Theoretical basis for EBPP

In the 1870s, psychology emerged as a discipline distinct from philosophy, inspired by the idea that natural science and objective laboratory research could displace the dominance of religion over mind and spirit (Nilsen, 2014). A substantial body of knowledge was quickly accumulated, but psychology lacked overarching theories that could integrate valuable but fragmented findings (Schjelderup, 1927). In other scientific disciplines, theories are often combined to form overarching explanatory models. The role of theory is to describe how empirically established phenomena are related to one another and thus aid in formulating laws (Spence, 1944). However, the thinkers of psychology never managed to formulate theories that could be integrated into a coherent whole (Koch, 1993). Instead, biological, cognitive, psychodynamic, behaviourist, humanistic and sociocultural approaches emerged, offering different and often incommensurable explanations for the driving forces of human thoughts, feelings and behaviours (Grennes, 2009). To date, no one has demonstrated that any established theory is more valid than another or managed to integrate them into a unified theoretical framework (Alexander & Shelton, 2014; Hillix & L’Abate, 2012; Melchert, 2016). As a science, psychology has thus become a rather loose collection of disciplines and theories as opposed to a unified subject domain (Saugstad, 2009).

The lack of theoretical cohesion is also reflected in the diversity of treatment approaches in clinical psychology. Most treatment methods are grounded in one or more theoretical frameworks that provide different answers to the question of what triggers and maintains mental disorders (Alexander & Shelton, 2014; Sauer-Zavala et al., 2016). For example, humanistic-existential theory is based on the idea that humans are fundamentally motivated by self-actualisation and that mental health problems develop when life events and relational experiences create a gap between the person one perceives oneself to be and the person one seeks to be (Maslow, 1962). Regardless of what the problem is, therapy in this perspective involves creating a healing context and relationship in which the patient is better able to foster personal growth (Rogers, 1951). In psychodynamic theory, mental disorders develop when expressions of healthy emotions, such as anger or sexual desire, trigger negative emotions, such as anxiety, shame or self-blame, leading to the development of defence mechanisms to avoid inner conflict (McCullough et al., 2003). This triangle of conflict (healthy emotions → negative emotions → defence mechanisms) (Malan, 1979) constitutes a universal framework for psychodynamic therapy (McCullough et al., 2003). Regardless of the specific problems presented by the patient, therapy focuses on identifying and managing defence mechanisms, experiencing healthy emotions without the burden of negative emotions, improving the quality of interpersonal relationships and fostering a more positive self-image (McCullough et al., 2003).

In psychotherapy, applying a ‘treatment method’ involves using theory to construct a narrative about the patient’s problems and then adopting problem-solving strategies that align with that theory (Truscott, 2010). In other words, choosing a treatment method is an implicit endorsement of a theoretical model that explains why mental disorders occur and persist, while relegating other theories to the background (Allport, 1961). This may partly explain why the issue of choosing treatment methods has sparked irreconcilable ‘culture wars’ in psychotherapy research (Norcross et al., 2006; Ryum & Halvorsen, 2014). In 1907, William James proposed an approach to mediating between theories that give conflicting answers about what is true. In Pragmatism (James, 1907/2012), he challenged the notion that an objective truth exists about the human psyche, shrouded in darkness and waiting for the right theoretical spotlight to reveal it. Instead of trying to prove that something is objectively true, consideration should be given to its usefulness: ‘Ideas become true just in so far as they help us to get into satisfactory relation with other parts of our experience’ (James, 1907/2012, p. 28). James also believed that the criterion for judging whether something is true should be considered based on the consequences of choosing one over the other: ‘If no practical difference whatever can be traced, then the alternatives mean practically the same thing and all dispute is idle’(James, 1907/2012, p. 22). From a pragmatic perspective, the focus shifts from whether humanistic or psychodynamic theory provides the true explanation of psychopathology to whether the methods derived from these theories are effective in practice.

Clinical psychology thus operates with multiple and diverse universal explanatory models for mental disorders, and none of them can be shown to be more valid than others (Melchert, 2016). Consequently, choosing treatment methods based on theory becomes more a matter of adherence to interpretive traditions than one of scientific foundation: the theories serve as a basis for understanding and interpreting patients’ problems, but none can be considered more objectively true than others. The foundational report for EBPP does not take a position on questions about psychological theories. Instead, it asserts that the purpose of evidence-based practice is to foster effective psychological practices by learning what works for whom (APA Presidential Task Force on Evidence-Based Practice, 2006). This can be understood as neo-pragmatism, which posits that the primary task of psychological science is to document which practices produce desired outcomes as opposed to identifying universal truths (Messer & Wachtel, 2009). Thus, as with EST, the question of what ‘works’ is fundamental in EBPP. However, EBPP has shifted its focus away from the question of which treatment methods work for specific disorders, due largely to the difficulties encountered when studying the effectiveness of treatment.

Efficacy, effect and evidence-based practice

To show whether something ‘works’ in science, the fundamental task is to establish efficacy. Efficacy involves uncovering causal relationships and identifying the prerequisites for claiming that something (e.g., a treatment) causes something else (e.g., a reduction of illness/improved health). In order to discuss efficacy, certain prerequisites must be met: it must be possible to describe what causes the change and what exactly is changing.

The question of cause can quickly become complex, even in a case of what seems to be a straightforward causal relationship. For example, a lit match can cause a fire, but it can only be deemed to be the cause if the necessary supporting factors that enable the effect (oxygen, combustible material) are present. Oxygen and combustible material cannot create a fire without the ignition charge. The match is therefore a necessary but insufficient condition for causing the fire, and it is the presence of other supporting factors that creates an environment sufficient to enable the lit match to cause an effect. Furthermore, other possible causes with their own combination of supporting factors can also create a fire (e.g., an electrical short circuit). A similar situation exists in clinical psychology in that various treatment methods can lead to change for the same problem, provided that certain supporting factors are present. However, determining efficacy becomes much more complicated because we do not know exactly what we are treating or exactly which combination of supporting factors is necessary to enable treatment effect.

Effect of treatment

Discussions of efficacy in clinical psychology often involve the assumption that psychotherapy is the cause of an observed change. A prerequisite for this assumption is that the treatment context allows for the realisation of the outcomes that the therapy seeks to achieve. For example, psychotherapy aimed at improving relationships relies on the patient having people they are close to. Similarly, the patient needs to have the necessary financial resources and living conditions to dedicate sufficient time and effort to engage in therapeutic work to create change. In other words, some fundamental contextual supporting factors often need to be in place for psychotherapy to work (Cartwright & Hardie, 2012). Additionally, the treatment itself consists of strategies and techniques that are specific to a defined treatment method, as well as factors common to all effective treatments, such as the therapeutic alliance, expressed empathy and structured feedback (Cuijpers et al., 2019). Although there is disagreement about whether patient improvement in psychotherapy is primarily attributable to method (Barlow, 2004) or to common factors (Wampold, 2015), there is general consensus that both are necessary ingredients (Mulder et al., 2017). However, research has not been able to clarify how individual contributions of common factors and method factors contribute to improved outcomes in psychotherapy (Cuijpers et al., 2019). In other words, if research establishes the efficacy of a given psychotherapeutic treatment, the challenge of interpreting this finding persists; we cannot determine whether the effect is due to the method or to common factors, or which combination of supporting factors was needed to enable the treatment to have an effect (Cartwright & Hardie, 2012; Shadish et al., 2002). Thus, even if empirical evidence shows the efficacy of a treatment, it is not possible to know which specific parts of the treatment or its context actually caused the change and which parts enabled the cause.

Effect on mental disorders

In order to determine the efficacy of a treatment, there must be identifiable problems that the treatment successfully addresses. If, for example, an attempt is made to show that a treatment can successfully heal the phenomenon we call ‘depression’, it must be assumed that the symptoms – despite varying from patient to patient – are caused by underlying mechanisms that behave similarly in everyone with those symptoms (Rawlins, 2008). This is in line with a medical disease model where observable or self-reported problems are thought to be caused by an underlying condition and where it is this condition that is affected by the treatment (Hyland, 2011). This fundamental understanding of illness has been applied in ESTs in attempts to identify which methods are effective for different disorders (Wampold & Imel, 2015). However, most mental disorders deviate from this premise, and it has not been possible to identify underlying causes of mental disorders based on diagnostic classification (Bentall, 2004; Kendler et al., 2011). When symptoms cannot be systematically linked to an underlying cause, it means there may be multiple possible factors causing them in each individual case (Kendler et al., 2011). When a treatment proves to be effective for certain types of diagnoses, we can assume that action has been taken that has alleviated the symptoms, but in terms of why the symptoms occur and persist, we do not fully understand what we have treated or what the treatment has actually been effective for.

When trying to base the choice of method on empirical evidence of treatment efficacy, significant challenges thus arise in interpreting such findings, since we are unable to clearly define what parts of a treatment cause or enable change or what it is effective for (the disorder). How then can clinicians ensure that they select treatment methods that align with EBPP? Some proposed solutions are presented and discussed below.

Role of treatment methods in EBPP

Over forty years of efficacy research indicate that psychotherapy, based on a wide range of different theories, is effective for a broad spectrum of disorders (Barkham & Lambert, 2021). While we do not know which particular aspects of psychotherapy cause change or what, specifically, it is effective for, we can say with reasonable certainty that psychotherapy is effective for something. Research has also shown that there are minimal differences in the efficacy of various psychotherapeutic methods (Barkham & Lambert, 2021; Wampold & Imel, 2015). It could therefore be argued that neither the method nor the classification of a problem is important, and that ‘all methods of therapy when competently used are equally successful’ (Rosenzweig, 1936, p. 413). If this were the case, evidence-based practice would primarily entail identifying what makes therapists competent in practising such skills. For example, experiencing a trusting relationship with the treatment provider may be a universal need, regardless of the problem or symptom. This may also extend to the need to experience a sense of mastery and positive emotions that can foster hope for change and provide the motivation to pursue it, or to the need to have a safe space to practise what needs to be changed (Frank & Frank, 1991). Psychotherapy will thus be effective if a therapist creates an empathetic trusting relationship, instils hope and positive expectations that the treatment will lead to change, and offers specific strategies or actions to break dysfunctional thought, emotional or behavioural patterns (Wampold & Imel, 2015). These elements are referred to as common factors; aspects that, regardless of theoretical and methodological orientation, are essential for positive outcomes in all psychotherapy (Wampold, 2015). Some common factors entail characteristics or skills that are not directly linked to the treatment method, such as interpersonal skills (Anderson et al., 2009), alliance (Flückiger et al., 2018), empathy (Elliot et al., 2018) or unconditional positive regard (Farber et al., 2018). However, it is also worth noting that common factor models explicitly describe the necessity of using a therapeutic method that includes specific therapeutic strategies or actions (Mulder et al., 2017). Without knowledge of and training in a method, it is difficult to envision how a therapist can consistently provide a cogent rationale (i.e., a theory) explaining the patient’s problems, establish a shared understanding of the treatment goals and tasks and, consequently, foster hope and positive expectations for change. Furthermore, without training in one or more methods, challenges can also arise in proposing specific approaches to solving problems according to the explanatory models being applied. A cornerstone of EBPP is therefore that, as well as acquiring general therapeutic skills, therapists must also learn at least one treatment method.

Impact of the method on problems

Various forms of psychotherapy can be hypothesised to utilise the universal human ability of a subjectively experienced self to recognise, understand and change other aspects of self. To observe their own thoughts and feelings, people also rely on an irreducible sense of self that carries out this observation (Skjervheim, 1976). A common feature of several therapeutic methods is the use of theoretical concepts to help patients distinguish dysfunctional thoughts, feelings and behaviours from healthy ones and then change them. For example, metacognitive therapy (MCT) uses the construct ‘metathoughts’ to explain how a person can control and manage other dysfunctional thoughts (Nordahl, 2014). In acceptance and commitment therapy (ACT), the concept of ‘cognitive fusion’ is used to elucidate the potential difficulty of distinguishing between thoughts that are inappropriate representations of reality and thoughts that correctly reflect the world (Holden & Lenndin, 2014). In mentalisation-based therapy (MBT), the concept of ‘psychic equivalence’ is used to show how the patient’s experience of the world around them is not necessarily the same as the reality (Skårderud & Sommerfeldt, 2014). By using a psychotherapeutic method, the therapist interprets the patient’s problems through the theoretical concepts associated with that method, and such interpretations function as a schema that helps the patient organise and understand their problems (Hoffart, 1997; Rosenzweig, 1936). This schema can be seen as one of several ways to formulate an understanding of the problems, with each alternative offering different but potentially useful approaches to unravelling the complex psychological phenomena being treated (Rosenzweig, 1936). The different schemas can be considered equivalent approaches because mental disorders typically involve many different problems that influence and sustain each other in a mutually reinforcing network (Borsboom & Cramer, 2013). For example, sleep problems can reduce concentration and cognitive functioning, while rumination and worry can lead to sleep problems (Fried & Nesse, 2015). The fact that symptoms mutually influence each other may explain why between 73.8% and 98.2% of patients with a mental disorder also have at least one other disorder (Gadermann et al., 2012), and why having had one mental health diagnosis at some point in time increases the risk of developing others in the future (Plana-Ripoll et al., 2019). It can also mean that successfully influencing some symptoms can create positive ripple effects for other symptoms that the specific treatment does not directly address. In such a scenario, ‘it would matter relatively little whether the approach was made from the right or the left, at the top or bottom, since a change in the total organization would follow regardless of the particular significant point at which it was attacked’ (Rosenzweig, 1936, p. 414). A theoretically derived understanding of a patient’s presenting problems does not, therefore, need to fully encompass all the difficulties the patient has in order to be effective (Rosenzweig, 1936). This is supported by recent research, which indicates that different treatment methods can have direct and indirect effects on specific symptoms, and that improvement in certain types of problems can lead to subsequent improvement in others (Boschloo et al., 2019). Thus, even though treatment methods differ theoretically and methodologically, they can end up producing roughly the same positive outcomes (Boschloo et al., 2019; Dunlop et al., 2018; Stewart & Harkness, 2012).

From this perspective, the most important aspect for the clinician is that the method’s theoretical explanatory models and specific strategies facilitate the patient’s ability to identify, understand and address core elements of their set of problems.

Method choice based on efficacy research

The APA presidential task force statement on evidence-based practice states that EBPP must be based on the best available research (APA Presidential Task Force on Evidence-Based Practice, 2006). How, then, can research influence the choice of method? If it is not particularly important which method is chosen, can a therapist choose any treatment approach they like and still legitimately claim their practice is evidence based? A key point here is that, in order for something to be considered a treatment method, it must at a minimum include a theoretical model that explains how mental disorders work and describe therapeutic actions that align with such an explanation (Wampold et al., 1997; Wampold et al., 2010). Efficacy studies indicate that such methods often lead to positive outcomes compared with no treatment at all (Wampold & Imel, 2015). This means that treatment methods should be able to demonstrate efficacy at group level, serving as a minimum standard for therapists and service providers when considering incorporating a treatment method into their service delivery system. Although efficacy studies do not provide a basis for concluding what precise aspects of the treatment are effective or exactly which problems or processes it successfully targets, they nevertheless provide a robust indication of whether the method leads to improvement. Treatment methods that fail to demonstrate an effect at group level in such studies will probably not be helpful and could even be harmful. One implication is that efficacy studies should underpin decisions about whether to include specific treatment methods as a standard part of the service delivery system.

A prerequisite for conducting efficacy studies is testing the effectiveness of the method for specific types of problems, and in this regard diagnostics becomes relevant. Although diagnoses cannot be considered valid disease categories, they are, nevertheless, concrete descriptions of psychological issues, and research on the efficacy of methods tends to be based on these descriptions. Performing diagnostic investigations allows practitioners to reference existing research to identify which treatment methods have proven successful for patients with similar symptoms. Although the method cannot be said to cure specific diagnoses, the evidence shows that certain types of treatment methods have been effective for people with similar problems. Thus, efficacy research and diagnostics can help narrow down which treatment methods might be worth incorporating into a service delivery system.

Hypothesis-based method testing

Although efficacy studies show that many treatment methods are effective at group level, many patients do not achieve lasting improvement. For example, in studies that show significant effect on depression, 50% of patients will not improve (Craighead & Dunlop, 2014; Cuijpers et al., 2021). Effect at group level means that a treatment is beneficial for many but far from all. Even if a particular method (e.g., MBT) has been shown to have an effect on a disorder (e.g., a personality disorder (PD)), it does not mean that everyone with PD must receive MBT. Since the mechanisms driving change are not fully understood and since there may be a variety of problems underlying the diagnostic categories, it is not possible to know whether the next patient will benefit from the treatment or not. EBPP therefore requires that treatment be tailored to the patient’s characteristics, preferences and goals (APA Presidential Task Force on Evidence-Based Practice, 2006). A key consideration for using a treatment method is therefore how to evaluate its effectiveness for the individual patient throughout their treatment process. If we assume that the treatment method acts as a mental schema for understanding problems with accompanying strategies for problem solving, it is essential for its effectiveness that the patient understand, accept and perceive the method as meaningful for their problems (Wampold & Imel, 2015). The therapist’s ability to articulate and explain the rationale for the treatment in a convincing manner is therefore a key therapeutic skill (Anderson et al., 2009). When patients view a treatment method as logical, appropriate and effective, they also benefit more from the therapy (Constantino et al., 2018). Additionally, the treatment method’s explanatory model for mental disorders will be pivotal when formulating specific treatment goals. An important prerequisite for benefiting from treatment is that the patient agrees on which goals to aim for (Tryon et al., 2018). This indicates the importance of regular dialogue with the patient about their perceptions of the rationale for the treatment method, in order to establish a mutual understanding of whether the underlying assumptions for the treatment make sense to the patient (Constantino et al., 2018; Hoffart, 2003).

Within the theoretical framework of a treatment method, there are usually many possibilities for adapting the method to the patient’s personal worldview (Truscott, 2010). However, it may also be that the patient simply does not find the method’s explanatory models meaningful, thus rendering attempts at such adaptation alienating (Flaaten, 2021). In such cases, proceeding with specific therapeutic actions is unlikely to be beneficial, even if they are well presented within the theoretical framework. This is supported by studies showing lower dropout rates and better outcomes for treatment methods for which patients have expressed a preference (Swift et al., 2018).

The reciprocal influence between psychological symptoms means that many different methods can be effective for a given problem. For example, sixteen different treatment methods with different research-based theoretical foundations are supported by evidence demonstrating their effectiveness for treating depression (Division 12 of the American Psychological Association, 2021). However, it is also crucial that the rationale for the method actually helps the patient understand and begin the process of managing their set of problems. Whether it does or not is often not known until treatment is under way. An implication here might be that evidence-based practice should be conceptualised as a dynamic process of hypothesis-based testing of methods in consultation with the patient, where initial testing should largely be seen as part of an investigation phase.

Summary

Evidence-based practice is defined as ‘the integration of the best available research with clinical expertise in the context of patient characteristics, culture and preferences’ (Norwegian Psychological Association, 2007, p. 1127). I have focused on the use of treatment methods within this definition and concluded that the use of one or more treatment methods is a prerequisite for evidence-based practice. I have further concluded that research indicates that different treatment methods often work equally well, but that research should establish the efficacy of a treatment before the decision is made to incorporate it into a service delivery system. Additionally, I have emphasised that evidence-based practice in the application of treatment methods depends on the patient’s understanding and acceptance, and that regular dialogue is needed with the patient about their perceptions of the treatment method’s rationale in order to establish a mutual understanding of whether the underlying assumptions for the treatment are valid for them. Evidence-based treatment should therefore be viewed as a hypothesis-based testing of methods, where outcomes are continuously evaluated together with the patient, and where the treatment process may entail repeated attempts using different methods.