When psychologists move from observation to inferences

1

• When psychologists move from observation to
  inferences generalizations, there are inherent
  risks of idiosyncratic interpretations,
  overgeneralizations, confirmatory biases
  similar errors in judgement.
• (APA Presidential Task Force on Evidence Based
  Practice, 2006)

2
Theoretical empirical foundations for evidence
based interventions

• Prof Alex Blaszczynski
• 4th Year Honours lecture

3
Components of clinical psychology expertise

• Systematic assessment treatment planning
• Having a cogent rationale for clinical strategies
• Clinical decision making, implementation
  monitoring
• Interpersonal expertise (explains 5-8 of outcome
  variance)
• Self reflection acquisition of skills use of
  resources
• Evaluations application of research evidence in
  both basic applied psychological science
• Understanding individual cultural differences
  influences

4
Lecture outline

• Briefly describe theoretical models of psychology
• Discuss concept of empirically validated practice
  in psychology
• Outline criteria for determining well-established
  treatments
• Objective Understand need for the
  conscientious, explicit, judicious use of
  current best evidence in making decisions about
  the care of individual patients (Sackett et al.,
  1996)

5
References

• Roth, A. Fonagy, P. (1998). What works for
  whom? A critical review of psychotherapy
  research. Guildford Press London.
• Nathan, P. Gorman, J. (1998). A guide to
  treatments that work. Oxford University Press
  Oxford
• APS (2009). Psychological Treatment
  Evidence-based practice practice-based
  evidence. URL www.psychology.org.au/publications/
  inpsych/treatments

6
NHMRC Guidelines

• Research merit and integrity
• 3.3.1 Health care medical institutions should
  establish standards to determine when an
  innovative intervention requires systematic
  investigation to determine its safety efficacy
• 3.3.2 When such systematic investigation is
  required, it should be treated as clinical
  research needing formal consideration by an HREC

7
NHMRC Guidelines

• 3.3.3 Researchers should show that
• Research is directed to answering a specific
  question/s
• A scientifically valid hypothesis is being tested
  that offers a realistic possibility that the
  interventions studied will be at least as
  beneficial overall as standard treatment, taking
  into account effectiveness, burdens, costs
  risks
• Size profile of the sample to be recruited is
  adequate to answer the research question
• The research meets the relevant requirements of
  the CPMP/ICH Note for Guidance on Good Clinical
  Practice (CPMP/ICH-135/95), ISO 14155 Clinical
  Investigation of Medical Devices, and the TGA.

8
Principles of ethics in treatment interventions

• Beneficence Psychologists should promote the
  patients good prevent harm
• Respect for Autonomy Competent patients should
  be allowed to make their own decisions about
  treatment options recommended by psychologists
• Distributive Justice Treatment resources should
  be distributed fairly cost-effectively

9
What is a psychotherapeutic intervention?

• An interpersonal process designed to bring about
  modifications of feelings, cognitions, attitudes
  and behaviour which have proved troublesome to
  the person seeking help from a trained
  professional
• (Strupp, 1978)
• Over 400 different therapies described by Kazdin
  (1986)
• This figure has increased over the years

10

• Major classes of theoretically conceptually
  derived interventions based on a model of human
  behaviour
• Psychodynamic
• Behavioural Cognitive-Behavioural
• Interpersonal psychotherapy
• Strategic or Systematic psychotherapies
• Supportive Experiential psychotherapies
• Group therapies
• Counselling
• There is an ethical obligation to ensure that the
  intervention provided is effective efficacious

11

• Eysenck, H. (1952). The effects of psychotherapy
  Extended updated 1960 1969
• Seminal review of the effectiveness of
  psychotherapy
• Use of spontaneous remission rates in neurosis as
  benchmark against which to compare effects of
  psychotherapy
• Approx. 2/3 recover from illness within two years
  
• Improved with treatments
• Psychoanalysis 44
• Electrically 64
• GPs 72
• Treated untreated have comparable outcomes

12
Theoretical foundations Psychodynamic

• Intensive (one to two sessions per week), long
  term treatment directed to resolving intrapsychic
  unconscious conflicts
• Aim to restructure personality rather than
  directed to single symptom recovery
• Use of the talking cure free association
  interpretation of dreams
• Focus on discovering modifying defence
  mechanisms
• Hypothesized processed not observable or
  hypotheses unfalsifiable

13
Theoretical foundations Behavioural

• Derived from learning theory (ignored cognitions)
• Behaviours determined by schedules of
  reinforcement
• Pavlovian Classical Conditioning
• Skinnerian Operant Conditioning
• Based on data derived from animal studies
• Observable behaviours easily measured
• Hypotheses open to experimental testing hence
  falsifiable
• Wolpes Systematic Desensitization
• Conditioning experiments

14
Theoretical foundations Cognitive Behavioural

• Commonality with psychodynamic models is the
  focus on irrational cognitions but in contrast,
  these are assumed to be learnt (rather than due
  to unconscious forces) maintained by
  reinforcement schedules
• Treatment is directed to identifying, challenging
  modifying irrational or distorted cognitions
  monitoring outcomes
• Use of behavioural experiments to test cognitions
• Focus is on how maladaptive functioning is
  maintained by environment belief structures
• ? Shifts in cognitions follow behavioural
  exposure or vice versa
• Approximately 30 different models of cognitive
  therapy (for example, Beck, Ellis, Meichanbaum,
  Bandura)

15
Theoretical foundations

• Systemic therapy
• Neither symptoms or insights are focus for
  treatment intervention
• It is the system that generates problem behaviour
  is target for intervention
• Therapists task is to identify the role symptoms
  play maintaining dysfunctional system vice
  versa
• Counselling
• Not a unitary theoretical framework but defined
  by setting in which it takes place
• Client centered (empathy, unconditional regard,
  warmth)
• Focus on current problems using pragmatic problem
  or solution focused interventions

16
Primary questions

• Is the treatment efficacious effective?
• Is the treatment more efficacious effective
  than current gold standard interventions?
• Are there negative /or harmful side effects?
• Is the treatment more cost-effective than
  existing interventions?
• HOW DO WE DETERMINE THE ABOVE?
• Empirical data obtained through comparative /or
  evaluative research outcome studies

17
Scientist practitioner model

• American Psychological Association policy
  established in 1947 that psychologists should be
  trained in both science practice
• Application integration of experimental
  findings in practice
• This formed the strong foundation for evidence
  based interventions in psychology

18
Effects of Exposure to Trauma

• Severe incapacitating psychological distress
  (intrusive memories, hyperarousal avoidance)
• Normal, immediate response lasting minimum two
  days subsides within month - Acute stress
  disorder
• Symptoms persisting longer than one month PTSD
  or chronic psychiatric morbidity (sub-threshold
  PTSD)
• lt 1 month ASD
• 1-3 months Acute PTSD
• 3 months Chronic PTSD
• 30 may develop PTSD

19

• CRITICAL INCIDENT DEBRIEFING
• Single session intervention involving emotional
  processing/ventilation by encouraging
  recollection/reworking of traumatic event,
  accompanied by normalisation of emotional
  reaction
• Critical question Is one session of 'debriefing'
  following trauma effective in preventing PTSD?
• Reviewed 15 (11 usable) Randomized Control Trials
  on people exposed to traumatic event within last
  month
• (Rose, Bisson, Churchill, Wessely, 2001)

20

• Findings
• Single session debriefing did not prevent onset
  of PTSD or reduce psychological distress compared
  to control
• One trial reported significant increase in risk
  of PTSD at 1 year in debriefed (Odds Ratio 2.51)
• Debriefed group reported no reduction in PTSD
  severity at 1-4 months or 3 years
• No evidence that debriefing reduced general
  psychological morbidity, depression or anxiety,
  or superiority over education
• Short term PTSD Debriefing 16 vs. Controls 11
• Long term PTSD 21 vs.
  7

21
Psychological interventions for adults with PTSD

• Recommendations
• Adults with PTSD should be provided with
  trauma-focussed interventions (trauma-focussed
  CBT or eye movement desensitization EMDR and
  reprocessing, in addition to in vivo exposure)
• As available evidence does not support importance
  of eye movements in EMDR, it is recommended that
  practitioners who use EMDR be aware that
  treatment gains are more likely to be due to
  engagement with traumatic memory, cognitive
  processing and rehearsal of coping and mastery
  responses

(Australian Centre for Posttraumatic Mental
Health, 2007)
22
Psychological interventions for adults with PTSD

• Where symptoms have not responded to first line
  trauma-focussed interventions, consider
  alternative form of trauma-focussed
  interventions
• Non trauma-focussed interventions such as
  supportive counselling and relaxation should not
  be provided to adults with PTSD in preference to
  trauma-focussed interventions

(Australian Centre for Posttraumatic Mental
Health, 2007)
23

• Conclusion
• No evidence that single session individual
  psychological debriefing is useful treatment for
  prevention of PTSD.
• Structured psychological interventions such as
  psychological debriefing should not be offered on
  a routine basis
• Compulsory debriefing of victims of trauma should
  cease.
• Rose, Bisson, Churchill, Wessely (2001)

(Australian Centre for Posttraumatic Mental
Health, 2007)
24

• Many therapies still commonly used have not been
  found to be helpful in treatment of social
  anxiety disorder
• Relaxation therapy teaches individuals techniques
  to decrease anxiety. Studied carefully found to
  be not helpful. In some situations it can even
  make social anxiety worse
• Beta-blockers found to be helpful for performance
  social anxiety disorder, but not been helpful for
  generalized type of social anxiety disorder

25

• Autism described 60 years ago
• Now 111 treatments advanced
• Parents apply an average of 4 to 7 interventions
  simultaneously
• Sensory integration treatment widely used for
  autistic patients
• Technique developed by occupational
  therapist/clinical psychologist Dr. A. Jean Ayres
  based on observation that some children show
  excessive sensitivity to external stimulitouch,
  position in space, sound
• Hypothesised this was result of poor ability to
  process sensory messages received by the brain
• SIT involves graded tactile stimulation,
  balance exercises, exposure to soft music, use
  of weighted clothes
• Recent studies show it is no more beneficial than
  any other treatment

26
How do new treatment interventions evolve how
are they evaluated?
27
How do innovative treatments develop?
Scientist-practitioner modifies or introduces new
therapeutic component
28
Evidence-based best practice

• Integration of research findings with clinical
  expertise to promote effective efficacious
  treatment of clients
• Translation of empirically supported principles
  techniques into psychological practice
  assessment, case formulation, therapeutic
  relationships intervention

29
Chelmsford Hospital, 2 The Crescent Pennant
Hills, run by Dr Harry Bailey, a fashionable
highly thought of psychiatrist specializing in
deep-sleep therapy. Depressed patients put into
drug induced sleep for up to three weeks. ECT
administered while unconscious. Derived from
insulin coma therapy Outcome 1963-1980 1,127
patients treated 152 patients awarded 5.5
million damages 27 deaths 24 suicides Barry
Hart suffered double pneumonia, deep vein
thrombosis, pulmonary embolism anoxic brain
damage. Hart sued Dr Herron Chelmsford
awarded 60 000 for false imprisonment assault
and battery.
30
Efficacy

• Efficacy results achieved in a research setting
• Internal validity the extent to which a causal
  relationship can be determined between variables
  (intervention) outcome
• Require homogenous groups randomized into
  treatment monitoring outcomes
• Key point Need to ensure that the researcher is
  comparing apples with apples
• Treatment duration, length of therapist contact,
  plausibility of control intervention (e.g., brief
  CBT versus psychodynamic for depression)
• Regular access to supervision leading to enhanced
  quality of intervention
• Patient characteristics Severity duration of
  disorder, co-morbid disorders, biased referral
  sources (e.g., comparing same treatment in two
  difference clinics)

31
Effectiveness

• Effectiveness (clinical utility in routine
  practice)
• External validity Extent that causal
  relationships can be generalized to wider
  population
• Referrals not influenced by strict
  inclusion/exclusion criteria leading to greater
  heterogeneity in patient characteristics
• Treatment applied in less systematic fashion
• Variability in attendance
• Conclusion Research therapy appears more
  effective than everyday clinical practice

32
What factors contribute to incorrect conclusions
regarding treatment outcomes?

• Inappropriate research designs
• Incorrect diagnosis use of standardized DSM
  criteria
• Placebo effect placebo effect probably major
  factor
• Self limiting /or episodic nature of
  conditions
• Concurrent/complementary treatments
• Spontaneous remission
• Researcher allegiance
• Lack of manuals - integrity

33
Criteria for empirically validated studies
34
Criteria for empirically validated treatments

• At least two good group design experiments
  demonstrating efficacy in one or more of the
  following ways
• Superior (statistically) to medication or
  psychological placebo intervention
• Equivalent to an already established treatment in
  experiments with adequate sample size
• OR

35
Criteria for empirically validated treatments

• A large series of since case design experiments
  demonstrating efficacy. These must have
• Used good experimental designs
• Compared the intervention to another treatment as
  in IA above
• Further criteria for both I II
• Experiments must be conducted with treatment
  manuals
• Characteristics of the client samples must be
  clearly specified
• Effects must have been demonstrated by at least
  two different investigators/teams (Avoid research
  allegiance effect)

(American Psychological Association Task Force on
Psychological Interventions)
36
Probably efficacious treatments

• Two experiments showing the treatment is superior
  to a waiting list control group
• OR
• One or more experiments meeting the
  Well-Established Treatment Criteria IA or IB and
  IV but not V
• OR
• A small series of single case design experiments
  (ngt3) otherwise meeting Well-Established
  Treatment

37
Methodologies strategies in research design

• There is no ideal research design
• Design depends upon the aims opportunities
  presented to researchers
• Need to match aims to methods in context of
  resources funding
• Hierarchy of evidence established
• Randomized controlled outcome studies
• Controlled non-randomized trials experimental
  case series
• Cohort studies (groups of patients allocated to
  treatment)
• Case controlled studies (patients with similar
  outcomes grouped retrospective evaluation
  conducted

38
Randomized controlled outcome studies Type 1

• Involves random allocation of patients into one
  or more groups (experimental vs.
  control/placebo/standard treatment - 5 of times
  groups differ by chance)
• Prospective evaluation of outcome (pre-/baseline
  post-treatment follow-up)
• Researcher patient blind to group allocation
  (intervention received)
• Clear exclusion/inclusion criteria
• Sufficient power (sample size) to detect
  differences

(American Psychiatric Association Practice
Guidelines, 1995)
39
Randomized controlled outcome studies

• Consideration of clinical versus statistical
  significant outcomes
• All intervention groups treated identically
  except for experimental treatment
• Limitations
• Exposing patients to inert intervention when
  effective treatment available often thought
  unethical
• Patient bias includes those agreeing to take
  their chances in receiving active treatment

40
Clinical trial Type 2

• Prospective study where intervention is applied,
  compared to another intervention monitored
  longitudinally but some aspect of RCT missing
• Double blinding absent
• No random allocation to treatment groups
• Follow-up period relatively short to make full
  judgment of efficacy

41
Cohort or longitudinal Type 3

• Open treatment trials aimed at obtaining pilot
  data
• Participants followed up prospectively without
  any specific intervention
• Subject to experimenter bias

42
Case-control studyType 3

• Groups of patients identified in present
  retrospective information on response to
  treatment is elicited
• Subject to experimenter bias
• Uncontrolled data collection
• Retrospective recall bias

43
Review with data analysis study Type 4

• Review with secondary data analysis
  meta-analysis
• Negative or null finding studies tend not to be
  published leading to bias to over-inflate
  effectiveness
• Selective inclusion of data

44
Review without data analysis study Type 5

• Qualitative review with no secondary data
  analysis
• Negative or null finding studies tend not to be
  published leading to bias to over-inflate
  effectiveness
• Selective inclusion of literature biased toward
  researcher allegiance

45
Other Type 6

• Expert consensus, case reports, textbooks,
  opinion papers, Internet non-peer-reviewed
  articles
• Marginal value given potential biases,
  ideological influences, non-supported subjective
  judgments/opinions

46
Summary

• Clinical observations form source of innovative
  hypotheses
• Qualitative research used to generate new
  hypotheses
• Single-case studies useful in establishing causal
  relationships A B A design
• Process studies important to establish mechanism
  of change
• Naturalistic observational studies cannot be
  undertaken in experimental design
• RCTs
• Meta-analyses

47
Summary

• Clinical expertise
• What are the factors processes that allow
  experienced clinicians to develop effective
  formulations, diagnostic decision making,
  flexibility in treatment applications
• Patient characteristics
• Understanding components of motivation, comorbid
  psychopathology, personality traits, preferences
  for treatment, socio-economic socio-cultural
  factors that impinge upon outcomes

48

• Empirically supported therapies
• Start with treatment asks if it works with
  certain disorders under certain conditions
• Evidence-based practice
• Start with patient ask what research evidence
  will assist in achieving the best outcome