-
The importance of evidence based practice
- Psychology is invisible, abstract and difficult to define
- Performing psychotherapy is an art more than science.
- Therefore:
- ▪ Psychology needs to emphasise it’s scientific credentials to ensure that it is taken seriously in the 21st century
- ▪ This means that a core part of the first 4 years of psychology training is the most rigorous research training available. The highest compared to all other health practitioners
-
Effective interventions are not always intuitive
- Psychological debriefing after a traumatic event:
- Single session debriefing actually shown to be ineffective or even harmful
- Playing tetris after traumatic event beneficial.
Not just a problem in psychology. A popular knee surgery (arthroscopy) has been shown to be ineffective
-
Bias in thinking
- Psychologists are experts in thinking biases, but this does not make us immune.
- Our cognitive biases and our instinct to rationalize our beliefs with logical fallacies prevent us from seeing how things actually are.
- ▪ We need quality scientific evidence to reduce the chance of bias
- ▪ Bias in thinking and perceiving. Looking straight ahead at the lecture screen see if you can perceive the colour of the person’s clothes next to you.
- ▪ Colour vision stops at about 40 degrees.
- ▪ We cannot just rely on personal experience
-
Why we might think treatment is effective when it isn't
- 1. Due to natural course of diseases, regression to the mean, and the placebo effect, real signs and symptoms often improve- with or without treatment.
- 2. Patients and practitioners often convince themselves that treatment was effective- when it was not (due to confirmation bias and others human cognitive imperfections)
- 3. Personal evaluation of efficacy is quick and convincing, but properly controlled, scientific determinations can be slow, complex and costly
-
Efficacy of Psychotherapy
- Psychotherapy is more effective than no treatment
- ▪ Mega-analyses (meta-analyses of meta-analyses) (Grissom, 1996)
- ▪ Average effect size of Cohen’s d-.80. A “large” effect
- ▪ In other words, 69% “success rate”
- ▪ 15% of the outcome variance accounted for by therapy
- ▪ More effective than influenza vaccine, statins in cardiology, some surgeries
-
Relative Efficacy of Psychotherapies
- Are there differences in outcomes between treatment types?
- ▪ Mega-analyses of 32 meta-analyses of treatment approaches
- – Differences between therapies yielded an effective size of d=.20
- – A “small” effect accounting for tiny level of variance in the outcome
- – Results are consistent for anxiety and depression
- ▪ When researcher allegiance is taken into account, the relative efficacy seems to be negligible (d = 0 to .17)
- ▪ Dropout rates are equivalent across treatments (about 19%)
- ▪ Very controversial
-
-
Psychotherapy versus antidepressants
- Both psychotherapy and SSRIs are more effective than placebo in severe depression
- Psychotherapy and SSRIs do not differ in effectiveness for moderate to severe depression.
- -Depends on severity, chronic/non-chronic
- – Psychotherapy better in the long term
- – In under 18s therapy is better
- Psychotherapy has better long term outcomes at follow up
- Drop out rates are lower for psychotherapy.
- Psychotherapies create less resistance to multiple administrations. (ei. drop off in efficacy)
-
The placebo effect in psychotherapy
- The placebo effect is any treatment or component of treatment that is deliberately used for its nonspecific, psychological, or psychophysiological effect.
- The placebo effect may have greater implications for psychotherapy than any other form of treatment because both psychotherapy and the placebo effect function primarily through psychological mechanisms.
- Generally good research should demonstrate superiority over placebo (placebo controlled randomised controlled trial).
- It is nearly impossible to develop a truly suitable psychological placebo condition!
-
Problems with placebo in psychotherapy
- 1. A placebo is a theoretically inert intervention
- – in psychotherapy just talking about problems is theoretically important (in some theories)
- – Not a problem with typical placebos like sugar pills
- 2. In psychotherapy, all change is due to psychological factors
- 3. If a psychological placebo is administered and change occurs, there must exist a psychological mechanism responsible for that effect, whether or not we have a current theory to explain it. This “placebo” is thus not inert, a statement that is self-contradictory.
- 4. Cannot do a “double blind” study because in psychology research the therapist would know they are acting in a placebo condition. Therapist bias becomes a confound
- ▪ It turns out to be very difficult to create a psychological placebo condition whose credibility and expectancy generation in the client are equivalent to that of actual therapy conditions
-
A move away from placebo in psychotherapy research
- ▪ Instead, use wait list controls
- ▪ Or treatment as usual
-
Factors that reliably lead to positive outcomes that are:
– Inherent in any therapeutic situations, and
– Not specific to any one therapy approach
- Relationship: alliance, client feedback
- Therapist skills: allegiance to the therapeutic approach (belief in effectiveness); empathy.
- ▪ Client: Severity, expectations of benefit (a.k.a., placebo); belief in credibility of therapy and therapist
- ▪ Extra therapeutic
-
Common Factors: Therapist
- Who the therapist is accounts for about 7% of variance in patient outcomes
- 11% and 38% of therapists on average had patients who ended therapy worse off than when they started.
- Between 29% and 67% of therapists had patients who reliably got better.
- Factors that matter:
- ▪ Therapists belief in credibility of effectiveness of psychotherapy. D=.65 (medium)
- ▪ Empathy, as rated by client (effect size = .74, medium)
- ▪ Treatment rigidity is harmful
-
Collecting patient feedback
- Regularly monitoring and tracking patient treatment response with standardized self report scales.
- – Provide therapist with week to week summary of patient functioning
- – Improves therapist responsiveness to patients who are not getting better
- Therapist are notoriously poor at seeing who is getting better without measures.
- • In a large study (Lambert, 2007) of treatment efficacy, 20% of patients deteriorated, but only 8% of patients who deteriorated were identified by therapists!
- – Effects of therapists collecting client feedback on outcomes = medium effect size (d=.55)
- – Providing therapists with feedback (rather than just collecting without therapist monitoring) reduced the number of patients who deteriorated by half! (20% vs 9%)
|
|
