Mental Health

Depression - diagnosis and management

NICE 2022 substantially revised its approach to depression, recommending psychological therapies ahead of antidepressants for less severe presentations. NZ practice diverges from this in one unavoidable way: the psychological therapies NICE recommends as first-line are largely inaccessible in NZ primary care.

Reviewed May 2026 BPAC NZ / MoH RANZCP CPG 2015 NICE NG222 2022

Scope: unipolar depression in adults in primary care and community settings. Excludes bipolar disorder, psychotic depression, perinatal depression (separate guideline), and child/adolescent depression.

Where they agree

PHQ-9 is a validated and recommended tool for screening, diagnosis, and monitoring treatment response
SSRIs are the recommended first-line antidepressant class for moderate-to-severe depression
Physical activity has a well-evidenced role in reducing depressive symptoms and should be discussed at every encounter⁷
Continue antidepressants for at least 6 months after remission; after two or more episodes, discuss long-term maintenance therapy
Antidepressants should be tapered gradually when stopping, not discontinued abruptly; discontinuation symptoms can be severe and prolonged⁶
Assess suicide risk at initial presentation and at key transitions (starting antidepressants, returning after a missed appointment, when not improving)
Routinely assess for and treat comorbid conditions, particularly anxiety, alcohol use, chronic pain, and sleep disturbance

Where they diverge

NICE 2022 recommends psychological therapy (individual guided self-help, CBT, behavioural activation) as first-line for less severe depression, before antidepressants; NZ primary care reaches for antidepressants earlier due to severely limited psychology access
AU Better Access provides up to 10 Medicare-subsidised psychology sessions per year; NZ has no equivalent - the access gap is the defining practical difference between NZ and AU care
NICE 2022 uses a functional impairment framework rather than a simple mild/moderate/severe severity classification; NZ and AU continue to use severity-based stepped care language
NICE 2022 explicitly states antidepressant discontinuation symptoms can be "severe and prolonged" and recommends tapering over months; NZ and AU guidance has been slower to reflect this evidence
Newer antidepressants (agomelatine, vortioxetine) with evidence of efficacy are not funded in NZ

Comparison

Domain	NZ - BPAC / MoH	AU - RANZCP CPG	International - NICE NG222 2022
Diagnosis and severity classification	Clinical diagnosis based on DSM-5 or ICD-11 criteria: five or more depressive symptoms for ≥2 weeks, including depressed mood or anhedonia. PHQ-9 used to screen and monitor severity (mild 5-9, moderate 10-14, moderately severe 15-19, severe 20-27). BPAC emphasises that PHQ-9 alone is not a diagnosis; clinical context, duration, and functional impairment are essential components of the assessment.	Similar. RANZCP recommends a thorough biopsychosocial assessment including medical history, medication review, substance use, social stressors, and family history. Severity classification broadly consistent with DSM-5 (mild, moderate, severe) with functional impairment as a key determinant of treatment intensity. Always assess for bipolar spectrum features before commencing antidepressants.	NICE NG222 moved away from categorising depression as mild/moderate/severe as the primary treatment-decision framework. Instead uses a combination of symptom count and degree of functional impairment to determine treatment step. Acknowledges that many people fall between diagnostic categories and that a pragmatic, shared-decision approach to treatment intensity is preferable to rigid threshold-based prescribing.
First-line treatment - less severe depression	Stepped care model: active monitoring, psychoeducation, lifestyle advice (sleep, exercise, alcohol), and self-help resources as initial steps for mild presentations. In practice, antidepressants are offered at or near the first presentation in NZ primary care far more often than guideline sequencing implies, because funded psychological therapy is not reliably available. This is a system constraint, not a guideline recommendation.Access limited	Stepped care with more realistic access to psychological therapy. RANZCP recommends psychological therapy (CBT, behavioural activation, interpersonal therapy) for mild-to-moderate depression, with antidepressants alongside or after if therapy alone is insufficient. Better Access (Medicare) provides up to 10 subsidised sessions per year with a psychologist or mental health social worker under a GP mental health care plan.Better Access	NICE NG222 made a significant 2022 change: for people with less functional impairment, the preferred first-line treatments are: 1. Individual guided self-help (low-intensity CBT-based) 2. Group CBT or group behavioural activation 3. Individual CBT or behavioural activation Antidepressants are not the recommended first treatment for less severe depression unless the person prefers them, has a history of moderate-to-severe depression, or psychological therapies are not accessible or acceptable. This represents a more conservative position on antidepressants than previous NICE guidance.
First-line treatment - moderate to severe depression	For moderate-to-severe depression: an SSRI plus active support. Refer to a mental health service if: high suicide risk, psychotic features, severe functional impairment, inadequate response to two antidepressant trials, or diagnostic uncertainty. Offer a mental health care plan. Document shared decision-making around treatment choice. Set a clear review date (2-4 weeks after starting an antidepressant).	RANZCP: combined pharmacotherapy and psychological therapy for moderate-to-severe depression produces better outcomes than either alone. An SSRI as the antidepressant of first choice, alongside a referral for psychological therapy under a mental health care plan (GP Mental Health Treatment Plan). Consider inpatient or crisis team referral for: psychotic depression, high suicide risk, severe self-neglect, or treatment-refractory presentations.	NICE NG222: for people with more significant functional impairment, offer combined antidepressant and high-intensity psychological therapy (individual CBT, behavioural activation, interpersonal therapy, or mindfulness-based CBT), or antidepressant alone if therapy is not accessible or preferred. NICE does not recommend antidepressant monotherapy without discussion of psychological therapy as a concurrent option.
First-line antidepressant choice	SSRIs are first-line. Funded options (no access restrictions): Citalopram 20-40mg dailyFunded Fluoxetine 20-60mg dailyFunded Paroxetine 20-50mg dailyFunded (avoid in pregnancy; highest discontinuation risk of all SSRIs) Sertraline 50-200mg dailyFunded Second-line funded: venlafaxine (SNRI)Funded, mirtazapineFunded. Escitalopram and duloxetine: verify current PHARMAC funding status.Check PHARMAC	Same SSRI first-line approach. PBS options include all the SSRIs listed plus escitalopram and duloxetine.PBS RANZCP notes that network meta-analysis evidence places escitalopram and sertraline among the best-tolerated and most effective SSRIs as a class.⁵ No single SSRI is clearly superior; choice should be guided by side-effect profile, prior response, drug interactions, and patient preference.	NICE NG222: do not recommend one antidepressant over another as a class. Choice should be based on individual factors: prior response, contraindications, drug interactions, discontinuation profile, and patient preference including concerns about weight gain or sexual dysfunction. NICE notes the Cipriani et al 2018 Lancet network meta-analysis found all 21 antidepressants are more effective than placebo, with modest differences in efficacy between drugs.⁵ Sertraline and escitalopram have a favourable efficacy/tolerability balance for most patients.
Access to psychological therapy	Funded access to talking therapy in NZ primary care is severely limited and varies significantly by region: Primary Mental Health and Addiction (PMHA) initiative: provides brief interventions (typically 6 sessions) through PHO-based counsellors; not universally available; wait times vary widely. EAP services: available for employed people through workplace programmes; not universal. Clearhead, Calm, Groov: free digital CBT tools (mixed evidence; suitable adjunct, not a substitute for therapy). Private psychology: $150-250/session; not subsidised.No equivalent to Better Access	Better Access initiative: up to 10 Medicare-subsidised sessions per calendar year with a psychologist (or 10 with a mental health occupational therapist or social worker) under a GP Mental Health Treatment Plan.10 sessions/year Gap payments apply (psychologists set their own fees; Medicare rebates $136.35 per session as of 2024; out-of-pocket gap varies). Head to Health provides digital mental health resources. Community mental health centres in most states.	NICE NG222 was written assuming psychological therapy is accessible. The guideline explicitly recommends CBT, behavioural activation, interpersonal therapy, mindfulness-based CBT, and group-based therapies as first-line options. NICE does not address what clinicians should do when these therapies are unavailable. The gap between NICE guidance and NZ access reality is the most significant divergence for NZ primary care practitioners reading this guideline.
Duration of treatment and maintenance	Continue antidepressants for at least 6 months after remission for a first episode. After two or more episodes: discuss long-term maintenance, typically at least 2 years. Review the decision annually. BPAC: frame maintenance conversations around relapse prevention, not indefinite medication. Many patients can eventually discontinue safely with gradual tapering.	RANZCP: same 6-month post-remission minimum for a first episode. After recurrent episodes, the benefit-risk balance shifts toward long-term maintenance. Consider patient preference, residual symptoms, and functional recovery in the decision. Psychological therapy (particularly mindfulness-based CBT) has a role in relapse prevention as an alternative or adjunct to ongoing pharmacotherapy.	NICE NG222: continue for at least 6 months after remission. For recurrent depression (3 or more episodes, or significant residual symptoms): discuss long-term pharmacotherapy or mindfulness-based CBT as a relapse prevention strategy. NICE places greater emphasis on actively discussing the long-term plan, including eventual discontinuation, at each review.
Antidepressant discontinuation and tapering	Taper antidepressants gradually; do not stop abruptly. BPAC advises reducing the dose over several weeks to months, depending on duration of treatment and sensitivity of the individual. Warn patients that discontinuation symptoms (dizziness, nausea, flu-like symptoms, electric shock sensations, irritability, insomnia) are common and do not indicate relapse. Paroxetine and venlafaxine have the highest discontinuation risk due to short half-life and noradrenergic effects. Fluoxetine's long half-life makes it easier to discontinue.	RANZCP: taper over weeks to months; acknowledge that some patients require very gradual reduction. Switching to fluoxetine before tapering can ease discontinuation from shorter half-life SSRIs. Psychological support during discontinuation is recommended where available.	NICE NG222 made a landmark change to discontinuation guidance. The 2022 version explicitly states that discontinuation symptoms can be "severe and prolonged" and recommends tapering over months, with hyperbolic dose reductions (larger steps at higher doses, smaller steps at lower doses) in people who have been on treatment long-term.⁶ This represents a direct acknowledgement that the previous NICE position (4-week taper) understated the clinical problem, based on a systematic review by Davies and Read (2019).⁶ NZ guidance has been slower to reflect this explicitly.
Treatment-resistant depression	Defined as failure to respond to two adequate antidepressant trials (adequate dose for 6 weeks). Options: Switch to a different class (SSRI to SNRI, or to mirtazapine); augment with lithiumFunded; augment with low-dose quetiapine or aripiprazoleFunded; ECT for severe, life-threatening, or medication-refractory presentations.Funded via specialist Esketamine (Spravato): not funded NZ.Not funded Refer to psychiatry at this stage.	Similar augmentation pathway. RANZCP recommends structured assessment before labelling treatment-resistant: confirm diagnosis, check adherence and drug interactions, address comorbidities, consider psychosocial factors. Augmentation options include lithium, atypical antipsychotics, and thyroid hormone. Esketamine (Spravato): TGA-approved for treatment-resistant depression; available through specialist centres with significant out-of-pocket cost (not PBS-subsidised).TGA approved, not PBS	NICE NG222: after two failed antidepressant trials, offer: switching to another antidepressant, combining antidepressants (with monitoring), augmenting with lithium or an atypical antipsychotic, or adding high-intensity psychological therapy if not already provided. ECT for severe, life-threatening, or rapidly deteriorating depression unresponsive to medication. NICE does not recommend esketamine as a routine option and notes limited long-term safety and efficacy data.
Special populations - pregnancy and postpartum	Untreated depression in pregnancy carries significant risks (preterm birth, low birthweight, impaired bonding, maternal suicide). The risk of untreated depression usually outweighs antidepressant risk for moderate-to-severe presentations. Fluoxetine has the most pregnancy safety data.Funded Avoid paroxetine in the first trimester (cardiac septal defects, though absolute risk is low). All SSRIs carry risk of neonatal adaptation syndrome (transient, not teratogenic). Discuss risks and document consent. Refer to O&G or perinatal mental health if uncertain.	RANZCP: same hierarchy. Sertraline and fluoxetine are generally preferred in pregnancy. Collaborative care with obstetrics. Postpartum: SSRIs are compatible with breastfeeding (sertraline has the lowest infant exposure via breast milk of the commonly used SSRIs).	NICE NG222: always discuss with the woman the risks of both treating and not treating. Psychological therapies are preferred first-line in pregnancy if accessible and acceptable. Where antidepressants are required: use the lowest effective dose; avoid paroxetine in the first trimester where possible; monitor for neonatal adaptation syndrome. NICE provides detailed guidance in a separate perinatal mental health guideline (NG201).
Special populations - older adults	Start low, go slow. Use lower starting doses (typically half the usual adult starting dose). Citalopram and escitalopram: avoid doses above 20mg daily in older adults due to QTc-prolonging effects. Check for falls risk (SSRIs increase falls risk, as does untreated depression). Screen for cognitive impairment - depression and early dementia coexist frequently. Drug interactions are more common with polypharmacy. Consider stopping a potentially depressogenic drug (beta-blockers, steroids, some antihypertensives) before adding an antidepressant.	RANZCP: same principles. Mirtazapine is useful in older adults with poor appetite and sleep disturbance. SNRIs (venlafaxine, duloxetine) preferred over TCAs. TCAs are effective but poorly tolerated and carry significant falls, arrhythmia, and anticholinergic risks in older people. Avoid amitriptyline as an antidepressant in older adults (still has a role in neuropathic pain at low dose).	NICE NG222: same pharmacokinetic cautions. NICE recommends the same stepped care model for older adults, adapted for comorbidity and polypharmacy. Physical activity and social engagement are particularly important non-pharmacological components in older adults. Do not assume that low mood in older people is an inevitable part of ageing.

NZ clinical context

PHARMAC funding and population-specific considerations

The psychology access gap: what this means in practice

NICE NG222 is a well-constructed guideline, but it was written for a system in which psychological therapy is accessible. In NZ primary care, it largely is not. The available routes are:

PMHA (Primary Mental Health and Addiction) services: Brief intervention counselling, usually 3-6 sessions, available through most PHOs. Wait times vary from weeks to months. Not available in all regions. The He Ara Oranga inquiry (2018) recommended major expansion of these services, and funding has increased, but access remains uneven.²
Employee Assistance Programmes (EAP): 3-6 sessions funded through employers. Only accessible to people in employment; not available to the unemployed, those on benefits, or those in informal employment.
Free digital resources: Clearhead, Calm, Groov, and the Staying on Track programme offer CBT-based digital self-help at no cost. These are genuine adjuncts and appropriate for mild depression with low risk, but they are not a substitute for a therapeutic relationship.
Private psychology: $150-250 per session. Without insurance, a 12-session course of CBT costs $1800-3000. This is not accessible to a significant proportion of the NZ population.

The practical implication is that for many NZ patients with mild-to-moderate depression, an antidepressant will be the most realistic first treatment offered. This is not a guideline recommendation - it is a system reality. Where psychology access is genuinely available, it should be prioritised for mild-to-moderate depression in line with the NICE NG222 evidence base.

Funded antidepressants: what is and is not available

The following are funded without access restrictions:

Citalopram 20-40mg: note the cardiac (QTc) risk at higher doses; maximum 20mg in patients over 65, patients on QTc-prolonging drugs, or those with hepatic impairment.
Fluoxetine 20-60mg: longest half-life (easiest to discontinue); preferred in pregnancy; activating (better for hypersomnic presentations; less good for anxiety-dominant presentations).
Paroxetine 20-50mg: effective but the highest discontinuation risk of the funded SSRIs and the most drug-drug interactions (strong CYP2D6 inhibitor). Avoid in pregnancy.
Sertraline 50-200mg: favourable efficacy/tolerability profile; lowest breast milk transfer; a sensible default for most presentations.
Venlafaxine (SNRI): useful for depression with comorbid anxiety or neuropathic pain; blood pressure monitoring required at higher doses.
Mirtazapine: sedating (take at night); good for insomnia, poor appetite, and agitation-dominant presentations; weight gain is common; useful augmentation agent.
Amitriptyline and nortriptyline: effective antidepressants but poorly tolerated at therapeutic doses; reserve for neuropathic pain at low dose, or under specialist guidance.

Escitalopram, duloxetine, agomelatine, and vortioxetine: verify PHARMAC status before prescribing. Agomelatine and vortioxetine are likely unfunded; they have evidence of efficacy but no clear superiority to funded agents in most presentations.⁵

Maori and Pacific mental health: prevalence, access, and cultural safety

Depression is more prevalent among Maori and Pacific peoples and more likely to present at greater severity due to delayed engagement with primary care. Key considerations:

Stigma around mental health is common across many Pacific communities and in some Maori communities. The word "depression" may not resonate; asking about wairua (spiritual wellbeing), whakama (shame), or how the person is sleeping, eating, and feeling in their body may open conversations that a direct PHQ-9 does not.
Kaupapa Maori mental health services (eg Tui Ora, Te Rau Ora) offer culturally grounded therapeutic approaches. Knowing what is available in your region and making warm referrals is more effective than handing out a list of services.
Financial hardship, housing instability, and discrimination are major contributors to depression in these populations. Antidepressants address neurobiology; they do not address social determinants. A mental health care plan that ignores the social context is incomplete.
Pacific peoples are also at disproportionate risk for metabolic side effects (weight gain) from antidepressants and atypical antipsychotics used in augmentation. This should factor into drug choice and monitoring.

Antidepressant discontinuation: the clinical reality

Discontinuation symptoms are common, frequently confused with relapse, and often underestimated in duration. The Davies and Read 2019 systematic review found that 56% of people experience discontinuation symptoms when stopping antidepressants, and 46% of those describe them as severe.⁶ The previous clinical assumption that a 4-week taper is adequate for most patients is not evidence-based.

In practice, a patient who has been on an SSRI for 2 or more years should be offered:

A hyperbolic taper: larger dose reductions early in the taper, smaller reductions at lower doses (because antidepressant receptor occupancy follows a hyperbolic, not linear, dose-response curve)
A timeline measured in months, not weeks
The option to pause and hold at any dose that feels difficult before continuing reduction
Clear framing: discontinuation symptoms (dizziness, electric shock sensations, irritability, insomnia) are not relapse; they resolve within days to weeks of the dose stabilising. Relapse comes on gradually and persists.

Fluoxetine is the most forgiving SSRI to stop because its long half-life (including active metabolite norfluoxetine) effectively provides its own slow taper. Paroxetine is the most difficult. Switching to fluoxetine before tapering off a short half-life SSRI is a practical strategy in patients who have struggled to stop.

Bottom line for NZ practice

What this means in the consulting room

Use PHQ-9 to screen and monitor, but diagnose clinically. A PHQ-9 of 8 in someone who is otherwise functioning well is different from a PHQ-9 of 8 in someone who has stopped leaving the house. Severity is about functional impairment, not score alone.
For mild depression with adequate function: active monitoring, psychoeducation, and brief self-help (digital tools, exercise prescription, sleep hygiene) are legitimate first steps. NICE NG222 is right that antidepressants are not the default here. Book a 2-week follow-up and use that contact as a decision point, not an afterthought.
When starting an SSRI: sertraline or citalopram are sensible defaults for most adults. Review at 2-4 weeks for tolerability, 6-8 weeks for efficacy. If no response at 8 weeks on an adequate dose, switch or augment - do not simply continue indefinitely.
Before stopping an antidepressant: plan the taper collaboratively with the patient. For anyone on treatment for more than a year, months not weeks. Make sure they know what discontinuation symptoms feel like and how to distinguish them from relapse - this conversation prevents most of the distress and unnecessary re-prescribing that follows abrupt discontinuation.
If a patient is clearly appropriate for psychological therapy and funded access is not available: document what you referred to and why alternatives were not accessible. Consider a formal mental health care plan even if the immediate pathway is pharmacotherapy - it frames the approach as more than "here are some pills" and sets a review structure.

Sources

BPAC NZ. Depression in adults. Available from: bpac.org.nz [accessed 2025].
New Zealand Government. He Ara Oranga: report of the Government Inquiry into Mental Health and Addiction. Wellington: Ministry of Health; 2018.
Malhi GS, Bell E, Bassett D, et al. The 2020 Royal Australian and New Zealand College of Psychiatrists clinical practice guidelines for mood disorders. Aust N Z J Psychiatry. 2021;55(1):7-117.
National Institute for Health and Care Excellence. Depression in adults: treatment and management (NG222). London: NICE; 2022. Available from: nice.org.uk
Cipriani A, Furukawa TA, Salanti G, et al. Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder: a systematic review and network meta-analysis. Lancet. 2018;391(10128):1357-66.
Davies J, Read J. A systematic review into the incidence, severity and duration of antidepressant withdrawal effects: are guidelines evidence-based? Addict Behav. 2019;97:111-21.
Kvam S, Kleppe CL, Nordhus IH, Hovland A. Exercise as a treatment for depression: a meta-analysis. J Affect Disord. 2016;202:67-86.
Bloch MH, Hannestad J. Omega-3 fatty acids for the treatment of depression: systematic review and meta-analysis. Mol Psychiatry. 2012;17(12):1272-82.