Terug

Best Practices in Psychosis Treatment

BEST PRACTICES IN
PSYCHOSIS TREATMENT

A current, person-centred and recovery-oriented review

Medication • Psychological therapies • Early intervention • Community rehabilitation • Physical health • Digital innovation

STABILISE

RECOVER

RECONNECT

2026 perspective
Psychosis treatment is not a single intervention. Best outcomes come from early, collaborative, phase-specific care that combines antipsychotic medication when indicated with psychological therapy, family work, physical-health protection, substance-use treatment, social recovery and practical support. The aim is not simply symptom suppression, but a meaningful and self-directed life.

Updated August 2026

Prepared as an educational overview for Neurohaven


Executive Summary

Psychosis is a syndrome rather than a single diagnosis. It can occur in schizophrenia-spectrum disorders, bipolar disorder, severe depression, substance- or medication-related states, neurological and medical illness, and brief or stress-related conditions. Treatment therefore begins with careful assessment of cause, severity, risk, physical health, substance use and the person’s own goals. The original Neurohaven page correctly emphasised combined care; this 2026 revision strengthens that framework and updates areas where the evidence base or clinical emphasis has moved.

What matters most in 2026

Clinical implication

Early intervention

First-episode psychosis should enter a specialist early-intervention pathway rapidly, with medication, CBTp, family intervention and vocational/educational support integrated rather than delivered sequentially.

Shared medication decisions

Drug choice should be based on previous response, side-effect vulnerability, physical health, patient preference, route and practical adherence—not on a simple “typical versus atypical” hierarchy.

Physical health from day one

Weight, waist circumference, BP, pulse, glucose/HbA1c, lipids, prolactin, movement disorder assessment and relevant ECG monitoring are core treatment, not optional add-ons.

Clozapine without avoidable delay

After two adequate antipsychotic trials fail, treatment-resistant schizophrenia should trigger a clozapine discussion and pathway rather than repeated non-clozapine switching.

Recovery is functional

Housing, employment/education, relationships, cognition, trauma, substance use and physical health are treatment outcomes alongside hallucinations and delusions.

Technology is adjunctive

Apps, digital phenotyping, VR and AI are increasingly studied, but most relapse-prediction systems still lack sufficient external validation for autonomous clinical decision-making.

Experimental means experimental

Psychedelics are not a routine treatment for psychosis. Research remains highly cautious because of potential psychosis exacerbation and the historical exclusion of psychotic disorders from trials.

Guideline status in the UK
NICE CG178 remains the principal adult psychosis and schizophrenia guideline. NICE reviewed it again in July 2025 and did not add metacognitive therapy at that time. NICE now lists psychosis and schizophrenia guidance among areas with review work in progress, so clinicians should check current NICE and BNF advice when making treatment decisions.

1. Pharmacological Treatments

1.1 Antipsychotics: the core principle

Antipsychotics remain the main pharmacological treatment for schizophrenia-spectrum psychosis and are commonly used in acute psychosis while diagnostic clarification proceeds. The modern approach is an explicit, individual therapeutic trial: agree target symptoms and goals, document expected benefits and risks, start at an appropriate dose, titrate carefully, monitor response and adverse effects, and review whether the medicine is delivering meaningful benefit.

Choosing an antipsychotic

Consider

Why it matters

Previous response and tolerability

A person’s own history is often more informative than class labels.

Metabolic risk

Olanzapine and clozapine are among the agents with greater weight/metabolic burden; risk varies across individual drugs.

Movement-disorder risk

Higher-potency dopamine D2 blockade can increase dystonia, parkinsonism and akathisia; tardive dyskinesia remains a long-term concern.

Prolactin

Risperidone/paliperidone and some other agents can substantially increase prolactin.

Cardiac factors

Consider QT prolongation, tachycardia, orthostatic hypotension and individual cardiovascular history.

Sedation/activation

These effects can determine daytime function, driving, work and adherence.

Route and adherence

Oral, orodispersible, liquid and long-acting injectable formulations may suit different preferences and needs.

Pregnancy, age, comorbidity and interactions

Choice and monitoring must be individualised; smoking status is especially relevant to clozapine and olanzapine metabolism.

1.2 “Typical” versus “atypical”: useful shorthand, but too crude for prescribing

The first-generation/second-generation distinction remains familiar, but it can obscure large differences between individual drugs. Both groups can be effective for positive psychotic symptoms. A more useful conversation compares the specific adverse-effect and practical profile of each medicine. Clozapine is the major efficacy exception because of its role in treatment-resistant schizophrenia.

Domain

Examples of clinically important effects

Management principles

Extrapyramidal symptoms

Dystonia, parkinsonism, akathisia

Review dose and diagnosis; consider switching; use symptom-specific treatment where appropriate.

Tardive dyskinesia

Persistent choreiform or stereotyped movements

Detect early; review dopamine-blocking exposure; specialist management and VMAT2-inhibitor options where available/appropriate.

Metabolic

Weight gain, dyslipidaemia, insulin resistance/diabetes

Baseline and serial monitoring; lifestyle support; consider lower-risk agent; treat obesity/diabetes according to current guidance; metformin has a growing prevention evidence base.

Hyperprolactinaemia

Sexual dysfunction, galactorrhoea, menstrual disturbance, bone-health implications

Measure when indicated/baseline per guidance; review symptoms and drug choice; consider prolactin-sparing strategies.

Cardiovascular

QT prolongation, orthostasis, tachycardia

Risk assessment, ECG where indicated, review interacting drugs and electrolyte disturbance.

Sedation/anticholinergic

Sleepiness, dry mouth, constipation, blurred vision, cognitive burden

Dose timing, dose reduction/switching, symptom treatment and active constipation prevention where relevant.

Clozapine-specific

Neutropenia/agranulocytosis, myocarditis/cardiomyopathy, seizures, hypersalivation, severe constipation/ileus, metabolic effects

Mandatory haematological monitoring plus active physical-health, bowel, cardiac and toxicity surveillance; consider plasma levels in defined situations.

A major 2026 emphasis: prevent metabolic harm earlier
A 2025 evidence-based guideline concluded that metformin is the best-studied pharmacological intervention for prevention of antipsychotic-induced weight gain and proposed co-commencement in selected higher-risk situations. This is an important emerging practice direction, but it is not a blanket instruction to prescribe metformin to everyone starting an antipsychotic; decisions should follow current local/NICE/BNF guidance and individual risk-benefit discussion.

1.3 Long-acting injectable antipsychotics (LAIs)

LAIs should not be framed as punishment for “non-compliance”. They are a formulation choice that can reduce the burden of daily dosing, make missed treatment more visible, and suit people who prefer regular injections. NICE recommends considering depot/LAI treatment after an acute episode when the person prefers it or where avoiding covert non-adherence is a clinical priority. Shared decision-making remains essential.

1.4 Clozapine and treatment-resistant schizophrenia

Clozapine should be considered when schizophrenia has not responded adequately to two sequential adequate antipsychotic trials, at least one being a non-clozapine second-generation agent under NICE guidance. Delayed clozapine exposure can leave people with persistent symptoms cycling through less effective strategies. Before declaring treatment resistance, clinicians should confirm diagnosis, adherence, adequate dose/duration, substance use, interactions, physical illness and psychosocial contributors.

·       Blood monitoring is mandatory because of neutropenia/agranulocytosis risk.

·       Constipation is a potentially life-threatening clozapine adverse effect and requires active prevention, questioning and treatment—not passive monitoring.

·       Consider myocarditis/cardiomyopathy, seizures, aspiration/pneumonia, hypersalivation and metabolic complications in ongoing monitoring.

·       Smoking cessation or resumption, significant infection (including pneumonia), and interacting medicines can substantially alter clozapine concentrations; blood-level monitoring is useful in selected situations.

1.5 Baseline and ongoing physical-health monitoring

Measure / review

Typical NICE framework

Weight

Baseline; weekly for first 6 weeks; 12 weeks; 1 year; then annually.

Waist circumference

Baseline and annually.

Pulse and blood pressure

Baseline; 12 weeks; 1 year; then annually.

HbA1c or fasting glucose; lipids

Baseline; 12 weeks; 1 year; then annually.

Prolactin

Baseline, then symptom-/drug-guided follow-up.

Movement disorders

Baseline and regularly, particularly during titration and if symptoms emerge.

ECG

Before treatment when indicated by SPC, cardiovascular risk/history, or inpatient admission; repeat according to drug/risk context.

Smoking, diet, activity, nutrition

Assess from baseline and revisit regularly.

Overall physical health

Secondary care retains responsibility for at least the first 12 months or until stable, whichever is longer, under NICE CG178.

2. Psychotherapy and Psychological Interventions

2.1 CBT for psychosis (CBTp)

CBTp is not simply “challenging delusions”. Good CBTp develops a collaborative formulation of experiences, distress, appraisals, safety behaviours, sleep, mood, trauma and social context. The aim is often to reduce distress and disability, strengthen coping, test alternative explanations and support personally meaningful goals. NICE recommends offering CBT to people with psychosis or schizophrenia, including during acute and later phases when the person can engage.

2.2 Family intervention and psychoeducation

Family intervention can reduce relapse and improve the quality of the home environment. Modern practice avoids blaming relatives and avoids reducing the work to “compliance education”. It should support communication, problem solving, relapse planning, carers’ own needs and a shared understanding of psychosis. Consent and confidentiality boundaries should be explicit.

2.3 Trauma-informed and trauma-focused care

Trauma is common among people who experience psychosis. Services should assume that coercion, threat, discrimination and previous adversity may shape how a person experiences assessment and treatment. Trauma-informed care means safety, choice, collaboration, transparency and minimising avoidable coercion. When PTSD is present, trauma-focused therapies can be considered by appropriately trained clinicians; psychosis is not automatically a reason to withhold evidence-based PTSD treatment, but timing, stability and pacing matter.

2.4 Cognitive remediation, social cognition and skills-based rehabilitation

Cognitive difficulties—attention, working memory, processing speed and executive function—can be major drivers of disability even when positive symptoms improve. Cognitive remediation and skills-based rehabilitation are increasingly used to target functioning, particularly when tied to real-world goals such as education or employment. Social skills training may help selected people with explicit social-function goals.

Metacognitive therapy: promising, but not yet a NICE recommendation
NICE conducted exceptional surveillance in July 2025 specifically on metacognitive therapy (MCT) for adult psychosis and decided not to update CG178 to include it at that time. It can be discussed as an evolving evidence area rather than presented as established NICE-standard treatment.

3. Community and Rehabilitation Interventions

3.1 Recovery-oriented community care

The strongest contemporary model is not “hospital versus community” but continuity across settings. People may need acute inpatient care at times, but recovery usually depends on what happens afterwards: stable housing, reliable follow-up, meaningful activity, physical healthcare, relationships and rapid access if warning signs return. WHO continues to emphasise community-based care, autonomy and human rights, while noting that most people with psychosis globally still lack access to specialist mental healthcare.

3.2 Assertive and intensive community approaches

Assertive Community Treatment and related intensive case-management models are designed for people with severe illness, repeated admissions, homelessness, disengagement or complex social needs. Their value comes from flexible multidisciplinary outreach and persistence rather than simply increasing appointment frequency.

3.3 Employment, education and housing

Functional recovery should be treated as a primary outcome. Individual Placement and Support (IPS) has become the dominant evidence-based supported-employment model: it aims for competitive employment quickly, based on personal preference, with ongoing support. Supported education is particularly important in first-episode services. Stable housing is likewise a treatment intervention because housing insecurity directly undermines engagement, sleep, medication continuity and physical safety.

3.4 Peer support and shared decision-making

Peer workers can offer experiential knowledge, hope, practical navigation and a less hierarchical relationship with services. Evidence is stronger for engagement, empowerment and personal recovery outcomes than for dramatic symptom reduction. Shared decision-making should extend to medication, psychological therapy, family involvement, crisis planning, employment, substance use and physical-health priorities.

4. Early Intervention Strategies

4.1 First-episode psychosis: treat the first years as a critical period

Early Intervention in Psychosis (EIP) services are one of the major advances in modern psychosis care. The core principle is rapid, youth-friendly, multidisciplinary intervention before educational, occupational and social losses become entrenched. NICE states that EIP services should be accessible to all people with a first episode or first presentation of psychosis irrespective of age or duration of untreated psychosis.

Core component

What good care looks like

Rapid assessment

Exclude delirium, intoxication/withdrawal, neurological/medical causes and affective psychosis while addressing immediate risk.

Low-burden medication strategy

Shared choice, cautious dosing, proactive adverse-effect prevention and clear review of benefit.

CBTp

Available early and linked to distress, functioning and personal goals.

Family intervention

Offered where family/carers are involved, with consent and clear confidentiality boundaries.

Substance-use intervention

Integrated treatment—particularly for cannabis and stimulants—rather than exclusion from psychosis services.

Education/employment

Supported education/IPS from the beginning, not after “clinical recovery”.

Physical health

Prevent rapid weight/metabolic deterioration from the first prescription.

Relapse prevention

Individual early-warning signs, crisis plan, sleep/substance monitoring, and rapid route back into care.

4.2 Clinical high risk (CHR-P) and prevention

People with attenuated psychotic symptoms or other high-risk features require careful specialist assessment, but risk states are probabilistic: many do not develop a psychotic disorder. Current practice therefore favours monitoring, treatment of anxiety/depression/substance use, CBT and family support rather than routine antipsychotic prescribing solely to prevent psychosis. Avoid deterministic language that can create unnecessary stigma.

5. Physical Health, Lifestyle and Complementary Approaches

The older term “holistic” can be useful, but it should not imply that evidence-based medical treatment is being replaced. In 2026 the more important shift is that physical health and lifestyle are recognised as central components of psychosis care because cardiometabolic disease, smoking, poor sleep, inactivity and treatment adverse effects contribute substantially to premature mortality.

5.1 High-value interventions

·       Structured physical activity adapted to ability and preference.

·       Healthy eating and weight-management support from the beginning of antipsychotic treatment.

·       Smoking cessation with awareness that stopping or restarting tobacco smoking can change clozapine and olanzapine exposure.

·       Sleep assessment and treatment, including circadian disruption and obstructive sleep apnoea where relevant.

·       Dental, sexual and reproductive health, vaccination and routine primary-care prevention.

·       Integrated treatment of alcohol, cannabis, stimulants and other substances.

5.2 Mindfulness, arts and complementary therapies

Mindfulness-based and acceptance-oriented approaches can help some people change their relationship to distressing voices or thoughts, but delivery should be adapted to psychosis and guided by clinicians familiar with the population. Art, music and other creative therapies may support expression, engagement and social connection. Supplements and complementary medicines should not be represented as antipsychotic alternatives; evidence is variable, product quality can be uncertain and interactions are possible.

6. Guidelines and Standards

Organisation / framework

Current relevance

NICE CG178 (UK)

Still the central adult psychosis/schizophrenia prevention and management guideline; reviewed July 2025. Emphasises EIP, shared antipsychotic choice, CBTp, family intervention, physical-health monitoring, LAIs when appropriate and clozapine after two adequate trials.

NICE NG181

Rehabilitation for adults with complex psychosis; useful for persistent symptoms, complex rehabilitation needs, clozapine optimisation and multidisciplinary care.

NICE QS80

Quality standard for adult psychosis/schizophrenia, including physical-health assessment and access to key interventions.

APA schizophrenia guideline

Supports antipsychotic treatment, clozapine for treatment resistance/suicidality, LAIs when preferred or adherence is uncertain, CBTp, psychoeducation, supported employment and coordinated specialty care.

WHO

Emphasises medication plus psychosocial rehabilitation, psychoeducation, family interventions and community-based, rights-oriented services; the 2025 fact sheet highlights the large global treatment gap.

7.1 Precision prescribing and pharmacogenomics

Research increasingly seeks to predict who will respond to which antipsychotic and who is most vulnerable to adverse effects using clinical variables, genetics, blood biomarkers, neuroimaging and machine learning. These approaches are scientifically important but are not yet reliable enough to replace careful clinical trials of treatment in routine care.

7.2 Digital phenotyping, smartphones and wearables

Smartphone and wearable research can measure activity, sleep, mobility, communication patterns and self-reported symptoms. A 2025 review identified 27 unique monitoring apps across 54 studies, but replication was rare and sample sizes were generally modest. 2026 systematic-review evidence suggests some relapse-prediction models can identify risk weeks in advance, but high risk of bias, heterogeneous definitions and limited external validation remain major barriers. These systems should therefore augment—not replace—clinical contact and consent-based relapse planning.

7.3 Artificial intelligence

AI is being explored for relapse prediction, outcome prediction, clinical decision support, language analysis and service triage. Performance in published studies varies substantially, and many models are trained on small or unrepresentative datasets. Before clinical deployment, systems need external validation, transparent governance, privacy protection, bias testing and clarity about who acts on an alert. AI-generated interpretations should not be treated as autonomous diagnostic decisions.

7.4 Virtual reality and digitally delivered therapy

VR-assisted psychological interventions are an active research area, particularly for paranoia, social avoidance and functional practice. Digital CBT and remote therapy can improve access for some people. The key question is not whether the technology is novel, but whether it produces durable improvements in distress, functioning and engagement without worsening isolation or creating unacceptable privacy risks.

7.5 Psychedelics: a research question, not a psychosis treatment

Important safety distinction
Classic psychedelics and related psychedelic-assisted therapies should not currently be described as a treatment for schizophrenia or active psychosis. People with schizophrenia-spectrum disorders have historically been excluded from trials because of concern about precipitating or worsening psychosis. A 2024 systematic review questioned whether exclusion must be absolute in all future research, but also found limited, low-quality evidence and recommended a conservative approach. Any future study in this population belongs in tightly controlled specialist research, not routine clinical practice.

7.6 New pharmacology beyond dopamine D2 blockade

The field is moving beyond the assumption that all effective antipsychotic treatment must primarily work through dopamine D2 receptor antagonism. New mechanisms—including muscarinic, glutamatergic and other targets—are reshaping research and, internationally, beginning to influence the treatment landscape. Their eventual UK role will depend on licensing, NICE appraisal, comparative effectiveness, safety, cost and real-world experience. They are best presented as a rapidly developing pharmacology pipeline rather than immediate replacements for established treatment.

8. A Practical Integrated Treatment Pathway

Phase

Priorities

1. Immediate assessment

Safety, capacity, safeguarding, intoxication/withdrawal, delirium/neurology, physical examination, substance use, affective symptoms, medication history and collateral information.

2. Engagement and formulation

Explain uncertainty; agree goals; involve family/carers with consent; understand trauma, culture, neurodevelopmental factors and practical stressors.

3. Acute treatment

Antipsychotic where indicated; treat mood/catatonia/substance withdrawal/medical causes as appropriate; minimise coercion; begin psychological support and sleep restoration.

4. Early recovery

CBTp, family intervention, physical-health plan, substance-use work, education/employment support, medication review and early-warning-sign plan.

5. Maintenance

Lowest effective treatment burden consistent with stability; annual medication and physical-health review; LAI if preferred/beneficial; address cognition, relationships, housing and meaningful activity.

6. Persistent symptoms

Re-check diagnosis, adherence, substance use and adequate trials; initiate clozapine promptly when criteria are met; use complex-psychosis rehabilitation expertise if needed.

7. Long-term recovery

Shared decision-making, relapse prevention, social and occupational recovery, primary/secondary care coordination, physical-health protection and a clear route back to specialist help.

9. Common Misconceptions

Misconception

More accurate 2026 framing

“Psychosis means schizophrenia.”

Psychosis is a syndrome with many psychiatric, substance-related, neurological and medical causes.

“Medication is the whole treatment.”

Medication is often important, but psychological, family, vocational, housing, substance-use and physical-health interventions materially affect outcomes.

“Atypicals are simply better than typicals.”

Differences between individual drugs and side-effect profiles matter more than the broad class label; clozapine is a special efficacy case.

“LAIs are only for people who refuse tablets.”

LAIs are a legitimate shared-choice formulation and may suit convenience, preference or relapse-prevention goals.

“Nothing can be done about antipsychotic weight gain.”

Early monitoring, lifestyle intervention, drug choice and increasingly proactive pharmacological prevention strategies can reduce harm.

“Trauma therapy is always unsafe in psychosis.”

With appropriate assessment, pacing and expertise, co-occurring PTSD can be treated; trauma-informed care should be routine.

“AI can now predict relapse reliably.”

Promising signals exist, but real-world validation, bias, privacy and workflow integration remain unresolved.

“Psychedelics are a new treatment for psychosis.”

No. This is experimental research territory and may carry psychosis risk; routine use is not supported.

10. When Urgent or Emergency Assessment Is Needed

Psychosis can become an emergency when there is immediate danger, severe behavioural disturbance, inability to care for basic needs, severe self-neglect, suicidal intent, violent intent, rapidly fluctuating consciousness, suspected delirium, severe intoxication/withdrawal, catatonia, severe medication toxicity, or a serious physical-health problem. New-onset psychosis with fever, seizures, focal neurological signs, marked confusion or autonomic instability requires urgent medical assessment for organic causes.

Conclusion

The central message of modern psychosis treatment is integration. Antipsychotics remain essential for many people, but outcomes improve when treatment also protects physical health, begins early, includes psychological and family interventions, treats substance use, restores social roles and supports autonomy. Clozapine should be used when treatment resistance is established rather than indefinitely postponed. New technologies and novel pharmacology are widening the field, but the standard for adoption should remain the same: demonstrable benefit, acceptable harm, patient choice, equity and real-world clinical value.

Selected Current References and Guidance

  1. NICE. Psychosis and schizophrenia in adults: prevention and management (CG178). Current recommendations; guideline reviewed July 2025. Link
  2. NICE. July 2025 exceptional surveillance of CG178: metacognitive therapy. Link
  3. NICE. Rehabilitation for adults with complex psychosis (NG181). Link
  4. NICE. Psychosis and schizophrenia in adults quality standard (QS80). Link
  5. World Health Organization. Schizophrenia fact sheet. 6 October 2025. Link
  6. American Psychiatric Association. The APA Practice Guideline for the Treatment of Patients With Schizophrenia. Link
  7. MHRA. Clozapine and other antipsychotics: monitoring blood concentrations for toxicity. Link
  8. Carolan A, et al. Metformin for the Prevention of Antipsychotic-Induced Weight Gain: Guideline Development and Consensus Validation. Schizophrenia Bulletin. 2025. Link
  9. Hau C, et al. Smartphone monitoring and digital phenotyping apps for schizophrenia: A review of the academic literature. Schizophrenia Research. 2025. Link
  10. Digital phenotyping for predicting relapse in psychiatric disorders: a systematic review of passive sensing approaches. 2026. Link
  11. Tay JL, et al. Accuracy of machine learning methods in predicting prognosis of patients with psychotic spectrum disorders: a systematic review. BMJ Open. 2025. Link
  12. Reconsidering evidence for psychedelic-induced psychosis: systematic review and meta-analysis. 2024. Link

About this revision
This document is an updated and reorganised version of the supplied Neurohaven page “Best Practices in Psychosis Treatment”. The source page already covered antipsychotic treatment, CBTp, family work, trauma-informed care, community rehabilitation, early intervention, lifestyle care, major guidelines and emerging technology. The 2026 revision retains that breadth while correcting overstatements, updating guideline status and strengthening physical-health, clozapine, digital-governance and experimental-treatment cautions.