequency ranking. Archive significance alone was not an admission rule. The first decision identifies the sentence’s sense: an individual use remains coherent when “mental health crisis” is replaced by “a psychiatric emergency”; a societal use remains coherent when replaced by “a national problem.” The second decision records the institutional work attached to that language.

PhRMA’s statement supplies its public rationale, not evidence of motive or treatment effectiveness. The joint youth emergency declaration supplies a professional claim, not an independent prevalence measure. KFF/CNN’s 90% figure measures reported agreement with the national crisis claim, rather than independently diagnosing prevalence, severity, or a trend.

Role assignments concern the documented claim or institutional act, and they do not establish private motive, endorse an attributed cause, or imply that the selected records cover the entire field. Relationships are documented ties; position is ordinal; persistence records duration rather than influence.

Profession, government, and service agencies

The APA’s stakeholder role identifies institutional interest without proving motive. Congress and the White House are grouped for federal legislation and administration; professional standing remains distinct from political responsibility. The 1981 retreat is partial disowning because the federal role continued. The later Congress, Surgeon General, White House, HHS, and SAMHSA profiles distinguish legislation, advisory authority, executive coordination, regulation, and service implementation. The Congress profile does not attribute a position to every member. The Surgeon General and White House additions each rest on one supporting Timeline entry.

Public and former-patient claims

Survey respondents appear more often than direct public speakers in the collected newspaper records. This reflects the sampling frame and does not show that public claims were absent. Packard, the patient movement, and Deegan entered as later context rather than title-search results. Deegan’s ownership basis is a definitional claim from someone classified by the diagnostic system. The veterans’ profile likewise records non-professionals gaining definitional authority, supported by one Timeline entry.

Advocacy, press, and markets

NAMI is an operative because it organizes family experience into public programs. Active Minds and the Kennedy Forum carry claims through organizational networks without establishing a shared brand strategy. The press is both venue and claimant. Industry’s interest supports stakeholder coding; the selected PhRMA statement also supports an operative role. Insurance is principally a stakeholder in these records, where its presence concerns policy conduct. The KFF/CNN poll is operative claims-making because reported agreement becomes a public statistic.

Contested causes and separate disputes

HHS’s claims about prescribing and Fromm-Reichmann’s parent-blaming account are recorded without endorsement. The absence of credible evidence for the mass-shooting claim remains stated in the profile. RFK Jr.’s claims from federal office belong under HHS rather than the Kennedy family’s advocacy role. The MAHA movement holds a profile of its own, because the MAHA Institute makes the crisis claim in its own name outside government. It is coded as a lay petitioner on chiefly values, since its documented intervention rests on advocacy rather than credentials and no disclosure in this archive ties it to revenue. The FDA’s records of 1991, 2004 and 2025 stay with the federal science agencies, as the agency’s earlier records do, although the 2025 panel was convened under the same administration as the department’s plan. Vatz disputes the crisis label itself; the NRA uses it in a gun-policy dispute. These are different functions from maintaining a mental-health institution.

Later references

Meyer’s proposed mental-hygiene districts are treated as an antecedent of community-care policy. A hollow 1963 mark identifies that later connection, not personal activity by Meyer in 1963. Profile duration and evidence ticks therefore retain distinct meanings.

Each core item must link to its exact dated Timeline event or peer-reviewed evidence record. Undated relationships, general allegations, and shared interests alone do not qualify. The register of funding arrangements contains 34 unique records and 36 analytical placements. The 1981 program withdrawal and 2026 Kaiser settlement each support two readings. Four series organize the records: Public capacity, Payment & rights, Claimant resources, and Withdrawal & enforcement. The 2026 reading of payment for deprescribing rests on the department’s plan and on the CMS billing guidance of May 4, 2026, which the plan announced and which covers every kind of medication rather than psychiatric drugs alone. Each reading answers the same five questions, from Source to Effect. Its marks show what is documented at each step, alongside the argument and the limits of its evidence.

An answer may be directly documented, implied by an identified institution or instrument, proposed, refused, suspended, withdrawn, or not documented. Whether resources moved is kept separate from what an arrangement changed. Solid fill means documented; hatching means implied by the arrangement; a dashed outline means proposed or intended; a slash means blocked or withdrawn; a pale segment means not established. A complete row of marks means that all five stages are documented rather than that a policy succeeded, and no mark encodes an amount.

In the community-care reading, each of the seven events opens on its own record, and a second view lists the dated Timeline records the archive places beside it. Those records include the earlier proposals and later investments that supported other parts of service provision, and because the lists show where the archive reads a connection, they establish neither a causal sequence nor a common funder.

Four official Context references and the fourteen-year APA publication-sales series are supplementary. They are excluded from the 197-record Timeline, 34 core records, and 36 placements. The APA Form 990 filings published by ProPublica report aggregate publication sales, not sales of a particular manual. Revision-year markers do not explain a year’s revenue.

The opening U.S. series and WHO country profiles retain separate definitions and denominators, documented below. Neither is a sum of the financing records.

Amounts retain their stated dollars, fiscal years, reporting windows, recipients, and denominators. Appropriations, organizational receipts, physician payments, and publication sales are neither summed nor inflation-adjusted into a common total.

The WHO commercial-determinants framework identifies routes such as research funding, lobbying, and donations. Here, those routes are inspected only where a dated document or study supports a connection.

A disclosure, filing, contract, payment record, enforcement action, or study linking sponsorship to an outcome can support a specific claim. Investigative reporting and historical scholarship are assessed through their supporting records. Shared interests, funding, or access alone cannot establish coordination, capture, or a changed diagnostic criterion.

The table of Senate inquiry findings sets out what the inquiry, or an organization in its own account, reported for one period and one total, so its rows stand side by side without being compared. A share of funding from industry shows who provided resources but cannot show that any decision changed as a result.

The earlier figure was mental-health treatment spending, not all health spending. SAMHSA’s 2016 report gives $186.089 billion for mental-health treatment in 2014 and $2,915.343 billion for total health spending within its accounting scope: $186.089 ÷ $2,915.343 × 100 = 6.383%, rounded to 6.4%. Substance-use treatment is reported separately at $33.891 billion. Public mental-health spending is $110.313 billion and private spending is $75.776 billion, rounded to a 59% / 41% split.

The source is SAMHSA, Behavioral Health Spending and Use Accounts, 1986–2014 (HHS Publication No. SMA-16-4975, 2016), executive summary and Tables A.1–A.2. This estimate covers a different set of services and uses a different denominator from the BEA health-spending account used here. The chapter therefore does not join the two estimates into a trend. The smaller BEA dollar total does not establish a decline from the SAMHSA estimate.

The first table divides each U.S. share into the spending it counts and the total it divides by. The second sets aside, one step at a time, the parts of the spending difference that a narrower condition group or the long-term care settings could explain, so that what remains is the difference in the settings both accounts cover.

Placing both counts over the same total widens the gap to 2.78 times, since BEA’s spending over WHO’s total gives 3.19% and the model’s spending over BEA’s total gives 10.96%.

The care-setting values come from a query retrieved on October 4, 2026, which the research register keeps with its checksum. The model publishes them in 2021 dollars, and each is therefore restated in 2019 dollars in proportion to the model’s own 2019 total, which its share of GDP gives in current dollars. BEA’s category values come from its detailed disease workbook.

Each pair separates modeled mental care from reported current health spending. Both are percentages of GDP. A treatment share can change because either input changes; it does not by itself measure a change in services.

United States

United Kingdom

China

Canada

Germany

France

Japan

India

Australia

The Atlas forecast dataset supplies a second denominator for 2019. It uses a different release and price base; this is an alternative-denominator scenario, not a confidence interval or a replacement series.

Publication records: 2013 · DSM-5 · 2022 · DSM-5-TR · APA profile

This chapter follows financing arrangements within the records, although it cannot reconcile costs shifted between health care, housing, corrections, and households; isolate every recipient group; or reconstruct all the payers behind one person’s care. It establishes neither population mental-health change nor a funder’s influence over a particular professional decision.

The marks at each step show how well the arrangement is documented. They do not measure dollars or policy success. Legal duties and institutional changes count as effects; improved health requires separate evidence. Construction grants do not establish continuing staffing or access; an authorization does not establish delivered capacity. The sixteen-bed payment threshold does not establish clinical quality. Financial dependence and disclosed payments do not, alone, prove altered professional decisions.

SSI includes mental and physical impairments; its supplementary source does not isolate recipients qualifying through mental impairment. Parity-analysis duties do not establish every plan’s compliance. Medicare billing eligibility does not measure local enrollment or access. Publication-year markers do not establish what caused changes in APA revenue.

Record density

A bright or busy year describes the records collected, rather than the frequency or severity of a historical condition.

Arena totals

Six arenas apply to all 197 entries: 160 claims records and 37 journal articles assigned to academia. The arena-distribution figure uses only the 160 claims records. Evidence-map articles were selected through a different search and would otherwise make the distribution appear to migrate toward academia.

Relationships

A line identifies an event-backed tie, not influence inferred from spatial proximity.

Measurement

A defensible series establishes change in its own construct and population; unlike measures do not form one national mental-health trend.

The synthesis concerns institutional action around a public problem. It neither estimates one continuous national condition nor concludes that the country has too much mental-health care. Expanded eligibility and inadequate access for severe illness may coexist.

The archive retains source descriptions and available publication metadata rather than adding unverified fields to meet a formal citation style. Because many historical records lack complete metadata, missing authors, dates, public copies, or publication fields remain missing until supported by an identifiable source.

Research links prefer an identifiable DOI, PubMed record, or publisher page, while discovery links remain in the research record to preserve how a paper was found. A public bibliographic page may establish an article’s identity without providing its full text, so public-copy availability does not imply that every study’s methods were inspected. Where a historical item links to a corroborating account rather than the cited original, its source note states that substitution.