This site is intended for health professionals only

At the heart of general practice since 1960

Read the latest issue online

CAMHS won't see you now

Initial approaches to depressed patients

What is your definition of treatment-resistant depression?

I define it as a depressive illness that's continued after two full trials, at least six weeks each, of antidepressants. Initial treatment fails to get a response in about a quarter of patients.

Half of depressed people don't have their illness picked up immediately. Of those that are identified, 80 per cent have an ineffective dose of medication for less than the minimum required time. Relapse occurs in 60 per cent of patients who don't have a long enough course.

You have to give prolonged treatment for an absolute minimum of six to nine months. Some of the people we think are treatment-resistant may have undertreated illness.

It's never too late to talk

Is it too late for 'talking therapy' when someone's depression is not responding?

Not at all. Always give supportive psychotherapy. Don't expect changes but be there for the patient, listen to them, reflect on what they're saying. Don't embark on dynamic psychotherapy where you are aiming for big changes, that's for later when somebody is no longer clinically depressed.

Use cognitive behaviour therapists (CBT) if you have access to them. CBT can be used in many different ways; primarily by dealing with negative thoughts that perpetuate low mood. Challenge these negative thoughts and expectations. I'm sure many GPs do CBT themselves without realising it.

So it's all right to be a little challenging when faced with a very negative and pessimistic patient who's going downhill?

Yes, that doesn't mean you're unsympathetic. Gently challenge and encourage them to see other ways of looking at things. That can be done sensitively without offence.

How do you initially present treatment with medication?

You have to overcome the bad publicity that psychotropic medications have had. Three-quarters of patients think antidepressants are addictive. Antidepressants, like any other drug, have side-effects. Only 40 per cent of GP patients who have been prescribed antidepressants are still taking them after 12 weeks. A lot of encouragement is needed to get compliance, and to accept that side-effects are worth it if they feel less depressed.

Newer drugs don't get more people better ­ the same number get better whichever antidepressant you choose ­ but there will be particular people who respond better to a particular drug.

What's your first rule when approaching a depressed patient?

Spend time engaging the patient properly, because you will be asking them to do difficult things, like take medication and persist with treatment. It's a complete waste of time if the patient doesn't believe what you are suggesting is going to work.

To avoid getting to a position where treatment is less successful, what's your next best step with these patients?

It will pay off if you take a good history, assess their current mental state, assess whether they're in danger, make a treatment plan, and make it clear to the patient what you're going to be expecting of them. Take baseline bloods (FBC, U&Es, glucose, LFTs, TFTs and maybe calcium). Think of hidden alcoholism and hypothyroidism, which can both present as depressive illness.

Remember that in older people dementia can be one reason for not responding to antidepressant medication.

Always encourage exercise in somebody who is depressed. Activity will get their endorphins going. Encourage healthy eating, watching alcohol intake and adequate rest, especially if sleep is disturbed.

Discontinuing treatment

How significant is discontinuation syndrome?

Adverse events have been reported with many antidepressants. They are not the same as withdrawal. The SSRIs, particularly paroxetine, have had considerable adverse publicity about this.

Discontinuation symptoms usually start within one to two days of stopping the drug, sometimes with just one missed dose. Symptoms will stop within 24 hours of restarting medication.

To reduce the chances of discontinuation after two months' treatment, withdrawal should take one to two weeks. After

six or nine months of treatment the

drug should be decreased over six to

nine weeks.

After longer courses, decrease the dose stepwise by 25 per cent every four to six weeks. With SSRIs, reduce by 10mg weekly, finally alternating days on and off the lowest dose.

If symptoms are troublesome with SSRI reduction, the decrease can be covered by fluoxetine, which has a very long half-life.

What is the place for lithium treatment?

Not only is it a prophylactic agent, but lithium can also be used to lift somebody out of a depressed state.

It works in combination with antidepressants and sometimes can be a key factor in improving resistant depression.

Again, I wouldn't expect a GP to have to shoulder that responsibility on their own.

Underlying factors in depression

What else could be going on for the patient?

Think about social situations, such as domestic violence, debts or other worries. Is marital or family therapy necessary? Having a depressed patient in the family affects everybody. Try self-help groups such as Cruse Bereavement Care, Depression Alliance and Sane Line.

If that doesn't work, review the notes. Is there a past history? Is this really depression or another psychiatric disorder, such as alcoholism, drug abuse, dementia, an adjustment reaction? Which interventions worked previously?

Have you missed an organic disease? The next step is a secondary referral. Despite its bad press, electro-convulsive therapy still is the fastest and most reliable treatment for severe depression, and it does save lives. It can be less problematic than drug therapy, especially in the elderly.

GPs are well placed to look at social and family factors. Ask yourself what's the function of this illness? What changes have resulted from it, or been prevented? Is there a decision fuelling it? Are you looking after the right family member?

Treating resistant depression can be draining. We can run out of steam and not want to bother any more. Rather than being ashamed of that, recognise what is happening. Ask a colleague to see them. The patient you find very difficult somebody else will have no problem with, and vice versa. It can be packaged in a very positive way to the patient: 'I'm asking Doctor X to review you, who will have a different slant on this.'

Finally, remember depressive illness can remit naturally. People used to get better before we had antidepressants. Half the people with depression will get better in six months. Don't lose heart, it isn't your fault and it isn't the patient's fault if they don't get better promptly.

Pharmacological management

How do you choose the right antidepressant?

If the patient is very anxious and you don't think there's a big risk of suicide, tricyclics can be used. They help the patient sleep and you don't need to give a sedative with them. They're good anxiolytics and they come in very small doses. Amitriptyline and doxepin are available in 10mg tablets so the dose can be slowly titrated.

Somebody who's somatising is more likely to have side-effects. If the patient is in charge of titrating their own dose you'll get better compliance, they feel more in control. Tricyclics have useful anti-pain and anti-sickness effects as well. If somebody is waking up nauseous or with diarrhoea with their anxiety, tricyclics slow the gut down nicely.

If somebody feels 'tired all the time' with their depression, an SSRI is a better first choice because they tend to energise. The disadvantage is that initially, particularly with fluoxetine, you can increase anxiety. Warning them about this makes a big difference. If they don't know what to expect they'll maybe give up on it.

Another disadvantage: SSRIs do sometimes cause sexual dysfunction. That isn't greater than the tricyclics, but it doesn't get better over time.

If the patient is complaining of extreme difficulty sleeping and is very tense, mirtazapine can be very useful. It tends to have a better

sleep-inducing effect at 15mg than 30mg.

There can be weight gain; again, this is more manageable if the patient is warned.

Venlafaxine is an extremely good antidepressant which has some enlivening action. It is often useful in treatment-resistant depression.

Moclobemide is a reversible inhibitor of monoamine oxidase. That tends to be more enlivening, so again can be very useful with people who are tired with their depression.

A few patients, especially the middle-aged and elderly, who have previously responded to old-fashioned MAOIs won't respond to anything else. They are a worry, but if you can trust them to be safe and sensible the drugs can be very good.

The pendulum has swung away from the use of benzodiazepines. Do you feel there is a place for their use in the early days of treatment of depression, particularly to improve sleep pattern, such as when somebody's on a rather alerting SSRI?

Yes. Avoid the most addictive ones (such as lorazepam), but consider a low dose of diazepam or perhaps clonazepam. Emphasise that you're only expecting them to need this for three to five days. The major tranquillisers, perhaps trifluoperazine in low dose, or chlorpromazine, do work and can augment antidepressants.

Is there still a place, if not at the GP level, at consultant level for combinations of antidepressants?

There very definitely is, but I'm afraid this is based on clinical experience rather than evidence. We do combine tricyclics with SSRIs. It isn't fair to expect GPs to do this on their own.

A low dose, say 10mg or 20mg of doxepin at night with say citalopram or fluoxetine in the morning, gives a sleep-making effect at night plus the alerting effect in the morning. We may use reboxetine, which is more enlivening, first thing in the morning with mirtazapine at night.

Useful contacts

National Centre for Psychiatric Medicines Information at the Maudsley Hospital offers information on switching and discontinuing antidepressants.

Tel: 0207 919 2317

Depression Alliance

35 Westminster Bridge Road,

London SE1 7JB. Tel: 0207 633 0557

Cruse Bereavement Care

Helpline 0870 167 1677

Young people between the age of 12 and 18 should call freephone 0808 808 1677


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