Generalised Anxiety Disorder (GAD) — Guideline-Based Treatment: A Comprehensive Review

Overview and Diagnosis

NICE Guidelines (CG113) — Stepped Care Model

Step 1 — Identification and Recognition

• Identify and communicate the diagnosis clearly.
• Psychoeducation about the nature of anxiety.
• Active monitoring for patients with mild symptoms or who do not want immediate treatment.

Step 2 — Low-Intensity Interventions (Mild–Moderate GAD)

• Low-intensity CBT-based self-help (guided self-help programmes, bibliotherapy).
• Psychoeducational groups based on CBT principles.
• Facilitate access to Improving Access to Psychological Therapies (IAPT) services.

Step 3 — High-Intensity Interventions (GAD with Marked Functional Impairment, or Failure of Step 2)

• High-intensity psychological treatment: individual CBT (12–15 sessions, 1 hour each) or applied relaxation
• Pharmacological treatment (see below)

Step 4 — Highly Complex/Refractory GAD

• Risk of self-harm/suicide
• Significant comorbidity (substance misuse, personality disorder, complex physical health problems)
• Inadequate response to Step 3 interventions

NICE Pharmacological Recommendations

Maudsley Prescribing Guidelines (15th Edition)

Crisis Management

• Benzodiazepines: short-term only, maximum 2–4 weeks; some authorities argue risks are overstated in certain patients; never use as long-term treatment except in a very small number of severely disabled patients who fail all other options.

First-Line Treatment (in order of preference)

Note: Maudsley places pregabalin in the first-line group (unlike NICE, which places it third), based on its rapid onset and RCT evidence base — but flags substantial misuse potential.

Second-Line Treatment (less well-tolerated or weaker evidence; no order of preference)

Key Maudsley Practical Principles

• SSRIs and SNRIs: modest benefit usually seen within 6 weeks, continuing to increase over time.
• Optimal treatment duration: at least 1 year.
• A recent network meta-analysis found agomelatine or bupropion may be most effective overall; clear placebo effect seen for lorazepam and vortioxetine.
• Efficacy of SSRIs (but not SNRIs) increases across the licensed dose range in anxiety disorders.
• Pregabalin: do not stop abruptly — taper slowly to avoid seizures and severe withdrawal.
• SSRIs: do not stop abruptly — patients with anxiety are particularly sensitive to discontinuation symptoms; taper over several months.
• Monitor all patients on SSRIs for akathisia, increased anxiety, and suicidal ideation, especially those <30 years, comorbid depression, or known higher suicide risk.

CANMAT Guidelines (Katzman et al., 2014 — Anxiety, Trauma-and-Stressor-Related, and OCD Spectrum Disorders)

Pharmacotherapy — Line Recommendations

Key CANMAT departure: Pregabalin and agomelatine are co-first-line with SSRIs/SNRIs in CANMAT, unlike NICE where they are reserved for when SSRIs/SNRIs fail.

CANMAT Psychological Treatment

• CBT (individual or group) — Level 1 evidence, first-line
• Applied relaxation — effective, first-line equivalent
• Acceptance and Commitment Therapy (ACT) — emerging evidence
• Mindfulness-Based Cognitive Therapy (MBCT) — beneficial, especially relapse prevention
• Internet-delivered CBT (iCBT) — increasingly well-evidenced

CANMAT Duration and Monitoring

• Acute phase: 8–12 weeks
• Maintenance: continue for ≥3 months after remission (minimum); many experts recommend 12–24 months given high relapse rates
• Onset of benefit: 2–8 weeks for pharmacological agents; titrate dose every 1–2 weeks as tolerated

BAP Guidelines (British Association for Psychopharmacology — Baldwin et al., 2014)

• First-line: SSRIs and SNRIs (particularly venlafaxine XR, paroxetine, escitalopram, sertraline for GAD)
• Second-line: Pregabalin, buspirone
• CBT endorsed as first-line psychological intervention with same efficacy as pharmacotherapy
• Benzodiazepines: appropriate for acute/short-term use; long-term prescribing should be avoided
• Combination of CBT + pharmacotherapy may be superior to either alone in some patients

WFSBP Guidelines (World Federation of Societies of Biological Psychiatry)

• Category A (Most Evidence): Duloxetine, escitalopram, paroxetine, pregabalin, sertraline, venlafaxine XR
• Category B: Buspirone, hydroxyzine, lorazepam
• Category C: Imipramine, trazodone, quetiapine XR
• Recommended treatment duration: 12 months

Psychopharmacology Algorithm (Davidson et al., 2010 — referenced in Maudsley)

Step 1: SSRI or venlafaxine XR (4–6 weeks)
         ↓ (if inadequate response)
Step 2: Optimise dose (6–8 weeks at maximum tolerated dose)
         ↓ (if still inadequate)
Step 3: Switch to another SSRI/SNRI or pregabalin
         ↓ (if still inadequate)
Step 4: Augment: add pregabalin, buspirone, hydroxyzine, or atypical antipsychotic
         ↓ (if still inadequate)
Step 5: Consider TCA (imipramine), MAOi (phenelzine) in specialist setting

Specific Drug Details by Class

SSRIs

• Start at half the normal antidepressant starting dose to avoid initial anxiogenic effect
• Allow 6–8 weeks for assessment of response
• If partially effective, optimise to maximum tolerated dose before switching

SNRIs

• Same slow-start principle as SSRIs
• Venlafaxine: monitor blood pressure (dose-related hypertension above 150 mg/day)
• Duloxetine: CYP2D6 substrate/inhibitor — watch drug interactions

Pregabalin

• Starting dose: 75–150 mg/day in divided doses
• Target dose: 150–600 mg/day in 2–3 divided doses
• Onset of anxiolytic action: within the first week (faster than SSRIs/SNRIs)
• Mechanisms: binds α2δ subunit of voltage-gated calcium channels → reduces glutamate, norepinephrine, substance P release
• Do not stop abruptly — seizure risk; taper slowly over weeks
• Misuse and diversion potential (Schedule 3 in UK since 2019); caution in patients with substance use disorders
• MHRA 2022 warning: avoid in pregnancy (teratogenicity risk)

Buspirone

• Dose: 15–60 mg/day in divided doses (start 5 mg TID)
• Partial 5-HT1A agonist; not useful in patients previously taking benzodiazepines (patients feel it is inferior)
• Delayed onset: 2–6 weeks
• No dependence or sedation; suitable for patients with substance use disorders

Benzodiazepines (Crisis Use Only)

• For acute, severe, disabling anxiety only
• Maximum 2–4 weeks; use at lowest effective dose
• Diazepam, lorazepam, clonazepam most commonly used
• Not recommended for routine long-term treatment (tolerance, dependence, withdrawal, cognitive impairment)
• NICE: do not use for GAD outside crisis management

Quetiapine XR

• 50–300 mg/day
• Good anxiolytic efficacy; CANMAT second-line, Maudsley second-line
• Metabolic side effects, sedation, weight gain limit use
• Not initiated in primary care (UK guidance)
• Potentially effective in comorbid depression + GAD

Agomelatine

• 10–50 mg/day (at night)
• Melatonin receptor agonist + 5-HT2C antagonist
• Network meta-analysis suggests highest efficacy in GAD
• Prevents relapse over 6 months
• Liver function monitoring required
• Not licensed for anxiety in many countries — off-label use

Hydroxyzine

• 25–100 mg/day in divided doses
• H1 antihistamine with anxiolytic properties
• Rapid onset (sedative mechanism); suitable for short-term use
• Useful where benzodiazepine concerns exist but rapid relief needed

Psychological Treatments

CBT (Cognitive Behavioural Therapy)

• Most evidence-based psychological treatment for GAD
• Individual: 12–15 sessions (weekly, 60 min)
• Components: cognitive restructuring (challenging catastrophic/overestimated worry), behavioural experiments, intolerance of uncertainty work, worry postponement, relaxation
• Approximately 50–70% response rate

Applied Relaxation (Öst model)

• Progressive muscle relaxation → cue-controlled relaxation → applied in vivo
• Comparable efficacy to CBT in some trials (NICE endorses both equally)

ACT (Acceptance and Commitment Therapy)

• Focuses on psychological flexibility, acceptance, mindfulness rather than direct worry challenging
• Growing evidence base; useful where CBT is ineffective

Mindfulness-Based Approaches

• MBSR (Mindfulness-Based Stress Reduction) and MBCT
• Particularly useful for relapse prevention and comorbid depression

Digital and Internet-Delivered CBT (iCBT)

• FDA cleared DaylightRx (2024) — first prescription digital therapeutic for GAD
• Trials: >70% remission; sustained 6+ months
• Prescribed as adjunct to usual care; suitable for motivated patients

Special Populations

Older Adults (CCSMH/CANMAT Seniors Guidelines 2024)

• First-line: Escitalopram, sertraline, venlafaxine (favourable pharmacokinetics)
• Second-line: Duloxetine (CYP2D6 interactions), buspirone
• Avoid: Benzodiazepines (falls, cognitive impairment, paradoxical agitation) except if all first-line options fail (max 2–4 weeks)
• Avoid: Quetiapine routinely (poorly tolerated, metabolic effects)
• Dose: start 50% of usual adult dose; titrate carefully

Pregnancy

• Psychological treatments preferred as first-line
• If pharmacotherapy needed: SSRIs generally preferred (sertraline, escitalopram)
• Avoid pregabalin (MHRA 2022 teratogenicity warning)
• Avoid benzodiazepines (neonatal withdrawal, "floppy infant" syndrome)
• Discuss risks/benefits; document discussion

Comorbid Depression

• SSRIs and SNRIs treat both — preferred pharmacological choice
• CBT adapted for comorbid presentations
• Quetiapine adjunct useful in treatment-resistant cases with comorbid depression

Comorbid Substance Use Disorder

• Avoid benzodiazepines and pregabalin
• SSRIs/SNRIs preferred
• Buspirone: safe, no dependence potential

Treatment-Resistant GAD

1. Augmentation strategies (Maudsley/CANMAT):
   • Add pregabalin to SSRI/SNRI
   • Add quetiapine (note: Maudsley suggests probably not effective as adjunct to SSRI/SNRI; may work better as monotherapy)
   • Add buspirone
   • Add hydroxyzine
2. Switching to a different class (SSRI → SNRI → TCA)
3. Third-line agents (CANMAT): imipramine, mirtazapine, trazodone, divalproex
4. Specialist interventions (see recent systematic review — Schiele et al., Psychother Psychosom 2025, PMID 40946318):
   • Repetitive transcranial magnetic stimulation (rTMS) — emerging evidence (meta-analysis 2024: PMID 39208534)
   • Augmentation with atypical antipsychotics (aripiprazole, risperidone — third-line CANMAT adjuncts)
   • MAOIs (phenelzine) — in specialist setting only

Comparative Guideline Summary Table

Monitoring During Treatment

• Response assessment: at 4–6 weeks (initial response), at 12 weeks (full response evaluation)
• Use validated scales: GAD-7, HAM-A, PSWQ (Penn State Worry Questionnaire)
• If no improvement after 12 weeks at adequate dose → switch or augment
• Monitor for suicidal ideation in first weeks (especially age <30)
• Pregabalin: monitor for misuse, assess at each visit
• Agomelatine: LFTs at baseline, 3, 6, 12 weeks and then periodically
• Venlafaxine: blood pressure monitoring
• Quetiapine: metabolic monitoring (weight, glucose, lipids)

Emerging and Novel Treatments (2024–2025)

• Gepirone (Exxua): FDA-approved 2023 for MDD; trials in GAD ongoing — novel 5-HT1A agonist (buspirone successor)
• MM-120 (LSD d-tartrate): FDA Breakthrough Therapy Designation March 2024 for GAD; Phase 2 single-dose trials showed significant reduction in GAD symptoms lasting 12 weeks; Phase 3 launched late 2024
• DaylightRx: FDA-cleared prescription digital CBT app (2024) — >70% remission in trials; adjunct to usual care
• rTMS: Non-invasive brain stimulation — systematic review (PMID 39208534, 2024) confirms emerging efficacy; not yet routine

References (Key)

• The Maudsley Prescribing Guidelines in Psychiatry, 15th Edition, Box 3.6
• NICE CG113: Generalised anxiety disorder and panic disorder in adults: management (last reviewed May 2024)
• Katzman MA et al. Canadian clinical practice guidelines for the management of anxiety, posttraumatic stress and obsessive-compulsive disorders. BMC Psychiatry 2014;14(Suppl 1):S1
• Baldwin DS et al. Evidence-based pharmacological treatment of anxiety disorders. J Psychopharmacol 2014;28(5):403–439
• Schiele MA et al. Treatment-resistant anxiety disorders [Systematic Review]. Psychother Psychosom 2025 (PMID 40946318)
• Qi L et al. Non-invasive brain stimulation in GAD. J Psychiatr Res 2024 (PMID 39208534)