Sources of evidence about which therapies help
Scientific research studies
-Using systematic quantitative or qualitative methods
Synthesis of research literature
- Metanalyses
- Clinical Practice Guidelines
Consumer surveys
-Aggregation of independent feedback from those using the target service (by researchers or lobby groups)
Lived experience
- Writing and advocacy of individuals
- Advocacy of consumer groups
Practice-based evidence
- Aggregating outcome data from ordinary practitioners who are trying the intervention (not opinion)
Charman & Barkham (2005) Evidence-based practice
Evidence-based practice ->Top-down
Practice-based evidence -> Bottom-up
Evidence and the social gradient
Randomised Controlled Trials (RCTs)
‘Gold Standard’
Which treatment and which dose are best for a particular mental disorder?
- DSM-IV or ICD-10
- Manualised treatments
- Random allocation
- outcomes include recovery and relapse rates
“It is assumed that the active ingredient is the treatment”
Asay and Lambert (1999)
Factors accounting for patient improvements
Factors accounting for patient improvements 30% Theraputic relationship 40% Extratheraputic change 15% expectancy effects 15% theraputic techniques
National Institute of Mental Health
Elkin, 1994
The Treatment of Depression Collaborative Research Program.
- All groups (CBT, IPT, SSRI+CM and Placebo+CM) showed improvement.
- Therapist efficacy appears more significant than treatment efficacy.
Evidence-Based Practice three pillars
- Research Evidence
- Clinical Expertise
- Patient specifics
Shadish, Clark and Steiner (2008)
Can Nonrandomised Experiments Yield accurate Answers? A Randomised Experiment Comparing Random and Nonrandom Assignments
Initially (N=445) they completely randomly assigned them into either a “randomly assigned” group or a “self selected group”.
They either got to choose whether to learn maths or vocab or they didn’t get to choose maths or vocab.
Cumming (2012) video on p values
p values: If p reveals truth, a replication should yeild roughly the same p.
If you repeat an experiment how much will results vary?
Cannot trust p
RCTs pros and cons
Pros:
- easy to replicate
- removes some bias
- very structured
Cons
- where do we have flexibility to adapt our intervention to work for the client?
- people dropping out throws off the groups and sample size
8 Questions for critically
appraising the evidence
- Is the study’s research question relevant?
- Does the study add anything new?
- Was the study design appropriate for the research question?
- Did the study methods address the key
potential sources of bias?
- Meta-analyses – publication bias
- RCTs – lack of allocation concealment
- Cohort studies – identifying and measuring all important confounding factors
- Case studies – relevance - Was the study performed in line with
relevant protocols? - Were the statistical analyses performed
correctly for the stated hypothesis? - Do the results justify the conclusions?
- Are there any conflicts of interest?
