PSYC10004 Lecture Notes - Lecture 33: Ship'S Doctor, Clinical Trial, Scurvy
7. The Clinical Trial
Evidence-based Practice in Psychology
• Clinical Psychology is an evidence-based discipline
• Therapies must demonstrate effectiveness before being offered
• Clinical trial is the method by which the best evidence is gathered
• Clinical tials eod the pusuit of tuth ad asks a siple uestio, does this teatet
ok?
First Controlled Clinical Trial
• Naval surgeon James Lind treatment
for scurvy
• 1746
• Lind sought evidence instead of
practice based on opinion
What is a Clinical Trial
• systematic, controlled empirical approach to determining the effectiveness or efficacy of a
therapy
• in context of mental disorder, this may pertain to psychotherapeutic (talking therapy like
CBT) or psychopharmacological (medications, like antidepressant medication) interventions
• to a point, type of clinical trial that can be performed is determined by the type of
intervention that is being tested
Internal Validity: degree of confidence
of the relationship between IV and DV
(treatment and effect on symptoms).
External Validity: to what extent findings
can be generalise
find more resources at oneclass.com
find more resources at oneclass.com
What amount of evidence is needed?
• Citeia fo a ell-estalished teatet
o Two high quality clinical controlled trials with active control groups
o Experiments include treatment manuals
o Client characteristics are specified
o Therefore, for whom the therapy is efficacious has been well defined
o Effects must be demonstrated by at least 2 different and independent investigative
teams
• Citeia fo a poal-effiaious teatment:
o Two clinical controlled trials showing the treatment is superior to a wat-list control
o One or more experiments meeting the well-established criteria, except that they
were completed by one investigating team
o A small series of single subject design experiments (n=3 to n=9) that meet the well-
established criteria
• Citeia fo poisig teatet
o Positive support from any of the following:
▪ One well-controlled clinical trial and second study of a lesser design or
▪ A small number of single-case design studies or
▪ Two or more well-controlled studies by the same investigator
What type of evidence is needed?
• Quantitative measurement of symptomatology
o Scores on appropriate symptom measures must reduce on average for the
treatment sample following an intervention
• Evidence should be based on:
o Related measurements (pre/post treatment)
o Time precedence (treatment must occur before outcome measure)
o tight controls for alternative explanations
What type of treatments are investigated?
• Both psychopharmacological (psychiatry) and psychotherapeutic (clinical psychology)
therapies are investigated in clinical controlled trials
• The nature of the therapy constrains the nature of the possible controls in the trial, to a
point
• What is controlled in such a trial and why?
What are these otrols aout?
• Clinical trial control measures aim to rule potential explanations for an effect other than the
tested treatment
• In other words, the controlled aspect of clinical trials aims to strengthen internal validity
• Threats to internal validity of a clinical trial include any factors that may completely or in
part explain the relationship between the IV and DV
• Example:
o Participant characteristics
find more resources at oneclass.com
find more resources at oneclass.com
o Historical confounds affecting one group
o Maturation affecting one group more than the other
o Regression to the mean
o Sampling bias
o Placebo effects, differential attrition, testing effects, etc
Example of a threat to validity: placebo
Placebo effect: a beneficial affect associated with a treatment; however, that effect cannot be
attributed to the treatment itself
• Thought to e die oes elief i the teatet – the expectation of it working
• Nocebo effect: things get worse
• Certain characteristics of placebos and of participants can influence the effect
• Evidence suggests the effet a ok ee he people ae i the ko
Placebo control
• With a medication trial, a sugar pill can be used as a placebo
• Placebo control more challenging with psychotherapy
• Psychotherapy is thought to contain specific (therapeutic technique) and non-specific
aspects (therapeutic relationship, time spent, the effect of talking with someone)
• In a psychotherapy trial, some degree of control of placebo effect may be achieved by
engaging the control group in either a specific or non-specific control measure
o Eg. just talking with someone, or treatment as usual (TAU) or other therapy (OT).
TAU and OT approaches provide the advantage of comparison with existing
evidence-based treatment
The Current State of Clinical Trial Evidence in Psychology: CBT
• Efficacy of CBT examined in numerous domains of psychological practice, not just depression
• Hierarchy of Evidence – considerable weight placed upon systematic reviews and meta-
analyses of clinical trials
What is Topical RCTs for Psychotherapy?
• An emerging area of clinical trial research is online provision of psychotherapy
• Commonly, face-to-face therapies with well established efficacy and effectiveness are tested
in online sphere
o Eg. CBT for depression and anxiety
• While these therapies have been established, their utility when delivered by phone, text,
skype, when self-directed, or blended must be examined
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
First controlled clinical trial: naval surgeon james lind treatment for scurvy, 1746. Lind sought evidence instead of practice based on opinion. What is a clinical trial systematic, controlled empirical approach to determining the effectiveness or efficacy of a therapy in context of mental disorder, this may pertain to psychotherapeutic (talking therapy like. Cbt) or psychopharmacological (medications, like antidepressant medication) interventions to a point, type of clinical trial that can be performed is determined by the type of intervention that is being tested. Internal validity: degree of confidence of the relationship between iv and dv (treatment and effect on symptoms). External validity: to what extent findings can be generalise. What type of evidence is needed: quantitative measurement of symptomatology, scores on appropriate symptom measures must reduce on average for the treatment sample following an intervention, evidence should be based on, related measurements (pre/post treatment) Time precedence (treatment must occur before outcome measure) tight controls for alternative explanations.