Insulin coma therapy as a treatment for schizophrenia
01
What institutions accepted
Repeated insulin-induced hypoglycaemic coma was treated as a specific or basic therapy capable of producing remission or cure in schizophrenia.
02
What the evidence supports now
Controlled comparisons did not establish a specific therapeutic benefit from insulin coma itself in schizophrenia. Apparent improvement in early uncontrolled series could reflect patient selection, the natural course of illness, intensive nursing and social care, concurrent treatments, and other non-specific effects. Deliberately inducing profound hypoglycaemia also carried substantial risks, including convulsions, prolonged coma, neurological injury, and death. The historical correction is therefore not that no patient ever improved during an insulin-coma course, but that the insulin-induced coma was not shown to be the effective antischizophrenic mechanism institutions claimed.
03
Evidence for the overall belief
Open sources, locators and notes
- DocumentsHistorical beliefIn the shadow of the “psychopharmacological revolution”: Malaria fever, insulin coma, cardiazol and electroconvulsive therapy at the Vienna Psychiatric University Clinic, 1951–1969Wiener klinische Wochenschrift / Springer Nature
Where in the source: “Insulin coma therapy”, paragraphs on early-schizophrenia cure expectations, repeated coma courses, Arnold’s 1960 treatment plan and discontinuation around 1970.
Open-access historical study based on Vienna clinic records. It documents 965 insulin-treatment cases in 1951–1969, mostly insulin coma therapy; the expectation of curing early schizophrenia; a 1960 description of full insulin shock as the basic treatment; and continuation in Vienna to about 1970.
- DocumentsHistorical beliefInsulin coma therapy in schizophreniaJournal of the Royal Society of Medicine
Where in the source: p. 147, opening account of the specific-treatment claim; p. 148, professional support and dispute over whether insulin counteracted schizophrenia.
Documents the specific-treatment claim and the Board of Control and RMPA support that helped establish the practice, while also describing opposition.
- ContextualizesPresent correctionInsulin treatment of schizophrenia; a controlled studyThe Lancet / PubMed
Where in the source: PubMed PMID 13407078, bibliographic record; Lancet 1957, 272(6969):607–611. Results reported in Freudenthal and Moncrieff, PDF p. 7, and Jones, p. 148.
Primary 1957 controlled study by Ackner, Harris and Oldham, widely treated as a pivotal experimental test of whether insulin coma itself conferred a specific benefit in schizophrenia. PubMed supplies bibliographic metadata, not the results text. The outcome was checked through Freudenthal and Moncrieff (PDF p. 7) and Jones (p. 148); the original results were not directly reviewed.
- SupportsPresent correction‘A landmark in psychiatric progress’? The role of evidence in the rise and fall of insulin coma therapyHistory of Psychiatry / University College London
Where in the source: PDF pp. 4–8, “The rise of insulin coma therapy, 1937–53” and “The decline of insulin coma therapy, 1953–65”; p. 7, Ackner comparison and its limits.
Modern peer-reviewed review explaining that efficacy doubts predated randomized trials and that controlled comparisons failed to establish insulin coma as a specific effective treatment. The insulin-versus-barbiturate comparison could not determine the value of the coma regimen itself.
- SupportsPresent correctionInsulin coma therapy in schizophreniaJournal of the Royal Society of Medicine
Where in the source: p. 147, description of the coma regimen, convulsions, brain injury and mortality; p. 148, patient selection, concurrent treatment and intensive care.
Supports the account of treatment risks and explains why improvement during a course did not by itself establish an insulin-specific effect.
- SupportsImpactInsulin coma therapy in schizophreniaJournal of the Royal Society of Medicine
Where in the source: pp. 147–148, Board of Control sponsorship, 31 hospitals in 1938, intensive staffing and repeated treatment over weeks.
Historical review based on British records describing Board of Control sponsorship, use in 31 hospitals by 1938, the highly intensive treatment regimen, mortality and brain-injury risk, professional support, and the 1957 controlled study that found no specific insulin benefit.
- SupportsImpactIn the shadow of the “psychopharmacological revolution”: Malaria fever, insulin coma, cardiazol and electroconvulsive therapy at the Vienna Psychiatric University Clinic, 1951–1969Wiener klinische Wochenschrift / Springer Nature
Where in the source: “Background: the research project”, final paragraphs on selection and 965 insulin therapies; “Insulin coma therapy”, glucagon use and discontinuation around 1970.
The 965 cases concern the study’s selected diagnostic groups and mostly coma therapy. The study also supports continued Vienna use until about 1970 and the later use of glucagon to end coma.
- SupportsImpact‘A landmark in psychiatric progress’? The role of evidence in the rise and fall of insulin coma therapyHistory of Psychiatry / University College London
Where in the source: PDF p. 3, “What was insulin coma therapy?”; pp. 8–11, continuing support, easier drug administration and discussion of decline.
Supports international diffusion, the intensive regimen and the 0.5–4.5% range of historical mortality estimates. This range is not a local hospital rate; the account also cautions against attributing decline to trials alone.
04
Impact
Insulin coma therapy deliberately pushed psychiatric patients into profound hypoglycaemia, often repeatedly over weeks. It spread rapidly from Vienna into psychiatric hospitals internationally and became a highly staffed, institutional treatment for schizophrenia. Vienna continued using it until about 1970 even after controlled studies had undermined the claim of a specific therapeutic effect.
Context
Insulin coma therapy for schizophrenia was distinct from ordinary insulin treatment for diabetes. Patients were intentionally made severely hypoglycaemic and unconscious, then revived with glucose or, later in Vienna, glucagon.
Its decline had more than one cause. Scientific criticism and controlled trials challenged efficacy, while new antipsychotic drugs were also easier to administer. The disappearance of a treatment does not by itself prove why institutions abandoned it.
insulin-treatment cases recorded at Vienna's university psychiatric clinic from 1951–1969, mostly coma therapy
Count from the study's selected records across five diagnostic groups, not exclusively schizophrenia and not a count of individual coma inductions or injuries.
were using deep insulin coma therapy in England and Wales by 1938
range of mortality estimates reported in contemporary insulin-coma literature
Estimates from different historical reports, not a measured mortality rate for the Vienna clinic or The Retreat.
Historical totals and effects are presented as contextual evidence, not as universal causal claims.
Institutional episode 01
Austria — Vienna
Europe
Insulin coma therapy was developed in Vienna in the 1930s and became an established treatment for schizophrenia at the university psychiatric clinic. Clinicians expected early cases to be curable and prescribed repeated comas as a central treatment. Its use declined after neuroleptics appeared and efficacy was increasingly challenged, but Vienna persisted unusually long: a 2025/26 historical study found 965 insulin-treatment cases between 1951 and 1969, mostly coma therapy, and concluded that insulin coma treatment continued there until about 1970.
Institutions involved
- Psychiatric-Neurological University Clinic, Vienna
Institutional context
Manfred Sakel reported early results in Vienna in 1933; the historical clinic study places the development of insulin coma therapy there in the mid-1930s. The treatment became embedded in the university psychiatric clinic's practice and survived the arrival of neuroleptic drugs. During Hans Hoff's directorship, Ottokar H. Arnold still described correctly indicated full insulin shock treatment in 1960 as the basic treatment for schizophrenia. A modern review of the clinic's records found 965 insulin-treatment cases from 1951–1969, the great majority involving insulin coma therapy, and records continued into 1968; the historical study places final discontinuation at about 1970.
Why this persisted
Administrative adoptionVienna's university psychiatric clinic embedded repeated insulin-induced coma in routine schizophrenia treatment, prescribing courses of repeated coma inductions and retaining the basic-treatment rationale even after neuroleptic drugs became available.
How correction happened
UndocumentedThe treatment was discontinued around 1970 after prolonged scientific criticism and therapeutic change, but the existing record does not identify a dated Vienna institutional act formally repudiating insulin coma therapy.
Consequences
- Repeated deliberate induction of profound hypoglycaemia and coma in psychiatric patients
- Courses commonly involving dozens of coma inductions over several weeks
- Risk of convulsions, prolonged coma, neurological injury and death
- Continuation of an invasive treatment after its specific efficacy had become seriously disputed
Evidence for this episode
8 evidence linksOpen sources, locators and notes
- DocumentsInstitutional carrierIn the shadow of the “psychopharmacological revolution”: Malaria fever, insulin coma, cardiazol and electroconvulsive therapy at the Vienna Psychiatric University Clinic, 1951–1969Wiener klinische Wochenschrift / Springer Nature
Where in the source: “Background: the research project”, records and treatment counts; “Insulin coma therapy”, the Vienna clinic’s practice under Hans Hoff.
Open-access historical study based on Vienna clinic records. It documents 965 insulin-treatment cases in 1951–1969, mostly insulin coma therapy; the expectation of curing early schizophrenia; a 1960 description of full insulin shock as the basic treatment; and continuation in Vienna to about 1970.
- ContextualizesContextMalaria fever therapy: wide-spread administration in Vienna was an exceptional case — results from the historical commissionMedical University of Vienna
Where in the source: 24 June 2015 announcement, paragraphs on the 1951–1969 transition and parallel shock/fever therapies; subsequent statement of regret concerns malaria-strain maintenance.
Official university account of the historical commission confirming continued parallel use of insulin coma and other shock/fever therapies at the Vienna clinic during the transition to psychopharmacology. Its statement of regret concerns malaria-strain maintenance, not a formal withdrawal of insulin coma therapy.
- DocumentsMechanismIn the shadow of the “psychopharmacological revolution”: Malaria fever, insulin coma, cardiazol and electroconvulsive therapy at the Vienna Psychiatric University Clinic, 1951–1969Wiener klinische Wochenschrift / Springer Nature
Where in the source: “Insulin coma therapy”, daily courses of at least 50 comas and Arnold’s 1960 overall treatment plan during Hoff’s directorship.
Shows how the clinic prescribed repeated coma courses and continued to regard full insulin shock as basic treatment after neuroleptics became available.
- SupportsConsequencesIn the shadow of the “psychopharmacological revolution”: Malaria fever, insulin coma, cardiazol and electroconvulsive therapy at the Vienna Psychiatric University Clinic, 1951–1969Wiener klinische Wochenschrift / Springer Nature
Where in the source: “Insulin coma therapy”, repeated daily coma, declining annual counts through 1968, disputed efficacy and cessation around 1970.
Supports the repeated hypoglycaemic coma, extended courses and continued use despite disputed efficacy. It does not supply a Vienna-specific count of injuries or deaths.
- SupportsConsequencesInsulin coma therapy in schizophreniaJournal of the Royal Society of Medicine
Where in the source: p. 147, description of convulsions, brain injury and mortality during deep insulin coma treatment.
Supports the general treatment risks, including neurological injury. These are risks of the procedure, not documented adverse-event counts from Vienna.
- SupportsConsequences‘A landmark in psychiatric progress’? The role of evidence in the rise and fall of insulin coma therapyHistory of Psychiatry / University College London
Where in the source: PDF pp. 3 and 5, mortality, fits and protracted coma; pp. 4–8, scientific doubts and controlled comparisons.
Peer-reviewed history of the evidence showing long-running scientific doubts, later randomized comparisons, lack of demonstrated specific efficacy, and contemporary mortality estimates.
- ContextualizesContext‘A landmark in psychiatric progress’? The role of evidence in the rise and fall of insulin coma therapyHistory of Psychiatry / University College London
Where in the source: PDF p. 3, opening paragraph of “What was insulin coma therapy?”, Sakel’s 1933 report to the Medical Society of Vienna.
Dates the early report to 1933, preceding the mid-1930s chronology in the Vienna clinic study. The episode’s start is approximate, not a precisely established first-treatment date.
- ContextualizesInstitutional correctionIn the shadow of the “psychopharmacological revolution”: Malaria fever, insulin coma, cardiazol and electroconvulsive therapy at the Vienna Psychiatric University Clinic, 1951–1969Wiener klinische Wochenschrift / Springer Nature
Where in the source: “Insulin coma therapy”, final two paragraphs, end of use around 1970 after Hoff’s directorship.
Documents approximate cessation of use. The passage does not identify a dated institutional act formally repudiating the treatment.
Institutional episode 02
England and Wales
Europe
England and Wales rapidly institutionalized insulin coma treatment after the Board of Control investigated the Vienna method in 1936. By 1938 it was established in 31 hospitals and had strong professional support. The Retreat began using it in October that year. Criticism intensified in the 1950s, and a 1957 controlled trial found no significant difference between insulin coma and a comparison coma treatment, undermining the claim that insulin itself was a specific treatment for schizophrenia. Use declined around the end of the decade and into the 1960s, while some clinicians continued to defend it.
Institutions involved
- Board of Control for England and Wales
- The Retreat, York
- Royal Medico-Psychological Association
Institutional context
The Board of Control, which inspected and supervised mental hospitals, sent a medical commissioner to Vienna in 1936 and then published guidance on hypoglycaemic shock treatment. Institutional adoption was rapid: by 1938 deep insulin coma therapy was established in 31 hospitals in England and Wales. The Retreat introduced it in October 1938 after favourable reports at the Medico-Psychological Association's July meeting. Professional support continued through the 1940s and 1950s despite scientific criticism, and a leading textbook still endorsed the treatment in 1960. A 1957 randomized comparison found no significant advantage of insulin-induced over barbiturate-induced coma; it could not establish the value of the coma regimen itself. Use declined as antipsychotic drugs displaced the procedure. The approximate end date describes this transition, not a shared cessation date for all the institutions named here.
Why this persisted
Administrative adoptionBoard of Control investigation and guidance helped diffuse dedicated insulin-coma treatment units through the mental-hospital system, reinforced by RMPA meetings and publications. The Retreat's archive describes its adoption following favourable reports at the Association's 1938 meeting.
How correction happened
UndocumentedClinical use declined as controlled comparisons challenged a specific insulin effect and antipsychotic drugs displaced the procedure. The available sources do not identify a dated Board of Control or RMPA withdrawal, or establish when The Retreat stopped using it. Clinical decline is not evidence of a formal institutional repudiation.
Consequences
- Government-supported diffusion of insulin coma units through the mental-hospital system
- Repeated invasive treatment of people diagnosed with schizophrenia
- High staffing requirements and treatment courses extending over several weeks
- Exposure to severe hypoglycaemia, convulsions, neurological complications and mortality risk
Evidence for this episode
9 evidence linksOpen sources, locators and notes
- DocumentsInstitutional carrierInsulin coma therapy in schizophreniaJournal of the Royal Society of Medicine
Where in the source: pp. 147–148, Board of Control and RMPA support, Wilson’s visit and report, professional meetings and publications.
Historical review based on British records describing Board of Control sponsorship, use in 31 hospitals by 1938, the highly intensive treatment regimen, mortality and brain-injury risk, professional support, and the 1957 controlled study that found no specific insulin benefit.
- DocumentsInstitutional carrierThe Retreat Archive — Insulin Coma TherapyWellcome Collection
Where in the source: RET/6/25, “About this work — Description”: October 1938 introduction at The Retreat; catalogue dates 1940–1952 describe the holdings.
Archival catalogue documenting introduction of insulin coma therapy at The Retreat in 1938 after favourable reports at the Medico-Psychological Association's annual meeting. The description specifies October 1938. This evidence is the catalogue description, not the restricted patient records.
- DocumentsMechanismInsulin coma therapy in schizophreniaJournal of the Royal Society of Medicine
Where in the source: pp. 147–148, Board guidance and consultation; RMPA journal meetings, 24 papers over ten years and the 1950 treatment chapter.
Documents government guidance and professional publication channels that supported hospital adoption. These institutional mechanisms extend beyond the single hospital covered by the Retreat catalogue.
- DocumentsMechanismThe Retreat Archive — Insulin Coma TherapyWellcome Collection
Where in the source: RET/6/25, “About this work — Description”: July 1938 professional meeting followed by October 1938 adoption.
Archival catalogue documenting introduction of insulin coma therapy at The Retreat in 1938 after favourable reports at the Medico-Psychological Association's annual meeting. This supports the adoption mechanism at The Retreat specifically.
- SupportsConsequencesInsulin coma therapy in schizophreniaJournal of the Royal Society of Medicine
Where in the source: p. 147, repeated deep coma, convulsions, neurological harm, staffing and courses over weeks; pp. 147–148, hospital diffusion.
Supports all four consequence bullets: diffusion, repeated treatment, staffing and course length, and exposure to treatment risks. Course length does not establish a comparative increase in hospitalization.
- ContextualizesContext‘A landmark in psychiatric progress’? The role of evidence in the rise and fall of insulin coma therapyHistory of Psychiatry / University College London
Where in the source: PDF pp. 6–8, debate, the 1957 comparison, the 1960 textbook endorsement and the 1962 recommendation to stop.
Modern peer-reviewed review explaining that efficacy doubts predated randomized trials and that controlled comparisons failed to establish insulin coma as a specific effective treatment. It also documents continuing advocacy after the trial, so about 1960 is not a universal cessation date.
- SupportsConsequences‘A landmark in psychiatric progress’? The role of evidence in the rise and fall of insulin coma therapyHistory of Psychiatry / University College London
Where in the source: PDF pp. 3 and 5, intensive staffing, repeated coma, mortality, fits and protracted coma.
Corroborates repeated treatment and general risks; it does not establish a mortality rate or count of injuries at The Retreat.
- ContextualizesContextInsulin treatment of schizophrenia; a controlled studyThe Lancet / PubMed
Where in the source: PubMed PMID 13407078, bibliographic record; Lancet 1957, 272(6969):607–611. Results reported in Freudenthal and Moncrieff, PDF p. 7, and Jones, p. 148.
Primary 1957 controlled study by Ackner, Harris and Oldham, widely treated as a pivotal experimental test of whether insulin coma itself conferred a specific benefit in schizophrenia. PubMed supplies bibliographic metadata, not the results text. The outcome was checked through Freudenthal and Moncrieff (PDF p. 7) and Jones (p. 148); the original results were not directly reviewed.
- ContextualizesInstitutional correction‘A landmark in psychiatric progress’? The role of evidence in the rise and fall of insulin coma therapyHistory of Psychiatry / University College London
Where in the source: PDF pp. 7–11, conflicting textbook positions, replacement by drugs and discussion of multiple reasons for decline.
Supports clinical decline and disagreement over its reasons, not a dated Board of Control or RMPA withdrawal. It does not establish The Retreat’s last-use date.
Source registryEach source is registered once and can carry several precise evidence relationships.6 sources
- 016 evidence links
- 021 evidence link
- 03Insulin coma therapy in schizophreniaJournal of the Royal Society of Medicine7 evidence links
- 042 evidence links
- 052 evidence links
- 06‘A landmark in psychiatric progress’? The role of evidence in the rise and fall of insulin coma therapyHistory of Psychiatry / University College London7 evidence links