The problem with mental illness is that one is by definition talking about a condition that, until recently at least, did not display symptoms and physical changes that could be scientifically verified. There is no blood test for depression. If you hook up someone with dementia or epilepsy to a brain scanner or an EEG machine, atypical brain anatomy and electrical activity will be observed. But there is no equivalent set of diagnostic signs for something as clearly different from normal thought processes as active schizophrenia, let alone gradations in the severity of conditions most of us have had at some time, like depression.

This does not mean that subtle brain changes (in the activity of certain neurotransmitters, for instance) cannot be picked up; it means that there is not a gross pathology of the central nervous system process which has a clear cause, whether or not it is amenable to treatment.

There is also a philosophical dimension to the diagnosis of mental illness which is summed up in the saying that “if the human brain was simple enough to understand, we would be too simple to understand it”.  Mental ill-health affects not only our physical bodies but our social abilities and emotional being. This means that a pure ‘medical model’ where a physician – presumed not be affected or being part of the condition in question – diagnoses a condition and then applies expertise and treatments to effect a physical cure or relief.

Mental illness is both a physical condition in which very powerful changes can happen in the brain and body chemistry (as anyone who has had a panic attack or major depressive episode well knows), but is also a social illness; a disorder of the person’s ability to relate to others. Loneliness is the strongest single predictor of mental illness: up to three-quarters of people with depression feel lonely, and vice versa. It is also strongly associated with physical ill-health (House), and, of relevance to the present chapter, with unprotected sex with casual partners in gay men (Martin)

Perhaps because of this two very different treatment modalities have arisen for mental illness:

• psychopharmacology, or the prescription of drugs to alter mind states
• psychotherapy, ‘talking therapy’ in which a cure is effected by discussing one’s problems with a skilled practitioner. May be called counselling if of a short-term or less intensive nature.

These two modalities exist in uneasy relationship and it is not uncommon to encounter doctors who dismiss the effectiveness of psychotherapy for anything other than mild mental illness or psychotherapists who regard resorting to antidepressants as a defeat. In reality they are complementary; psychopharmacology deals with mental illness from the ‘bottom up’ by blunting or calming extreme emotions or mental states so that the person may begin to deal with their problems, while psychotherapy deals with the situation from the ‘top down’ by helping the individual to regain their lost sense of self-confidence and self-efficacy.

Different psychotherapies work in different ways according to how much stress they place on the patient-therapist relationship. For short-course therapies such as Cognitive-Behavioural Therapy (CBT), the therapist acts more like an instructor who prescribes different methods by which the patient can control negative thought patterns and behaviours. However for long-term and lifestyle problems where self-image may interfere with the patient’s motivation to get better, more stress is played on using the therapeutic encounter as a ‘model relationship’ in which patients may experience attention and understanding they may have lacked and work out experientially how to alter maladaptive ways of being with others.

A meta-review (Gill 2005) of studies of the effectiveness of antidepressants found that patients given antidepressants were 63% more likely to recover from depression than patients given a placebo. However it also found that patients given a placebo were 3.45 times more likely to recover than patients given no treatment. This nicely quantifies the extent to which drugs can alter mental states above and beyond what the patient may be able to do for themselves in ideal situations where the belief that they can get better is itself the thing that cures.

Psychotherapy is sometimes criticised as being a treatment only those already motivated to change can benefit from, and furthermore that it is a ‘western’ or ‘middle class’ approach that depends on the ability to talk about mental experience in terms that are culturally similar to the therapist. In fact there is a huge evidence base as the effectiveness of psychotherapy in all sorts of settings. To give one example of relevance to this chapter (Bolton 2003), 341 men and women from 30 villages in rural Uganda diagnosed with major depression were randomised to receive group interpersonal therapy or no treatment. After 16 weekly sessions, the proportion of subjects meeting the criteria for major depressive disorder fell from 86 to 6% in the psychotherapy group as compared with a fall from 94 to 55% in the control group, or in other words psychotherapy worked 5-6 times better than doing nothing.

The investigators comment that they decided on group psychotherapy because they could not afford to prescribe antidepressants in the African context. This is ironic, as in the UK NHS-provided psychotherapy has long waiting lists and patients are more likely to be given a prescription for psychotherapeutic drugs and left to cope with issues than provided with what is seen as a costly and time-intensive resource.

Cost-effectiveness is a difficult thing to measure in a situation where outcomes are so subjective, both on the part of patient and therapist. However several studies (Browne, Burnand, Scott) have calculated that different kinds of psychotherapy for depression saved anything from £275 over two years to £4500 over doing nothing. Miller (2003) calculated that the depression-associated cost per patient of counselling for mild to moderate depression was £302 for counselling as against £342 for antidepressant prescription, which was statistically equivalent. However this is still an area where the evidence is weak, though qualitative surveys almost universally report that patients value therapy.