Mrs Ward first went to her therapist, a clinical psychologist, to help her come to terms with an unhappy experience she suffered several years ago. At first she was wary of trusting him, she says, but after nine months of regular 50- minute sessions, her defences were worn down. She became wholly dependent on him. 'I felt I had been emotionally and mentally seduced. I loved him. In the end I would have believed him if he'd told me black was white.'
Mrs Ward is now sceptical of her therapist's approach, which permitted her dependency and left her in so much pain. 'It's like having a wound inside because you are missing someone so much.' She is not alone in her misery.
Jenny Fasal, a private psychotherapist based in London, and founder of the Prevention of Professional Abuse Network (Popan), an organisation set up to address the issue of abuse in therapy, believes the problem of 'addiction' is widespread. It occurs between client and counsellors of the same and the opposite sexes, and even in group therapy. It can happen whatever the type of therapy practised.
'People come to us saying: 'I got hooked,' ' says Ms Fasal. 'Or: 'I felt if I didn't see him/her I was going to die.' Or: 'I don't know what happened. I felt that once I'd seen her I just had to go back.' '
The intimacy of the counselling setting means that vulnerable and often lonely people can develop very warm feelings for their therapists, Ms Fasal says. If the situation is not handled carefully, clients can be left in far worse emotional states than when they started.
There are at least 450 recognised methods of psychotherapy in existence, and a variety of practitioners with a confusing range of methods, training and qualifications. People offering psychotherapy may be counsellors, trained psychotherapists, clinical psychologists, psychoanalysts or psychiatrists.
Client dependence is a recognised phenomenon. But while some therapists see this as a potential 'force for good', a creative analytical tool with which to work out deep-rooted problems, others will use all their skills to avoid it.
More disturbingly, it is acknowledged that some counsellors have a personal incentive to keep patients coming back. As Windy Dryden, Professor of Counselling at Goldsmiths' College, London University, points out, the motive need not be purely financial. 'Some therapists have the notion that the longer they keep their clients, the better they are as therapists,' he says. 'They use this to bolster their self-esteem. Others like the idea of having a following. They may not just see their clients in individual therapy but in a group, and a cult starts to build up, which is decidedly unhealthy.'
Professor Dryden believes, however, that dependence on the therapist, if dealt with correctly, can be helpful. The term for this phenomenon, in which the client-counsellor relationship mirrors childlike patterns of relating to significant adults, is transference.
Derek Hill, head of counselling at Relate, the marriage guidance service, says the organisation uses this approach only with a minority of clients who are seen by its more experienced therapists. The counsellor takes on the role of the parent figure and the client the child.
'This is allowed so that clients can rework issues not sorted out with their real parents,' he says. 'In long-term work there may be a passage during which there is a considerable dependence on the counsellor or therapist. The purpose is to help the client through 'adolescence', and when they are ready, to take on responsibility for themselves.'
At that point they will no longer need counselling, he adds. At the conclusion of therapy, the onus is on the therapist to work towards a 'good ending' and to help the client through the natural period of bereavement.
Mrs Ward believes this is where things with her therapist went wrong. He gave her three months' notice of his departure; by the time he left his job, in February 1991, she had been seeing him regularly for a year. 'He constantly reassured me it would be all right in the end, that we would go through a grieving process together when we agreed the time was right,' she says. But it did not work quite like that: 'For the last few sessions I just sat and cried.'
Mrs Ward went to another psychologist, a woman, but found it did not help. 'I didn't want to get too close because I didn't want a repeat of what had happened before,' she says. 'I found it very difficult to trust her.'
Craig Newnes, head of the psychology consultancy service at the Royal Shrewsbury Hospital, Shropshire, and the editor of Changes, an international journal of psychology and psychotherapy, has an interest in the ethics of psychotherapy. He believes that although trained psychoanalysts may be able to use their clients' dependence creatively, 'the majority of counsellors and therapists are not trained in how to work with dependence or how to end a relationship properly.'
Professor Dryden warns that there will be people who become dependent whatever the technique, and for whom little can be done. 'No matter how skilful you are, a small minority of patients is going to zip into a dependent role quite quickly.'
But some professionals are adamant that dependence should simply be prevented. Pamela Ashurst, consultant psychotherapist for Southampton health district, believes addiction is avoidable, and that a mystique surrounds 'dependent personalities'. 'If people become dependent on counselling one should always look for the problem in the counsellor rather than in the patient,' she says.
She thinks that if counsellors are sufficiently well trained they should be able to avoid dependency. It is important, she argues, to assess a person's independence in the outside world before entering into 'heavy work', and to prevent therapy becoming the centre of the person's life. In order to achieve this, clear goals must be agreed with the client and adhered to within a set framework of sessions.
Other counsellors believe that no matter how experienced, therapists should never work in isolation, but should always be under supervision from colleagues.
Craig Newnes believes it would help if all counsellors were more aware of their own power, conscious of the effects of being kind, decent and responsive to someone in a distressed state. Being more honest would keep the relationship in perspective. 'One of the painful things clients need to be told is that only a third of people will benefit from therapy, and a third will get worse,' he says. 'Lots of research has shown that 65 per cent of people either stayed the same or got better - and that was little different from a control group who did not have therapy at all.'
Clients should make sure they know their rights: that they can leave whenever they like, and can question the therapist's values and training.
As a result of her experiences, Melanie Ward recently set up a self-help group in the East Midlands for those with nowhere to turn to. Of the nine members, seven feel their difficulties stem from dependence on therapy, ranging from marriage guidance to full-blown psychotherapy. 'Many people suffer in silence and are frightened about telling their therapists because they feel stupid about becoming dependent,' Mrs Ward says.
She feels that people considering therapy should be warned of the dangers of dependence. 'They say it's just transference and it doesn't really hurt,' she says. 'But I would question whether clients can be weaned off their therapists painlessly, in whatever circumstances the therapy ends, because the power imbalance is so great.
'Therapists must take more responsibility for the relationships they are engendering. Tinkering with someone's psyche should carry a government health warning.'
The Prevention of Professional Abuse Network is at Flat 1, 20 Daleham Gardens, London NW3 5DA. Please enclose a stamped addressed envelope.