This site is intended for health professionals only

At the heart of general practice since 1960

Could you cope in aftermath of bioterrorist attack?

Prescribing advice from

Dr Ross Taylor,

a member

of the CSM expert group conducting an inquiry into

SSRI safety

Controversy has surrounded SSRIs for several years (see page 29). But because of their effectiveness and safety against overdosing they continue to be prescribed by GPs.

They are widely considered to be as effective as tricyclic antidepressants, which many GPs were accused of prescribing in ineffective doses, possibly because of their potential side-effects.

GPs are more prone to prescribe SSRIs in doses proved to be effective in clinical trials.

Prescribing of all antidepressant medicines rose from 9.9 million in England in 1993 to 23.3 million in 2002, with much of this due to increasing prescription of SSRIs.

It is not clear how much this increase is due to the high-profile 'Defeat Depression' campaign launched by the RCGP and Royal College of Psychiatry in 1993. The campaign grew out of concern that depression was under-recognised in general practice and long-term use of antidepressants was encouraged to prevent possible recurrence, depression being increasingly characterised as a chronic, relapsing condition.

Dr David Healy, director of psychological medicine at the University of Wales, and others now contend that there is little or no evidence for this relapsing model in most primary care depression. They point to trials of SSRIs in the prophylaxis of depression where patients who had recovered from depression on SSRI treatment were then re-randomised to placebo and withdrawl symptoms were misinterpreted as recurrence of anxiety/depression. Their contention is that SSRIs can produce dependence, similar to that of benzodiazepines, so that an unknown number of patients continue to take them in order to avoid the unpleasant consequences of trying to stop.

How effective are antidepressants?

On an intention-to-treat basis, short-term response rates for antidepressants are roughly double that of placebo (50-65 per cent as opposed to 25-30 per cent)1.

This relatively small overall effect might even be exaggerated because of various methodological and other difficulties associated with trials of antidepressant drugs. For example, a blind test is very difficult because the side-effects of antidepressants tend to reveal the active treatment. In the few trials where an active placebo has been used, that is an active pharmacological principle that produces side-effects, the drug versus placebo difference in effect has been even less.

Although, on average, the effect of SSRIs – and other antidepressants – is relatively marginal, this could conceal a major clinical effect in selected individuals. It would help, of course, if we could tell in advance which patient was going to be which and this is where there might be improvements in the future.

It may be that genetic differences are important in predisposing certain individuals towards adverse effects of these drugs, particularly akathisia and suicidal ideation, for example.

Rau et al2 have recently produced evidence to suggest deficiency of cytochrome P450 2DY (CYP2D6) increases the risk of adverse effects and early termination of antidepressant therapy whereas over-expression of the CYP2D6 enzyme increases the

chance that a patient will suffer

from lack of antidepressant response.

It is estimated up to 10 per cent of the population might be deficient in CYP2D6 while about 4 per cent might be rapid metabolisers. At least a third of patients will recover from mild depression within six-eight weeks through supportive care alone.

The placebo response may also be greatest in people with mild depression, with the placebo-drug difference becoming greater with increasing severity of depression3. Spontaneous recovery is common in milder depression and most frequent early on, so the longer we delay antidepressant treatment the more likely we are to reach those who truly require medication.

Practical implications

Many now wonder whether GPs are going too far and are in fact 'medicalising' unhappiness rather than treating true depressive illness. A particularly critical review by Drug and Therapeutics Bulletin10 reported evidence that 'most depression that GPs treat is below, or just reaches, the minimum diagnostic criteria for major depression'.

The criteria for the diagnosis of depression vary and are generally derived from criteria originally developed for research studies – not particularly helpful for real-life general practice. PRODIGY11 guidance gives three main scenarios for initiation of treatment based on how far they fulfil the DSM-IV criteria for major depression (see table above).

For a diagnosis of major depression the symptoms must also not be due to a physical/organic factor (for example substance abuse) or illness and not be better explained by bereavement (although this can be complicated by major depression).

The two other scenarios are then:

lDysthymia where there is a chronic milder depressive disorder, with depressive symptoms for at least two years not meeting the DSM-IV criteria for major depression

lMilder depression with recent-onset depressive symptoms not meeting the DSM-IV criteria.

PRODIGY recommends antidepressant drugs as first-line treatment for major depression and dysthymia but not for milder depression, unless there is a previous history of a major depressive episode. For milder depression, support and simple problem solving will resolve most episodes.

The current draft NICE guideline12 does not recommend antidepressants for the initial treatment of mild depression because the risk-benefit ratio is poor, unless there is a previous history of moderate or severe depression (this may be the start of a more severe episode), but, where treatment is required for more severe episodes, it recommends that SSRIs are used routinely. Generic fluoxetine is recommended, because its longer half-life means it is associated with fewer discontinuation symptoms, although it is also associated with a higher propensity for drug interactions. Specific advice is given for where a depressed patient being treated with an SSRI develops increased agitation early in treatment.

This may be new to some doctors, and includes providing appropriate information and 'if the patient prefers, either change to a different antidepressant or consider a brief period of concomitant treatment with a benzodiazepine followed by clinical review within two weeks'.

Latest developments

The events leading to the creation of a CSM working group on SSRIs are outlined in the box on page 29. The work of the group is not yet complete but so far has resulted in:

•The MHRA initiating a Europe-wide review of the risk-benefit of paroxetine.

•Revised SPC and patient information leaflet for paroxetine including warnings about use in under-18s and updated warnings relating to withdrawal reactions and suicidal behaviour.

•An article in the September edition of Current Problems in Pharmacovigilance reminding prescribers of CSM advice on the use of paroxetine in under-18s and of the licensing status in children of other SSRIs and related antidepressants.

•A factsheet on SSRIs to aid discussion between prescribers and patients.

•New advice in the British National Formulary on paediatric use of paroxetine.

•Commencement of a research study using GP records examining the possible association between SSRIs and suicide/suicidal thoughts and behaviour.

•A reminder to use the recommended 20mg dose of paroxetine in adults and that there is no evidence of increased efficacy for higher doses in the treatment of depression.

How my own prescribing

has changed

Initiating therapy

I am now more reluctant to start antidepressant therapy. A high percentage of patients discontinue therapy after the first prescription. Whether this is because they feel better or experience adverse effects, they have been put at risk of adverse effects without any possible benefit.

I would like first of all to eliminate this group as far as possible and so I now rarely prescribe any antidepressant at a first visit but spend more time talking over a series of sessions and prescribe only as a last resort.

My practice is fortunate in already operating 10-minute appointments but some practices may need to increase their consultation time.

Elderly patients

Where an antidepressant seems indicated in elderly patients I would still tend towards SSRIs as a potentially safer option than tricyclics, particularly in those with known cardiovascular disease.

However, these are also exactly the sorts of patient who might have been advised to take low-dose aspirin and there is some suggestion SSRIs increase the risk of bleeding arising from NSAIDs including aspirin13. Lofepramine and trazodone might be the best alternatives.

Children and adolescents

I think it would now be unwise for any kind of antidepressant to be initiated by a GP for this group. There might have to be a lower threshold for referral to secondary care – certainly this is what the CSM advice indicates.

In cases that can confidently be considered mild, with no suicidal ideation, I would be inclined again to spend several sessions talking to the patient with and without parents if necessary.

Where obvious precipitants could be identified, for example bullying or other school-related difficulties, it might be possible to deal with these at primary care level, at least in the first instance.

If there were long waits for secondary care it would be necessary to provide regular supportive contacts. Although no antidepressant is licensed for use in children, the CSM considers fluoxetine to be effective, but suggests all antidepressants are used only with specialist advice.

Patients already on SSRIs

Until the NICE guidance is published I would advise patients already on SSRIs to continue to take them for six months following remission of depression.

The current draft guidance states patients who have had two or more depressive episodes recently and experienced functional impairment during such episodes should continue taking antidepressants for two years. So patients who have been on antidepressants for more than two years should be reviewed and consideration given to withdrawing treatment which would generally need to be done by graduated reduction.

Where dosages make that difficult – for example 20mg fluoxetine – these longer-acting drugs can be given at wider intervals such as alternate days, twice weekly or even weekly.

Ross Taylor is a CSM member and senior lecturer at the department of general practice and primary care, University of Aberdeen – he stresses the opinions in this article are his own views and not those of the CSM

PRODIGY criteria for major depression

•Depressed mood

•Loss of interest or pleasure in almost all activities

••Significant weight loss or gain, or change in appetite

•Insomnia or hypersomnia

•Psychomotor agitation or retardation (observable by others)

•Fatigue or loss of energy

•Feelings of worthlessness or excessive or inappropriate guilt

•Diminished ability to think or concentrate, or indecisiveness

•Recurrent thoughts of death (not just fear of dying) or

suicidal thoughts/actions

Major depression can be diagnosed when at least five of these symptoms (including at least one of the first two listed) are present during the last two weeks for most of the day, or nearly every day, and these cause clinically significant distress or impairment in functioning.

The chronology and controversy of SSRIs


CSM reports on the occurrence of increased anxiety in the early stages of treatment with fluvoxamine, a year after its launch in the UK4.

June 1990

A CSM review follows the publication of a case series by Teicher et al5 which describes severe, intrusive and suicidal ideas involving thoughts of self-harm in six patients, two-seven weeks after starting fluoxetine.

It concludes that there is no evidence that fluoxetine was the cause of violent behaviour or suicidal acts (as documented by and supported by Drug and Therapeutics Bulletin6).

December 1990

Paroxetine licensed in the UK with withdrawal reactions included in the product information.


CSM highlights the occurrence of 39 reports of extrapyramidal reactions (particularly facial dystonias) in the early stages of treatment with paroxetine and 78 cases of withdrawal reactions 'including dizziness, sweating, nausea, insomnia, tremor and confusion'7.

CSM recommended that 'paroxetine should not normally be discontinued abruptly'.


CSM advises that core SPC for all SSRIs should reflect the general experience that suicidal behaviour increases in the early stages of treatment.


CSM concludes evidence on suicide not sufficient to confirm a

causal association, although an effect in a small high-risk population cannot be ruled out. Recommends that akathisia be added to the SSRI SPCs.

October 2002

BBC TV Panorama programme8 Secrets of Seroxat reflects growing public concern about possible dependency and other adverse effects, including agitation and suicide in individual patients who had not previously given any indication of suicidal ideation, and the occurrence of a highly unpleasant electric shock-like sensation, or 'electric head', on withdrawal.

December 2002

Dr David Healy presents his research in relation to suicidal behaviour to a group of CSM and external experts. The CSM and experts conclude that the evidence does not justify a change to the regulatory position but that changes to the UK Seroxat patient information leaflet are required to clarify warnings on withdrawal reactions. They also recommend further work to investigate suicidal behaviour via the general practice research database and a re-analysis of clinical trial data on fluoxetine.

March 2003

A further meeting of this same group is cancelled after the independence of two members of the group is called into

question because of their declared interests in pharmaceutical companies. The group dissolves.

April 2003

CSM convenes a formal expert working group.

September 2003

The new group issues its interim report9 after reconsidering all the evidence on withdrawal reactions and suicidal behaviour. It also urgently considers new data on the safety of paroxetine in the treatment of major depressive disorder in children and adolescents (and, subsequently, similar data for other SSRIs).


01 Schulberg et al, 1999, quoted in NICE Draft guidance, 7.1, p.141

02 Rau T et al. CYP2D6 genotype: Impact on adverse effects and no response during treatment with antidepressants – a pilot study. Clinical Pharmacology and Therapeutics 2004;75(5):386-93. See also associated editorial by Wedlund PJ and Leon J. Cytochrome P4502D6 and antidepressant toxicity and response: what is the evidence?

Clinical Pharmacology and Therapeutics 2004;75(5):373-4

03 Angst 1992, Khan et al 2002, quoted in the NICE draft guidance

4 Committee on Safety of Medicines. Fluvoxamine (Faverin):

Adverse Reaction Profile. Current Problems 1988; 22: May 1988

05 Teicher MH et al. Emergence of intense suicidal preoccupation during fluoxetine treatment. American Journal of Psychiatry 1990; 147: 207-10

06 Consumers Association. Fluoxetine, Suicide and Aggression. Drug and Therapeutics Bulletin 1992;30:5-6.6

07 Committee on Safety of Medicines/Medicines Control Agency.

Dystonia and withdrawal symptoms with paroxetine (Seroxat).

Current Problems 1993; 19: 1

08 Medawar C et al. Panorama and user reporting of ADRs: Consumer intelligence matters in clinical practice and post-marketing surveillance. International Journal of Risk and Safety in Medicine 2002; 15: 161-9

09 Publicly available through the CSM website (September 2003)

10 Mild depression in general practice: time for a rethink? Drug and Therapeutics Bulletin, volume 41 No 8, August 2003 p1-5


12 NICE Draft Guidance

13 BNF 47, March 2004, page 627

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say