The efficacy and effectiveness of psychotherapy - summary of chapter 6 of Bergin and Garfield’s Handbook of psychotherapy and behavior change By Lambert, M.J.

M. J. Lambert (Ed.), Bergin and Garfield’s Handbook of psychotherapy and behavior change
By Lambert, M.J. (2013).
Chapter 6
The efficacy and effectiveness of psychotherapy.

The effectiveness of psychotherapy

Is psychotherapy efficacious?

Cohen’s d is the standardized difference between the mans of groups.
The aim of d and related statistics is to describe the magnitude of treatment response.
An effect size of 0 indicates the complete lack of differences.
An effect size of 1.0 indicates that one group, on average, is one standard deviation superior to the other group.

Meta-analysis is essentially a statistical means to test hypotheses by synthesizing the results of a set of studies addressing the same research question.
In meta-analysis an effect is calculated for each study and then aggregated. The aggregate is then tested against zero.
If an aggregate effect for treatment versus no-treatment is significantly greater than zero, it can be concluded that the treatment is more effective than no treatment.

Meta-analysis can also be used to determine whether there are moderators of the effects obtained from the various studies.
Meta-analysis can be used to identify and test moderating variables.

Broad meta-analysis of therapy efficacy

The evidence from meta-analysis indicates that the psychological, educational, and behavioural treatments studies by meta-analysis generally have positive effects.

The use of meta-analysis to summarize efficacy literature is critical for the field.

Meta-analysis are just as prone to poor methods and misinterpretations as other methods of research.
There are three main threats to their validity

• File drawer problem
  The tendency for studies with small or no effects to never be published
• The garbage in, garbage out problem
  Mixing poor-quality and high-quality studies
• The apples and oranges problem
  Combining studies of very different phenomena

Meta-analysis focused on particular disorders

Mood disorders

Numerous meta-analytic reviews suggest that patients undergoing many diverse kinds of psychotherapy for depression surpass no-treatment and wait-list control patients.
Results in treating depression have shown that most psychological treatments that have been studied produce substantial effects, in terms of symptom reduction, and increased well-being.
Psychological treatments are effective in specific populations, including adults, older adults, women with postpartum depression, and patients with both depression and general medical disorders.

Psychotherapy has a small but significant effect on chronic depression when compared to control groups.

Adding psychotherapies to a mood stabilizer regimen in bipolar disorder has been shown to reduce rates of relapse over 1 to 2 years.
Therapies that specifically target increases in medication adherence, teach self-monitoring and early intervention with emergent episodes, and enhance interpersonal functioning ad family communication help to prevent relapse.

Overall, the effects of psychotherapy and pharmacology are comparable at termination, although there is some evidence that medications may be superior to psychotherapy for dysthymia.
Mediation can prevent depression symptom relapse as long as medication is not discontinued.
The combination of psychotherapy and medication appears to be somewhat more effective than treatment with pharmacotherapy alone.
Pharmacotherapy, especially selective serotonin reuptake inhibitors, for those with chronic depression, appears to be more effective than psychotherapy alone.
Combined treatments appears to be more effective than pharmacotherapy alone.

CBT treatments exhibited an average relapse rate of 29.5%, while the antidepressant medication groups average relapse rate at 60%.

Anxiety disorders

The most dramatic and consistent findings over time have been reported with the behavioural and cognitive-behavioural treatment of panic disorder.

Generalized anxiety disorder (GAD)

Patients with CBT have been shown to benefit from treatment.
On average, about 50% of clients with GAD achieve high end-rate functioning.
CBT and pharmacotherapy seem to be equally effective.

CBT for GAD has failed to include interventions that target emotional avoidance and discomfort.
Adding techniques specifically designed to help CAD clients deeply experience and process uncomfortable emotions may help them reduce their chronic worrying.

Posttraumatic stress disorder (PTSD)

Psychological treatments from various theoretical perspectives have been found to be effective for PTSD in pervious reviews.
Specific therapies, such as CBT, exposure-based therapy, and cognitive therapy are equally effective. They are more effective than supportive techniques.

PTSD symptoms will improve through trauma-focused CBT (TF-CBT) and EMDR.
Trauma-focused treatments are superior to non-trauma-focused treatments for PTSD.

Obsessive compulsive disorder (OCD)

Patients respond to exposure and response prevention for treating OCD.
An active ingredient includes exposure to situations that provoke obsessions and compulsive acts, while inhibiting the expression of the compulsive act.

Social anxiety disorder

This is a disorder where group psychotherapy comes into prominence.
Treatments for social anxiety are moderately to highly effective.

Do patients make changes that are clinically meaningful?

The effect size statistics that are derived from the typical efficacy study estimate change in standardized units.
Discussing change in this way obscures whether the change actually made by each patient is clinically important.

The most popular statistical method for estimating meaningful change possess two criteria that each patient must meet before their change is considered meaningful

• Treated clients make statistically reliable improvements as a result of treatment
• Treated clients are statistically indistinguishable from ‘normal’ or non-deviant peers following treatment

Cost offset

A related body of research that suggest treatment effects produce clinically significant change and broad effects across a person’s life is that on medical cost offset.
Medical cost offset studies suggest that many patients who enter psychotherapy have concurrent medical and psychological needs.
Those who participated in psychological interventions were less likely to use inpatient medical services.
The savings were greatest among those patients who over utilized services and for behaviour medicine interventions, but can also be found across treatment types and particular medical and psychological conditions.

Life-functioning outcomes can result from psychological interventions.

Does therapy exceed placebo?

Many authors entirely reject the placebo concept in psychotherapy research because it is not conceptually consistent with testing the efficacy of psychological mechanisms of change.

Placebo’s have also been labelled nonspecific factors.
Contemporary psychotherapy outcome studies seldom use the placebo term, preferring instead to control for all factors that are common to all treatments not just expectancy effects.
Others have suggested the term common factors in recognition that many therapies have ingredients that are not unique to a specific treatment but healing nonetheless.

Psychotherapy surpasses placebo effects sometimes by large margins.
It goes beyond the installation of hope.
The effects are variable.

Do patients maintain their gains?

Many patients who undergo therapy achieve healthy adjustments for long periods of time.

Maintenance of change seems unrelated to specific type of treatment, or even treatment length.

Strategies to maximize maintenance

Factors that influence relapse rates (many in MDD)

• Number of depressive episodes
• Severity of initial depressive episode
• Response to acute treatment
• Residual symptoms following treatment
• Psychopathological and medical comorbidities
• Family history of depression and other psychopathology
• Treatment type and duration
• Treatment adherence
• The therapeutic alliance

Minfulness-based cognitive therapy (MBCT) was developed specifically to reduce relapse and recurrence in depression.

Relapse rates for psychotherapy (27%) are almost half when compared to pharmacotherapy (57%) in depression, unless pharmacotherapy is continued.

How much therapy is necessary?

Patients remain in therapy until they recover is a good description of the relationship between dose and individual  patient recovery.

Although patients improved during treatment, patients’ rate of change varied as a function of total dose of treatment.
Total dose had a nonlinear relationship with the likelihood of clinical significant change.

A sizable portion of patients reliably improve after 7 sessions.
75% of patients will meet more rigorous criteria for success after about 50 sessions of treatment.
Limiting treatment sessions to less than 20 will mean that about 50% of patients will not achieve a substantial benefit from treatment.  

Sudden gains and related patterns of change

An individual’s recovery path is likely to differ from the group mean in some way.
Many different courses of progress are present even with similar disorders, seemingly similar client, and similar psychotherapies.

Three requirements of sudden gains. An intersession change that is:

• Large in absolute terms
• Large relative to depressive symptom severity before the gain
• Large relative to symptom fluctuations preceding and following the gain

Typically, sudden gains accounted for at least 50% of the total change in the course of therapy.

Do some patients get worse?

Some patients are worse at the time therapy is terminated than when they started.
Negative outcomes can be observed across a variety of treatment modalities as well as across theoretical orientations.
This does not mean that all instances of worsening are the product of therapy.

About two-thirds of adults who enter treatment in RCTs have a positive outcome in about 14 sessions, but about a third either show no benefit or worsen.

In general, those who did not respond to treatment provided special challenges to practitioners and these tended to take three general forms

• Poor motivation
• Complicated problems
• Resistance to therapist prescriptions and suggestions

Therapist mistakes in delivery ay have interfered with patient outcome.

In order to enhance treatment outcomes, treatment protocols needed to cover more aspects of dysfunction and more treatment targets tailored to specific clients.

The literature on negative effects suggests that although the studies contain many methodological shortcomings and ambiguities, the evidence that psychotherapy can and does harm a portion of those it is intended to help is substantial.

Does efficacy research generalize to practice?

The efficacy of treatment is determined by a clinical trial or trials in which many variables are carefully controlled in order to demonstrate that the causal relationship between the treatment and outcome are relatively unambiguous.

Efficacy studies emphasize the internal validity of experimental design through a variety of means including:

• Controlling the types of patients included in the study
• Using manuals to standardize treatment delivery
• Training therapist prior to the study, monitoring therapist adherence to the treatment during the study, and supervising therapist to ensure they do not deviate from the treatment protocol
• Managing the ‘dose’ of treatment through analysis that include only patients who have received a specified amount of treatment
• Random assignment of clients to treatment
• The use of blinding procedures for raters

The effectiveness of a treatment is considered in clinical situations when the intervention is implemented without the same level of internal validity.
Effectiveness studies emphasize the external validity of the experimental design and attempt to demonstrate that the treatment can be equally beneficial in a clinical setting.
Typically

• Clients are not as carefully selected
• Treatment dose is less controlled
• Therapist adherence is neither monitored nor modified to be a pure-form, manually determined treatment
• Therapist tend to be those working in the settings and may or may not receive the same level of prestudy training as that of the efficacy study

It appears that RCTs can and do generalize to clinical settings, but do not always.
The likelihood of this occurring depends to a great deal on the degree to which the studies are similar in design to the RCT.

Comparison and causative factors

Is one treatment more effective van another?

The use of manuals to specify treatment activities characteristic of the different schools results in objectively discriminable therapist behaviours that are true to conceptions of what is wrong and how it can be changed.

In general, the results from comparative, dismantling, and components analysis studies suggest that general equivalence of treatments based on different theories and techniques.

Is one therapist more effective than another therapist?

There are differences of effectiveness between therapists.
Three therapist qualities were identified as distinguishing the more helpful form the less helpful therapists

• The therapist’s adjustment, skill, and interest in helping patients
• The purity of the treatment they offered
• The quality of the therapist/patient relationship

The way particular therapists characteristically react to challenging client presentations partially explains why some therapist are more effective than others.

Common factors and outcome

The general finding of no difference or very little difference in the outcome of therapy for clients who have participated in highly diverse therapies has a number of alternative explanations

• Different therapies can achieve similar goals through different processes
• Different outcomes do occur but are not detected by past research strategies
• Different therapies embody common factors that are curative, though not emphasized by the theory of change central to a particular school.

Common factors

• Support factors
  Catharsis/release of tension
  Mitigation of isolation
  Structure/organization
  Positive relationship
  Reassurance
  Safe environment
  Identification with therapist
  Therapeutic alliance
  Therapist/client active participation
  Recognition of therapist expertness
  Therapist warmth, respect, empathy, acceptance, genuineness
  Trust/open exploration
• Learning factors
  Advice
  Affective re-experiencing
  Assimilating problematic experiences
  Cognitive learning
  Corrective emotional experience
  Feedback
  Insight
  Rationale
  Exploration of internal frame of reference
  Changing expectations of personal effectiveness
  Reframing self-perceptions
• Active factors
  Facing fears
  Cognitive mastery
  Encouragement of experimenting with new behaviours
  Taking risks
  Mastery efforts
  Modeling
  Practice
  Reality testing
  Success experiences
  Working through
  Behavioural/ emotion al regulation

The therapist-patient relationship is a common factor critical to positive outcome.

Relationship variables not only predict positive change but may produce it as well.

The search for larger effects

Using feedback and problem-solving tools to enhance treatment outcomes

Therapists were often unaware of patients’ unexpressed reactions.
Patients were particularly likely to hide negative feelings.

Conclusions

Research in psychotherapy outcomes has resulted in 12 important conclusions:

• Many formal, theory-driven psychotherapies have demonstrable effects on a variety of clients.
• Those clients that undergo formal treatment have better outcomes than individuals who are on a wait list or who receive no treatment
• Patients who undergo control treatments meant to simulate aspects of being treated show considerable improvement depending on the procedures used.
  • Bona fire interventions are superior at promoting change than control conditions
    The degree of inferiority depends on the nature of the control condition
• The effects of treatment are not only statistically significant but also clinically meaningful
• We now have better general estimates of the amount of therapy needed in order to bring about clinically meaningful change
  • For patients who begin therapy in the dysfunctional range, 50% can be expected to achieve clinically significant change after about 20 sessions
  • More than 50 sessions are needed for 75% of patients
• The effects of therapy tend to be lasting
• Although research continues to support the efficacy of those therapies that have been rigorously tested, differences in outcome between various forms of therapy are not as pronounced as might have been hoped for
• Given the growing evidence that there are probably some specific technique effects as well as large common effects across treatments, the vast majority of therapist have become eclectic in orientation.
• Positive affective relationships and positive interpersonal encounters, that characterize most psychotherapy and are common across therapies, still loom large as stimulators of patient improvement.
• Although the individual therapist can play a surprisingly large role in treatment outcome even when treatment is being offered within the stipulations of manual-guided therapy, recognition of the important place held be a therapist should in no way be construed as suggesting that technical proficiency has no unique contribution to make.
• The development and use of meta-analytic procedures for integrating outcome research is a methodological advancement that has enabled scholars and clinicians to better understand research findings.
• Although the broad, positive statements about psychotherapy can be made with more confidence than ever before, it is still important to point out that average positive effects mask considerable variability in outcomes.