Psychiatric/mental health nursing is often like a house divided. Even the double name commonly used to describe our specialty—psychiatric/mental health—reflects possible ambivalence or duality of purpose, or both. It should not surprise anyone that our attempts to implement evidence-based practice mirror this struggle.

Mental health nurses initially defined our practice as an interpersonal therapeutic process to assist the client in growth.^1–3 This was in synchrony with psychiatry's earlier focus on psychotherapy as the treatment for mental illness. The division in mental health nursing occurred when psychiatry moved to a more biological perspective. Should mental health nursing follow this route or continue to focus on counselling and process issues? Both the “therapeutic relationship” and “biology” camps have used evidence-based arguments to buttress their positions. It could be argued that all the psychiatric/mental health professions experience this struggle, but it has been a particularly open struggle in nursing.

The struggle is present in our literature. Gournay's article, “Schizophrenia: a review of the contemporary literature and implications for mental health nursing theory, practice and education,” is an example of this struggle.⁴ The main purpose of this article was to review the literature on schizophrenia from a biological perspective, including the aetiology, epidemiology, and neuropsychology of schizophrenia. A good chunk of the paper, however, focused on berating nurses for using interpersonal nursing theory. This paper was a continuation of an earlier article criticising the use of nursing theory in mental health nursing.⁵ Similarly, McCrone also wrote of the effect of biological psychiatry on nursing and noted “In the future, psychiatric nurses will need to know normal neuroanatomy, physiology, psychoneuroendocrinology, and immunology to understand deviations occurring with mental disorders” (p 46).⁶ These resulted in various volleys between the camps. For example, Dawson reviewed the articles included in Gournay's review and concluded that the evidence for a biological understanding was overstated by Gournay.⁷ Barker and Reynolds criticised Gournay's understanding of models and theories.⁸ In reply to the concern that Peplau's interpersonal theory did not include, for example, a biological basis for hallucinations, it was pointed out that none of the biological theories pays any attention to Peplau's theory of interpersonal relations.

Poor psychiatric/mental health nurses reading through this may be scratching their heads wondering what the evidence is for any of the discussion.

An attempt to move beyond the nursing literature yields similar controversy and uncertain conclusions. The classic meta-analysis of psychotherapy outcomes from Smith and Glass found only a modest effect size.⁹ Cooper & Hedges, however, compared this modest effect size of r=0.32 that sounded the death knell of psychotherapy with such effect sizes as those for 3'-azido-3'-deoxythymidine in patients with AIDS (r=0.23), where the effect was considered so strong that the clinical trial was closed prematurely.¹⁰ Huxley and others reviewed 32 peer reviewed articles relating to the effectiveness of psychosocial treatments in bipolar disorder.¹¹ They concluded that although the studies were few and of variable quality, important gains were found such as increased clinical stability, fewer hospital readmissions, and improved function.

More recently, Klein re-examined 4 meta-analyses comparing psychotherapy with pharmacology.¹² The original meta-analyses all found favourable results for psychotherapy, but Klein found numerous flaws in the studies used in the meta-analyses. Most disturbing was that few of the studies actually included a direct comparison of psychotherapy and pharmacology at all. Klein concludes “…the currently available data simply do not allow for a proper meta-analysis that can address the relative merits of pharmacology and psychotherapy” (p1210).

Surely one could argue that these articles in the psychiatry literature show that a similar struggle exists between biological and psychotherapeutic approaches in the mental health literature generally. The difference with nursing is that the struggle goes beyond the evaluation of whether the biological or interpersonal approach makes more sense based on the evidence. It is also based on the more philosophical question: should nursing follow medicine or position itself in an alternative/complementary position? Nursing, like other professions, endeavours to implement strategies that are known to be effective. In addition to being effective interventions, however, do they not also need to be nursing interventions? For example, even if psychopharmacology offered great benefit, does that mean that this should become the focus of psychiatric/mental health nursing as well as psychiatric medicine? If the evidence does not exist for a nursing intervention, does this reflect an ineffective intervention, or an understudied intervention?

When the psychiatrists, social workers, psychologists, nurses, and others use the best evidence-based practice available, the client should receive the best possible care. The focus of each profession, however, will be unique but complementary to address the range of issues faced by the client. For nursing, this focus has traditionally been the development of therapeutic relationships to foster growth. This focus would not change because a biological understanding is useful, but could change based on whether a therapeutic relationship continues to be a need of clients. What is needed to evolve our focus is continued study and examination of the evidence related to our current psychiatric/mental health nursing practice: therapeutic relationships.

The current Therapeutic Relationships: From Hospital to Community study is an example of research that examines a nursing intervention.¹³ The study involves a randomised cluster design with 26 psychiatric wards and 350 clients. The intervention focuses on the development of a safety net of interpersonal relationships. This nursing intervention is complementary to, but different from, the psychiatric treatment that is received. This study is one of a small but growing number of efforts to scientifically evaluate nursing interventions.

The arguments using the evidence to define an appropriate focus for psychiatric/mental health nursing seem reminiscent of the famous quote of the Scottish writer and poet, Andrew Lang (1844–1912): “… he uses statistics as a drunken man uses lampposts: for support rather than illumination.” To move towards illumination we need to ask questions arising from our practice. To develop evidence-based psychiatric/mental health nursing practice, we need to evaluate psychiatric/mental health nursing strategies. The results of well designed research will provide the light to guide our way to providing the best possible services for our clients.