The 7 Ds of Depression Stephen Bazire, Chief Pharmacist Norfolk and Waveney Mental Health Partnership NHS Trust.

The 7 Ds of Depression

Stephen Bazire,Chief Pharmacist

Norfolk and Waveney Mental Health Partnership NHS Trust

Declarations of interest (last 3 years)

AstraZeneca – lectures, Bipolar Conference 2005 BMS/Otsuka - Abilify consultancy, lectures Cephalon - one day at ECNP 2004 Denfleet – 2 advisory boards Janssen-Cilag - ECNP 2005, 2006; lectures; APA 2004; CINP

regional 2006 Lilly – lectures, chair meetings Lundbeck - (books) Organon - Advisory Boards, lectures, (books) Sanofi-Synthelabo - lectures, thermal mug Servier – Advisory Board Wyeth – Advisory Board, lectures

UK Psychiatric Pharmacy Group User groups, Norfolk Mental Health Alliance, NICE

Depression

Diagnosis Does it matter if you treat? Drug-induced Drug Dose Duration Discontinuation

Depression - diagnosis

A) Five of the following: Weight changes Sleep disturbance Loss of interest/pleasure Depressed mood Psychomotor agitation or

retardation Recurrent thoughts of death or

suicide Fatigue or loss of energy Feelings of worthlessness Poor concentration

Not organic Not normal

reaction to death of a loved one

No delusions or hallucinations in absence of mood symptoms

Not superimposed on schizophrenia

Is depression on the rise?

Stirling County Study (Atlantic Canada) Same diagnostic criteria used Representative adult population samples

1952, n=1003 1970, n=1201 1992, n=1396

Current incidence 5% for MDD in all years Increased rate in younger women

(Murphy et al, Arch Gen Psych 2000, 57, 209-15)

Lundby study suggested rise in depression in 1950s has levelled off

Higher incidence in women (n=3563, prospective longitudinal study, Mattisson et al, J Aff Dis 2005, 87, 151-60)

Underuse of antidepressants

Survey 1988-1994 (USA) n=7589, aged 17-39 4.1% (312) had DSM-III major depression

Only 7.4% of these were having antidepressants “Telling prescriber the symptoms increased

chance of having an antidepressant 10-fold”

Druss et al, J Clin Psych 2000, 61, 234-7

Suicide

Suicide is a fatal outcome of psychiatric illness Suicide practically does not occur without the presence of

mental illness, most commonly depression, then alcoholism.

Depressed individuals who have committed suicide are seldom treated with antidepressants

Does increased antidepressant use reduces suicide?

14 studies say yes, 2 say no Effect may be even greater in bipolar disorder

where the lithium effect is greater Göran Isaacson, Acta Psychiatr Scand 2006;114:149-50

Effects of treating depression

Sweden: Two year educational programme (GPs on Gotland) Increased antidepressant use Reduced referral, sick leave and in-patient days for depression Significantly reduced suicide

(Rutz et al, Acta Psych Scand 1989;80:151-4) Sweden: Annual on-going educational programme (GPs in Jämtland county) Antidepressant use increased from 25% below national average to the

same level Suicide decreased to the national average

(Henriksson and Isacsson, Acta Psychiatr Scand 2006;114:159-67)Denmark: Suicide rate (1995-1999) has dropped in all groups More markedly in people prescribed SSRIs or older antidepressants

(n=438,625) Compared to those not treated with antidepressants (n=1,199,057)

(4yrs, Søndergård et al, Acta Psychiatr Scand 2006;114:168-76)

The effect of antidepressants on suicide rates in USA

1985-1999 analysis: 20,000 pharmacies in USA Volume of prescriptions for antidepressants Overdose data Alcohol use per capita Employment

Suicide rate fell 13.5% from 12.38 to 10.71/100K Greater fall in women Antidepressant prescriptions rose from 35m to 144m pa 69% of increase was SSRIs

(Grunebaum et al, J Clin Psych 2004;65:1456-62).

10% increase in SSRIs reduced suicide rate by 1.4%

10% increase in 2nd genn reduced suicide rate by 1.2%

Counselling vs antidepressants

Mild to moderate depression, community Antidepressants vs generic counselling 4 treatment groups:

Randomised to antidepressants or counselling (n=103)

Patient preference to antidepressants or counselling (n=103)

Counselling vs antidepressants

Mild to moderate depression, community Antidepressants vs generic counselling 4 treatment groups:

Randomised to antidepressants or counselling (n=103) Patient preference to antidepressants or counselling

(n=103) No outcomes difference between groups (!)

Beck scores, Psychiatrists assessment Patients choosing counselling did slightly better than those

randomised to it Both seem equally effective in mild-to-moderate

depression (n=323, RCT, 8/52+12/12, Chilvers et al, BMJ 2001, 322, 722-75)

Drug-induced depression

Over 150 drugs reported to cause depression e.g:

Alcohol Benzodiazepines

e.g diazepam, clonazepam, temazepam, lorazepam

Antipsychotics Anticonvulsants

e.g. carbamazepine, lamotrigine, levetiracetam, pregabalin, topiramate

Anti-parkinsonian drugs Anticholinergics

H2 blockers Interferons (controversial) NSAIDs eg ibuprofen

Cardiovascular drugs e.g. beta-blockers, calcium

channel-blockers Antibiotics (rare) Baclofen (rare) Steroids (e.g. dexamethasone) Caffeine/caffeine withdrawal Oral contraceptives Simvastatin Dantrolene Tizanidine

Check doses, starting, stopping, previous

historyRef Psychotropic Drug Directory 2007, SPCs, BNF

Antidepressants available in UK

SSRIs: Citalopram (Cipramil), escitalopram (Cipralex), fluoxetine (Prozac), fluvoxamine (Faverin), paroxetine (Seroxat), sertraline (Lustral)

Mirtazapine (Zispin) Venlafaxine (Efexor) Tricyclics: amitriptyline, clomipramine (Anafranil),

dothiepin/dosulepin, doxepin (Sinequan), lofepramine (Gamanil), imipramine, maprotiline, nortriptyline, trimipramine (Surmontil)

Duloxetine (Cymbalta) Trazodone (Molipaxin) Reboxetine (Edronax) Moclobemide (Manerix) MAOIs: Phenelzine, isocarboxazid, tranylcypromine Mianserin, tryptophan, flupenthixol (Fluanxol) Agomelatine (2008), St. John’s wort

Comparative side effects of antidepressants

Anti-cholin-ergic

Cardiac Nausea Sed- ation

Over-dose

Pro-convuls-ant

Sexual dys- function

Tricyclics +++ ++ + ++ ++ + ++

(Es)citalopram (Cipramil/ Cipralex)

O O ++ O O O ++

Fluoxetine O O ++ O O O ++ Paroxetine (Seroxat) O O ++ O O O +++

Sertraline (Lustral) O O ++ O O O ++

Mirtazapine (Zispin) O O O ++ O O O

Reboxetine (Edronax) + + + O O O O

Duloxetine (Cymbalta) O O ++ O ? ? ++

Trazodone (Molipaxin) + + +++ ++ + O ++

Venlafaxine (Efexor) O ++ +++ + ? + ++

Bupropion (Zyban) + O + O ++ +++ O

MAOIs ++ ++ ++ O/+ ++ O +

Agomelatine (TBA)

Modes of action and receptors

SSRIs - 5-HT reuptake inhibitionMirtazapine - increased 5-HT and NE availability, 5-HT2 and 5-

HT3 antagonismVenlafaxine and duloxetine - 5-HT and NA reuptake inhibition

(venlafaxine variable, duloxetine similar)

Trazodone - 5-HT reuptake inhibition and some receptor antagonism

Tricyclics - 5-HT and NE reuptake inhibitionReboxetine - noradrenaline reuptake inhibitionFlupenthixol - Autoreceptor inhibitionMoclobemide - Reversible MAO-A inhibitionMAOIs - Inhibition of MAO-A and MAO-B enzymesAgomelatine - 5HT2C/2B antagonist and melatonin M1/2 agonist

Relevance: as long as it works, side effects

Selected antidepressant side effects

Anticholinergic – dry mouth, blurred vision, constipation Cardiac – prolonged QTc, postural hypotension, tachycardia, Nausea – initial, start with lower doses Sedation - mostly histaminergic effect Overdose toxicity – cardiac Pro-convulsant – bupropion at >300mg/d Sexual dysfunction – lower libido, ED, anorgasmia Anxiety (short-term esp. with SSRIs), appetite changes,

hyponatremia (except mirtazapine), diarrhoea, headache, sweating (esp. at night)

Many can be minimised by starting at lower doses

Usual therapeutic doses for depression

SSRIs: Citalopram (Cipramil) 20-40mg Escitalopram (Cipralex) 10mg Fluoxetine (Prozac) 20mg fluvoxamine (Faverin) 150-300mg? Paroxetine (Seroxat) 20-30mg Sertraline (Lustral) 50-100mgTricyclics: amitriptyline, clomipramine

(Anafranil), dothiepin/dosulepin, doxepin (Sinequan), imipramine, nortriptyline, trimipramine (Surmontil) – 125-150mg/d

Lofepramine (Gamanil) 140-210mg

Newer: Mirtazapine (Zispin) 30-45mg Venlafaxine (Efexor) 75-225mg Duloxetine (Cymbalta) 60-

120mg Trazodone (Molipaxin) 150mg? Reboxetine (Edronax) 8-12mg Moclobemide (Manerix) 300mgMAOIs: Phenelzine 45mg? Isocarboxazid 30mg? Tranylcypromine 30mg? Mianserin, tryptophan,

flupenthixol (Fluanxol), agomelatine (soon)

St. John’s wort

Onset of action of antidepressants

“Antidepressants take 4 weeks to work”

Wrong! 23% of all drug-placebo differences occur within

the first week and 57% were apparent by week 2 (s=47, n=8500, d/b, p/c, Pasternak and Zimmerman, J Clin Psych 2005, 66, 148-58)

“Time to substantial remission” may take 4 weeks in clinical trials

In 90% cases substantial improvement occurs within the first 2 weeks but that the benefit continues to build over several weeks.

(review by Mitchell, B J Psych 2006, 188, 105-6)

Markers of antidepressant response

• If no improvement (even minimal) after 4 weeks of a therapeutic dose, should switch to another one

• With minimal improvement, continue until week 6 but there is only benefit in continuing in about 10% pts

(n=593, Quitkin et al, Arch Gen Psychiatry 1996, 53, 785-92

If there is no response by 8 weeks then the trial should “be declared a failure”

(n=840, 12/52, open, Quitkin et al, Am J Psych 2003, 160, 734-40)

Only 58% people take antidepressants for more than 28 days

(n=829, Offson et al, Am J Psych 2006, 163, 101-8).

Duration of antidepressant therapy summary

40% of people may relapse after an index depressive episode within 2 years, and 60% within 5 years

First episode: Six months after recovery at same dose minimises risk

of relapse(n=839, RCT, one-year, Reimherr et al, Am J Psych 1998, 155, 1247-53

Second episode: 1-2 years

Third or subsequent episode: 3-5 years or longer

(Frank and Kupfer, Arch Gen Psych 1990 and 1992)

Depression relapse prevention

Full-dose vs. half-dose tricyclics

Frank et al, J Aff Dis 1993, 27, 139-45

0

20

40

60

80

100

0 6 12 18 24 30 36Months

% r

emain

ing w

ell

Full-dose TCA

Half-dose TCA

Relapse prevention

Frank and Kupfer, Arch Gen Psych 1990 and 1992

0

20

40

60

80

100

0 6 12 18 24 30 36 42 48 54 60Months

% s

till w

ell

Full-dose3-yr switchPlacebo

Imipramine Study

Frank E et al, Arch Gen Psychiatry 1990;47:1093-9

n=128

Long-term studies

6 months 1 year 2 years +

imipramine 1

fluoxetine 2

citalopram 3

paroxetine 4

sertraline 5

escitalopram

6

duloxetine 7

1. Frank E et al Arch Gen Psychiatry. 1990;47:1093-99 5. Wilson et al, 2003;182:492-72. Reimherr F et al Am J Psychiatry 1998;155:1247-53 6. Kornstein S et al J Clin Psych 2006; 1767-753. Franchini et al J Clin Psychiatry 1999;60:861-5 7. Detke M et al Euro Neuro Psych 2004;457-704. Reynolds C et al New Engl J Med. 2006;1130-8

‘Positive’ long-term studies 6 months 1 year 2 years +

imipramine 1

fluoxetine X2

citalopram X3

paroxetine ?4

sertraline X5

escitalopram

6

duloxetine 7

1. Frank E et al Arch Gen Psychiatry. 1990;47:1093-99 5. Wilson et al, 2003;182:492-4972. Reimherr F et al Am J Psychiatry. 1998;155:1247-53 6. Kornstein S et al J Clin Psych. 2006; 1767-17753. Franchini et al J Clin Psych. 1999;60:861-865 7. Detke M et al Euro Neuro Psych. 2004;457-4704. Reynolds C et al New Eng Jour of Med. 2006;1130-1138

Long-Term StudiesDrug Dosing Regime Results Notes

imipramine (160-200mg)

12 weeks treatment/ 17 weeks continuation/ 3 years maintenance

Relapse prevention over 3-5 years

Includes IPT

fluoxetine (20mg)

12-14 weeks treatment/ 50 weeks maintenance

Positive for 38 wks, no sig diff at 62 wks

4 treatment arms

citalopram

(20-40mg)

6 weeks treatment/ 4 months continuation/ 24 months maintenance

50% recurrence No placebo control. 20mg maint dose

sertraline

(50-200mg)

8 weeks treatment/ 16-20 weeks continuation/ 100 weeks maintenance

No significant difference to placebo at 100 weeks

Pts over 65

paroxetine

(10-40mg)

8 weeks treatment /16 weeks continuation/ 2 years maintenance

Significant difference between groups seen only after adjustment

Pts 70 yrs or over. Includes IPT.

Review of PREVENT Study

Prevention of Recurrent Episodes of Depression with Venlafaxine ER for

Two Years

Long-term treatment: venlafaxine XLPREVENT Study

Acute

(10 weeks)

Continuation

(6 months)

Maintenance A

(1 year)

Maintenance B

(1 year)

Venlafaxine Venlafaxine

Venlafaxine

Venlafaxine

Placebo

Placebo

Placebo

Responders

non-responders

non-responders

non-responders

non-responders

Discontinue Discontinue

Discontinue

Discontinue

Respo

nder

s

Respo

nder

sResponders

Responders

Responders

Simplified design adapted from Kornstein SG et al. Presented at the 6th International Forum on Mood and Anxiety Disorders, Vienna, Austria, November 29-December 1, 2006.

PREVENT: Primary Objectives

Acute Phase (10 weeks): Demonstrate venlafaxine XL efficacy in achieving

satisfactory therapeutic response/remission in MDD patients following 10 weeks of treatment

Continuation Phase (6 months): Demonstrate venlafaxine XL efficacy in sustaining response

(satisfactory therapeutic response/remission) - in MDD patients who achieved these responses in the acute phase

Maintenance Phases (2 consecutive 12-month phases): Compare venlafaxine XL and placebo in preventing

recurrence of depression in MDD patients who are responders after previous phases of venlafaxine XL treatment

Keller M et al Poster presented at ACNP Puerto Rico December 2004

Major Inclusion/Exclusion Criteria

Inclusions: 3 major depressive episodes in lifetime,

including 2 episodes in past 5 years* HAM-D17 of 18 at baseline Symptoms present for 1 monthExclusions: Previous failure to respond to an adequate

trial of venlafaxine In previous 3-years, failed 3 adequate trials of

2 classes of antidepressants*Interval ≥ 2 months between the end of previous episode and beginning of current episode.

Keller M et al Poster presented at ACNP Puerto Rico December 2004

Response/Remission definitions

Satisfactory Therapeutic Response

Total HAM-D17 ≤12 or minimum 50% reduction of total HAM-D17 compared to the acute baseline

Remission

HAM-D17 score ≤7

Sustained Remission

HAM-D17 score ≤7 sustained on two consecutive visits

Keller M et al Poster presented at ACNP Puerto Rico December 2004

Definitions used in post hoc analysis <225mg/d

Doses <225mg in acute or continuation phase Recurrence defined as:

Patients who had total score of HAM-D17>12 and a HAMD17 reduction no more than 50% from acute-phase baseline at two consecutive visits, or at the last valid visit in case of early discontinuation

Or Dose >225mg

Kornstein S et al Poster presented at IFMAD Vienna November 2006

Probability of recurrence – combined maintenance phases

by Dose Subgroup ( 225 mg)

Kornstein S et al Poster presented at IFMAD Vienna November 2006

Summary1-7+8

6 months

1 year 2 years +

imipramine1

fluoxetine2 No significant difference from placebo at 62 weeks

citalopram3 No placebo control

paroxetine4 No significant difference over placebo (unless

adjusted for IPT)

sertraline5 No significant difference to placebo at 100 weeks

escitalopram6

no published data

Duloxetine7 no published data

venlafaxine8

Discontinuing or switching antidepressants

Why discontinue or switch antidepressants?

Lack of efficacy Adverse effects Patient discontinues of own accord End of maintenance phase

What you can do if there is alack of response

1. Increase the dose 2. Switch antidepressants 3. Augment with:

another antidepressant mood stabiliser anxiolytic another drug e.g. pindolol, thyroxine

etc

1. Increasing the dose- types of dose-efficacy relationship

0

10

20

30

40

50

60

70

80

90

100

FlatCurvilinearLinearTherapeutic windowStepped

Summary of dose-response curves

Dose-efficacy Concentration-efficacy

Data Result Data Results

Tricyclics + Curvilinearor linear

+++ Linear, flat,curvilinear ortherapeuticwindow

MAOIs + N/K + Flat

SSRIs +++ Flat ++ Flat

Venlafaxine +++ Curvilinearor linear

++ Linear

Mirtazapine N/K N/K N/K N/K

Fluoxetine fixed-dose study

0

10

20

30

40

50

60

70

Placebo 5mg 20mg 40mg

ResponseRemission

Altamura Altamura et al, B J Psychet al, B J Psych 1988, 1988, 153153(Suppl 3), 109-(Suppl 3), 109-112112

Fluoxetine fixed-dose study

0

10

20

30

40

50

60

Placebo 20mg 40mg 60mg

ResponseRemission

Altamura Altamura et al, B J Psychet al, B J Psych 1988, 1988, 153153(Suppl 3), (Suppl 3), 109-112109-112

Venlafaxine dose-response

0

10

20

30

40

50

60

70

80

Placebo 75mg 225mg 375mg

Response at 6weeksRemission at 6weeks

Rudolph Rudolph et al, J Clin Psychiatryet al, J Clin Psychiatry 1998, 1998, 5959, 116-, 116-122122

The chances of success with increasing dose:

Limited: SSRIs (generally side effects limited) Mirtazapine (unknown)

Possible: Tricyclics (side effects increase) MAOIs (side effects and toxicity increase)

Probable: Venlafaxine (side effects increase)

Switching antidepressants

Factors to consider: Speed at which the

switch is needed Current dose of the

first drug Individual drugs

effects, transmitter effects, kinetics etc

Individual susceptibility to (additive) side-effects

Potential problems: Cholinergic rebound Antidepressant

discontinuation symptoms

Drug-drug interactions Discontinuation effects

from first drug interpreted as side-effects of the second

Serotonin Syndrome for drugs affecting serotonin

Serotonin syndrome

Definition - a toxic state caused by an increase in brain serotonin activity.

Symptoms 1. Neuromuscular

Restlessness Myoclonus Tremor and rigidity Hyperreflexia

2. Others Shivering/elevated temperature Arrhythmias etc.

Can be fatal due to cardiac collapse

Causes Most often with combined or consecutive treatment with SSRIs, tricyclics, MAOIs, tryptophan etc

Treatments Stop drugs - usually

resolves in no more than 24 hours

Symptomatic measures e.g. cooling, BDZs

Prevention take care when combining

or switching serotonergic antidepressants

Discontinuation phenomena

Characteristics: Commence within 1-3 days of stopping or

reducing doses Usually short-lived (1-2 weeks) Rapidly suppressed by re-introduction of

drug Distinct from relapse or recurrence, which

occur 2+ weeks after discontinuation Can occur even with missed doses

Discontinuation symptoms

Tricyclics: Cholinergic rebound

headache, restlessness, diarrhoea, nausea

‘flu-like symptoms, cramps lethargy sleep disturbances movement disorders

SNRI (venlafaxine): Fatigue, headache,

restlessness, nausea abdominal distension,

congested sinuses

“SSRI discontinuation”: Dizziness, light-headedness Sleep disturbances agitation, volatility electric shocks in the head nausea, fatigue, headache ‘flu-like symptoms

Mirtazapine & reboxetine: Little or nothing reported MAOIs: Confusion, delirium,

psychosis

Paroxetine discontinuation

0

20

40

60

80

100

120

Time since last dose

Pla

sma leve

l Daily paroxetine

ParoxetinediscontinuationNormal decay

“MedEd” technique

Craving? Tolerance? Withdrawal? Immediateeffect?

Alcohol

Opiates

Caffeine

Nicotine

Cannabis x

Hypnotics x

Antidepressants x ? () x

Antipsychotics x x () ()

Lithium x x x

Insulin x x ()

STAR*DSequenced Treatment Alternatives to Relieve Depression

Largest ever RCT in depression Independent, funded by NIMH

Focus on REMISSION Inclusions:

MDD, chronic depression, co-morbidities (61.5%) Exclusions:

OCD, eating disorders, bipolar, prior non-response, pregnant, breast-feeding, recent hospitalisation for SM, psychotic depression

Drugs/strategies chosen as safest, easiest and most used

STAR*D Level 1 – open SSRI

Remitters followed up over one year Non-responders or intolerant entered level 2 Some would also have been partial responders Patients had choice in which type of strategy

Level 2 – switch or augment Level 3 – switch or augment Level 4 – switch

STAR*D – conclusionsLevel 1: Give people 8/52 to respond Push citalopram to 40(-60)mg/d Level 2: Switching to another SSRI is as effective as other switches Switching may be better if no response Augmentation may be better if partial/incomplete

response but could actually just be delayed SSRI response (other RCTs)

Level 3: No significant difference between mirtazapine or

nortriptyline in response or tolerability Augmentation looks better Don’t know if these would have been better earlier Level 4: Response limited

Depression in bipolar disorder

Bipolar depression: is more resistant and longer-lasting

up to 50% may still be depressed at one year (Hlastala et al, Depress Anxiety 1997, 5, 73–83)

may respond to mood stabilisers e.g. lithium, valproate, carbamazepine etc

is susceptible to manic switch, especially in first 12 weeks

use lowest switch risk drugs, eg. SSRIs, mirtazapine Beware of inducing a mixed state in bipolar III

risk of self-harm/suicide is high

Antidepressants in bipolar depression

Antidepressant (paroxetine <40mg/d or bupropion <375mg/d)

or placebo plus Mood stabiliser (lithium, valproate, carbamazepine or a

licensed antimanic agent e.g. olanzapine, risperidone, aripiprazole, quetiapine, ziprasidone)

Outcome aim was 8/52 euthymia: Mood stabiliser and antidepressant response 23.5% Mood stabiliser and placebo response 27.3%

Longer-term adjunctive antidepressants have no therapeutic advantage

but at least the antidepressant did not increase the risk of relapse or switch to mania nor have greater ADRs

(n=366, RCT, d/b, p/c, 26/52, STEP-BD, Sachs et al, N Engl J Med 2007;356:1-12)

Quetiapine in bipolar depression

BOLDER 1 n=542, MD episode in

Bipolar I or II Response

600mg/d = 58.2% 300mg/d = 57.6% placebo = 36%

Remission 52.9% vs 28.4% Treatment emergent mania

3-4% for both groups

(n=542, RCT, d/b, p/c, 8/52, Calabrese et al, Am J Psych 2005,

162, 1351-60)

BOLDER 2 Quetiapine 300mg and

600mg/d monotherapy equally effective in bipolar I and II depression

53% response in BD

(n=542, RCT, 8/52, p/c, Hirschfeld et al, J Clin Psychiatry

2006;67:355-62).

Lamotrigine

Lamotrigine 50–200mg/d monotherapy significantly more effective than placebo in bipolar I depression

n=195, RCT, Calabrese et al, J Clin Psychiatry 1999, 60, 79–88

Survival rates favoured lamotrigine, with 41% stable without relapse at 6/12 (cf 26% placebo)

Well-tolerated, may thus be useful in some rapid-cyclers n=324, open + n=182 d/b maintenance phase,

Calabrese et al, J Clin Psych 2000, 61, 841-50

Two unpublished, negative studiesUnlicensed in UK, and never will be

NICE mentioned for relapse prevention of bipolar depression

Conclusion

Depression is a chronic condition and antidepressants:

are effective in acute depression prevent relapse are not addictive nor dependence prone help correct a chemical imbalance have no major documented long-term harmful effects appear to be widely used sub-optimally

Resistant depression might be undiagnosed bipolar Education about antidepressant use should be integral

with all prescribing, as it improves attitudes and hence “concordance”