(10) Research methods and evidence-based practice in clinical psychology Flashcards
What is evidence based medicine? EBM
- The concept is about making sure that when decisions are made they are made on the basis of the most up-to-date, solid, reliable, scientific evidence.
- In the case of medicine or health care, these are the decisions about the care of individual patients.
What are the aims of EBM?
Aim: aims to apply the best available evidence gained from the scientific method to medical decision making.
-How? By ranking evidence based on: the quality of studies, the strength of their findings
NICE guidelines for Mental Health and example?
- 23 complete guidelines
- Example: three sets of guidelines for depression:
- Adults
- Children and young people
- Depression with a chronic physical health problem
What is the process of EBM reviews?
- Search
- Read, include & exclude studies.
- Appraise, according to set criteria to establish quality.
- Synthesise where possible, by pooling results across studies
Based in appraisal (How is it assessed?)
- Of published papers: Depends in quality of search And on good critical criteria
- Laboratory tests
- Clinical experience
How are Randomised Controlled Trials assessed?
- The gold standard
- Randomisation: Controls for unknown bias and confounding
- Blinding: single and double
- Control group
How are RCT’s often flawed?
- Incomplete randomisation (drop-out, allocation bias)
- Length of trial? (e.g. dynamic versus behaviour therapy)
- Blinding patients? Assessor?
- Choice of control group? (Are waiting lists ethical?)
- Choice of outcome? (Days off work might be a primary outcome to government, but not to patients…)
A randomized clinical trial has four elements:
- A treatment group and a control group
- Randomization
- Blinding
- Ethics
What are a treatment group and control group in a radomized clinical trial?
A treatment groupand acontrol group.The treatment group receives an experimental treatment. The control group might be given no treatment at all, aplacebo, or “treatment-as-usual.” The idea is that the control group gives the clinician something to compare the experimental results to.
Why is randomization used?
Randomizationto allocate patients to one of the two groups. This randomization is normally performed by a computer.
Why is blinding used?
Blindingto prevent the patient or researcher from knowing what group they are assigned to. In a blind trial, the participants don’t know what treatment they are receiving (which may be a placebo). In a double blind trial, neither the participant nor the researcher knows which group the participant is in. Trials are often double-blinded to avoidselection biason the part of the researcher.
RCT can be problematic when…
- Clearly successful intervention (implemented already or reviewed in meta-analysis): replace with head to head or non-inferiority trials.
- Unethical (AZT)
- Very large subject groups needed, or when cases are very rare (Examining the effect of counselling on ‘unifying’ personality in people with split personality)
- Arguably, when treatment cannot be standardised (think which psychological model would fare better for standardisation)
What are cohort studies?
- Cohort studiesare a type of medical research used to investigate the causes of disease, establishing links between risk factors and health outcomes.
- Cohort studiesare usually forward-looking - that is, they are “prospective”studies, or planned in
What is a famous cohort study?
Hill & Dolls’ famous ‘Doctors Study’ started in 1951, finished in 2001, and provided strong evidence the smoking was closely linked to lung cancer.
Aspects of a cohort study
- Longitudinal.
- Baseline measures.
- No intervention, observe follow up over time, measure outcome.
- Inform about causality.
- Often large, costly, and time consuming
Example of a systematic review of prospective cohorts
- Examination of GP consultations with patients with uncertain causes/ treatments
- Split GP behaviours into affective and cognitive reassurance
- Examines patients outcomes at follow up
What is a case control study?
-A study that compares patients who have a disease or outcome of interest (cases) with patients who do not have the disease or outcome (controls), and looks back retrospectively to compare how frequently the exposure to a risk factor is present in each group to determine the relationship between the risk factor and the disease
Key aspects of a case control study?
- Retrospective. (looking back on or dealing with past events or situations)
- Select ‘cases’ and match with control group. ‘caseness’ is the outcome.
- Measure exposure in the past to suspected factor.
- If Odds are significantly higher in cases, exposure factor may have contributed to developing disorder
What is sudden infant death syndrome? SIDS
- Sudden Infant Death Syndrome (SIDS)
- Sudden, unexpected, unexplained
- Rare: less than 1 in 1000.
- How to investigate?…
- Evidence now supports brain abnormalities (in some), viral infection (in some), position face down (in some…)
What did Brown et al discover in a cohort study about schizophrenia and flu in the second trimester?
- Schizophrenia and flu in second trimester
- First identified from cohort correlation: flu epidemic curve and later incidence increase.
- How to study???
- Brown et al (2004) were able to analyse blood serum from expecting mothers (59 children with Schizophrenia,105 matched controls)
What are cross sectional/correlation?
- A snapshot in time: all variables measured simultaneously.
- Statistic tests inform on the relationship between two variables: significance and grade of relationship (high, moderate, low).
What is Causality- Bradford Hill criteria?
- Time line
- Plausibility of model
- Research data from different designs/ samples
- Quality of relationship (R, d’, p)
- Dose-response
- Reversibility
Problems for Evidence-based Psychology
- Measurements: surrogate (e.g. behaviour for cognition), self-report, validity (stress?), reliability, cut-points
- Can’t blind
- Often ignore / can’t measure main issue (communication)
- Multi-faceted problems
What issues are there with experiments?
- Can’t always generalise from animals (Seligman et al, 1978)
- Can learn about internal processes (e.g. Stroop in Spider Phobics)
- High on reliability but what about validity?
What are the issues with publication bias?
- Replication (not original)
- Refutation (Type II error)
- Political (from experimenters or editors)
What are case reports?
- Individual patients (e.g. Freud)
- Form of a story
- Can form a case series
- Evidence? Controversial
Pros and cons of case studies?
- Necessary for rare disorders: split personality or neuropsychology
- But no control for bias or confounding
- Empirical quality can be improved in single subject designs (experimental, base-line, manipulation, outcome)
What is the hierarchy of evidence?
- Systematic reviews and meta-analysis
- Randomised Control Trials
- Cohort studies
- Case-control studies
- Cross-sectional studies
- Case reports
Strong evidence for EBP
-Experiments
-Genetic research
-BUT most interventions are based on case-studies / experience:
Treatment is never the same for two individuals
-Most therapists ‘eclectic’
Why EBP is popular…
- Challenges clinical experience / authority
- Forces clinicians to continue learning
- Disputes value of clinical decisions based on: anecdotes, press cutting, expert opinion
Critical Appraisal of Method: Example of an item: How Participants need to be considered
- Who is the study about?
- Recruitment
- Is it representative? (APD in prisons)
- Are inclusion criteria well defined? GP case notes for the tag ‘depression’, rather than diagnostic criteria
- Exclusion: Have groups been included that could unduly sway the results? Women’s’ anxiety levels, not excluding pregnant
