Mental Health Nursing interventions
The National Service Framework (NSF) (Department of Health, 1999a) sets out the current government strategy for all aspects of mental health for adults under the age of 65. One of the key elements of the framework is patient access to and provision of psychological therapies - clearly issues that directly concern mental health nurses.
The NSF attempts to be evidence based wherever possible and specifically mentions that the effectiveness of some therapies is supported by empirical evidence. These include dialectical behaviour therapy (DBT), cognitive behaviour therapy (CBT), and psychosocial interventions (PSI) for people with schizophrenia, which are often derived from CBT principles.
There is a shortage of mental health nurses adequately prepared to deliver CBT, and only 10% of mental health nurses have so far received adequate training in providing PSI (Brooker, 1999). This leaves the majority of mental health nurses with no training in ‘evidence-based’ interventions. The NSF states that staff training is needed but full implementation of the framework is expected to take between five and 10 years.
So where does this leave those mental health nurses who are currently offering psychological interventions to their patients which they feel may not be evidence based?
Many mental health nurses, especially CPNs, are involved in counselling, which can range from informal chats through to structured, one-to-one psychotherapy sessions. It is likely that the majority of mental health nurses, while not specifically trained in CBT, will have received some training in psychological interventions, often using non-directive or client-centred methods. Many will also have undertaken postregistration training in other models of counselling or psychotherapy.
Although there is a great deal of research into CBT, other models - especially humanistic or experiential therapies - are much less well researched (Greenberg et al, 1994). The evidence that supports CBT is a testament to the utility of the model. However, nurses involved in counselling who do not use CBT may be surprised to learn that there is a wealth of research evidence to support the use of alternative models.
A comforting paradox
One of the most consistent findings from 30 years of psychotherapy research is that models of psychotherapy achieve broadly similar outcomes, despite varying in their theoretical orientation. This is referred to as the ‘equivalent outcomes paradox’ (Stiles et al, 1986).
Different models of psychotherapy have both ‘specific’ and ‘non-specific’ effects. Specific effects are those factors unique to each model, such as the identification of negative thoughts in CBT or making interpretations between past and present relationships in psychodynamic psychotherapy. However, these specific techniques are estimated to account for only 12%-15% of the variance across therapies (Lambert, 1992).
Non-specific effects are the common factors present in all models of psychotherapy, irrespective of theoretical orientation. These include understanding, warmth, the instillation of hope and of feeling supported, as well as the ‘ritual’ associated with the provision of therapy.
Lambert and Bergin (1994) suggest that non-specific factors are one of the largest mediators of outcome and ‘should not be viewed as theoretically inert or trivial’. One of the most important of these common factors is the quality of the therapeutic alliance formed between the client and the therapist, which is strongly predictive of the outcome of the therapy (Roth and Parry, 1997).
An example of the equivalent outcomes paradox is provided by a large US study that compared the effectiveness of four treatments for depression: CBT; interpersonal therapy; imipramine plus clinical management; and a placebo plus clinical management. The study found that imipramine was the most effective treatment. The two psychotherapies came a close second, with virtually no difference in effectiveness between them and little evidence for the specific effects of either of the two therapies (Elkin et al, 1989).