Transcript Cognitive-Behaviour Therapy

Clinical interventions for patients
with alcohol disorders
David Kavanagh
School of Medicine
University of Queensland
July, 2005
Psychological treatments cover a wide
range
 Brief Interventions—
 often assessment, feedback on risk, advice,
 sometimes options, planning, contracting
 Motivation enhancement/motivational
interviewing—
 patient-centred, empathic, accepting
 encourages evaluation of current use
 Elicits internal conflict re current behavior
Psychological treatments cover a wide
range
 Cognitive-behavioural (CBT) or self-control approaches—
 Based on theoretical work of Bandura, Marlatt/Gordon, Rehm
 often include brief intervention elements, but also
 Alcohol self-monitoring
 Goal setting/planning/contracting
 Identification of risk situations, applying problem solving
 Other elements—may include
 Establishing roles/contexts that reward alcohol control
 Community reinforcement—others reward change; new social activities, roles
 Cognitive therapy—examines evidence for problematic thoughts
 Unrealistic, + alcohol expectancies; low self-efficacy, low expectancy of success
 Social Skills Training—discuss, demonstrate, practice, feedback, homework
 e.g. alcohol refusal; developing new relationships
 Behavior contracting with concrete rewards
 (although effects go when rewards stop)
 Behavioral Marital therapy
 focuses on problem solving, contracting, rewarding
Psychological treatments cover a wide
range
 Case management—
 a very wide range of procedures
 often focuses on multiple domains of need
 Client-centered—
 focuses on > self-acceptance
 Aversion therapies
 Some form of negative experience if drink
 E.g. covert sensitization
 12-step
 Principles established in Alcoholics Anonymous (AA)
 Including admission of problem; common abstinence goal; proximal goal
focus; reliance on higher power
 AA—also social support; social rewards for efforts; non-medical
A key research trial:
Project Match (1997)
 Two parallel trials—
 952 outpatients
 774 aftercare following inpatient treatment
 Each randomly to 12 weeks
 Cognitive-Behaviour Therapy (CBT)
 Motivation Enhancement training (MET)
 (4 sessions)
 12-Step (individual, not Alcoholics Anonymous)
 High attendance—average 2/3 sessions
 High follow-up: > 90% of living participants
 Significant, well sustained treatment effects
 Little difference between treatments
Are all treatments equally
effective?
Not all are equally supported by evidence
Metanalysis: alcohol treatment—
procedures evaluated in >2 studies
Rank Order (Quality x Outcome) % studies +
1. Brief Intervention
68
2. Motivation enhancement
71
3. (acamprosate)
100
4. (naltrexone/nalmafene)
83
5. Social Skills Training
68
6. Community reinforcement
100
7. Behavior contracting
80
8. Behavioral Marital
62
9. Case management
67
10. Self-monitoring
50
11. Cognitive therapy
12.5 Client-centered therapy
12.5 (disulfiram)
14.5 Aversion therapy, apneic
14.5 Covert sensitization
40
57
50
67
38
N studies
31
17
5
6
25
4
5
8
6
6
10
7
24
3
8
Miller & Wilbourne (2002) Addiction 97, 265-277
A similar story in clinical populations
Rank Order (Quality x Outcome)
% studies +
1.
2.
3.
4.
5.
6.
7.5
7.5
9.
10.
73%
63%
100%
83%
100%
80%
63%
67%
40%
50%
Brief Intervention
Social Skills Training
(GABA agonist- acamprosate)
(Opiate antagonist—naltrexone, nalmafene)
Community reinforcement
Behavior contracting
Behavioral marital therapy
Case management
Cognitive therapy
(disulfiram)
11. Motivation enhancement
12. Self-help
13. Client-centred
14. Aversion therapy, nausea
15.5 Aversion therapy, apneic
15.5 Covert sensitization
56%
67%
67%
40%
67%
38%
Miller & Wilbourne (2002) Addiction 97, 265-277
What is not supported?
 Confrontational approaches 0% +
 Hypnosis 0% +, relaxation training 17% +
 Education 27% + (general), 0% + (clinical)
Caveats
Depends on decisions made in the analysis
e.g. what category put treatment in
Effects are relative to the
control group used
 (no treatment/usual treatment etc)
 e.g. self control: includes comparison with same treatment by
another delivery method
nature of participants and setting
Most evidence is from North America
An example of one that does not do
very well
 12-step (33% +)/ AA (14% +), but
 hard to get a no-treatment control
 many studies with more powerful
treatments use this as a control
So…what do we take from this?
1. Brief intervention is effective
 Relies on
 existing skills
 motivational power of assessment summary,
professional advice
 Current support for motivational
interviewing stronger in opportunistic
than treatment seeking context
Review of brief interventions
 Moyer, Finney, Swearingen & Vergun
(2002) Addiction, 97, 279-292.
1. Comparison with control.
Review of brief interventions
 Moyer, Finney, Swearingen & Vergun (2002)
Addiction, 97, 279-292.
2. Comparison with more extended treatment
Other reviews of brief
intervention
 Berglund (2005)—Studies in primary care:
 Single session ES = 0.19 ‘small’-homogeneous
 Multiple sessions ES = 0.61 ‘moderate’- but heterogeneous
 Ballesteros et al. (2004)—primary care:
 Similar effect sizes for men (0.25), women (0.26)
 Whitlock et al. (2004)—multi-contact, primary care
 13-34% greater fall in alcohol intake/week than controls
 10-19% greater proportion at moderate or safe intake level
 Poikolainen (1999)—excluding high alcohol dependence, hospitalised:
 -70gm/week relative to controls, for 5-20 min—heterogeneous
 -65gm/week for multi-session—heterogeneous
 Cuijers et al. (2004)—mortality rates
 Relative risk = .47 (2.5-.89) for studies with verified rates
But…what is brief intervention
Substantial variability in what is
included in different studies
There is substantial assessment, often
extensive other treatment in many
What are active ingredients of
Brief Intervention?
Feedback of assessment/risk?
Advice?
Empathy/encouragement/approval?
In multiple contacts—
Repeated assessment?
Informal problem solving/cognitive therapy?
Heterogeneity reflective of less effective treatment
in some studies?
Other conclusions from
Miller/Wilbourne analysis
 Forms of psychological intervention with
social components show strong support
 Social skills training
 Community reinforcement
 Behavioural marital therapy
 Behavioural contracting, cognitive
therapy show reasonably high support
Metanalysis: alcohol treatment—
procedures evaluated in >2 studies
Rank Order (Quality x Outcome) % studies +
1. Brief Intervention
68
2. Motivation enhancement
71
3. (acamprosate)
100
4. (naltrexone/nalmafene)
83
5. Social Skills Training
68
6. Community reinforcement
100
7. Behavior contracting
80
8. Behavioral Marital
62
9. Case management
67
10. Self-monitoring
50
11. Cognitive therapy
40
12. Client-centered therapy
57
13. (disulfiram)
50
Miller & Wilbourne (2002) Addiction 97, 265-277
N studies
31
17
5
6
25
4
5
8
6
6
10
7
24
Specific treatments—
Berglund 2005
 Cognitive-behavioural
 Community reinforcement
Effect size
0.73
0.59
 Naltrexone + CBT
0.28 (> + supportive)
 Acamprosate
0.26
 Disulfiram
 Disulfiram + supervision/reinf.
0
0.53
(1 study— + CBT  + supportive)
(But again, echoing Project MATCH):
 Comparisons of active treatments typically n.s.d
 (raises issue—how maximise ‘nonspecific’ factors—expectancy/hope,
alliance…)
Issues: Engaging, maintaining
change
Evidence for opportunistic intervention shows
can generate motivation
But not in all people
Substantial room for improvement of these approaches
And dropouts high unless significant
effort/incentives to retain
Issues: Relapse
50% or more often return to problem use
Retrospective analyses—most commonly negative
emotion/interpersonal conflict
Prediction, problem solving re risk situations may
help
But many problem situations hard to predict
Decision to engage/avoid situation difficult
Who benefits from what?
Project Match
Assessed 10 matching attributes
Severity alcohol involvement
Conceptual level
Gender
Meaning seeking
Readiness to change
Psychiatric severity
Social support for drinking/abstinence
Sociopathy
Typology of problem
Project Match
Very few significant results: variable over time—
by chance?—e.g.:
Outpatients:
If < psychiatric severity, 12 step > abstinent days than CBT to
12 months
If high in anger, MET > outcomes than CBT at Post
If heavy drinking friends, better in 12-Step at 3 years
Aftercare:
If high alcohol dependence, 12 step > outcomes than CBT at
Post
 if low, better for CBT
If higher in anger, better in MET than CBT to 12 months
Lower readiness to change: > in MET than CBT at 12 months,
not later
So…
Little to help design better versions of existing
treatments from this
Hard to know who will benefit from shorter
treatments
Left with an approach of trying less intensive first
Issue—gives some people ‘failures’ before get most
powerful treatment
Also…
We know that alcohol dependence often
has a fluctuating course
Berglund (2005):
Unclear how treatment affects this course
Unclear whether treatments differentially
effective at different points of the trajectory
Issues: Craving
We now have some effective medications,
but
Craving can occur long after treatment stops
Environmental cues, or memory associations
Especially after a priming drink?
Really suggesting permanent medication?
What can help craving?
Habituation to alcohol cues
(is in any effective treatment, can also do in clinic)
However involves discomfort, disruption of
concurrent tasks
Initial avoidance of high cue situations
(only a temporary and partial solution)
We are working on a new approach that
may improve effectiveness
Issues: Cultural/ethnic
applicability
A problem or opportunity?
Consider cultural beliefs/practices that may be
beneficial
E.g. considerable current interest in meditation
mindfulness, derived from Buddhist tradition
Issues—access
Risky intake of alcohol involves
large numbers
spread across wide areas
The majority need an approach that is
inexpensive
accessible
acceptable if have low-level problems
Most cannot access specialist treatment
Bibliotherapy/letters/internet
 A manual can help if want to  drinking, but
 many do not read a long manual
 Berglund: Bibliotherapy ES = 0.19
 Letters can




be brief
look at one treatment strategy at a time
offer feedback on progress
retain low cost
 Electronic (internet, palm pilot, telephone) offers > flexibility,
immediacy, attraction, reminders
 But still some limits on access for some people
 Tendency to dip into it—need strategies to maximise appropriate use
Randomised Controlled Trials of
Correspondence Intervention
Sitharthan , Kavanagh & Sayer (1996)
Minimal intervention
Information on alcohol
self-monitoring, or
Full treatment--added other elements
Setting drinking goals
Ways to deal with urges/ temptations
Identifying high-risk situations + problem solving
Applying incentives for control
Changing lifestyle
Alcohol over 12 Months
(Standard Alcohol Units per Week)
60
50
40
Minimal-Men
Minimal-Women
Full-Men
Full-Women
30
20
10
0
0
4 Months
6 Months

Minimal
received Full
12 Months
Drinking Days
per week
7
6
5
4
3
2
1
0
0
4 Months
6 Months
12 Months
Kavanagh et al. (1998)
Full treatment immediately or
Wait List to 2 months  Full
Brief self-monitoring to 2 months  Full
Self-monitoring as long as possible
(Extended self-monitoring)
Alcohol Consumption over 12 Months
(Standard Alcohol Units per Week)
Uses participants at 2 months, average substitution for missing data.
45
40
35
30
25
20
15
Brief Self-Mon.
Extended S-M
Immediate Full
Wait List
10
5
0
0
2
4

Brief S-M, Wait
received Full
6
12

Extended S-M
received Full
Alcohol Consumption:
Participants Completing to 12 Months
45
40
35
30
25
20
15
Brief S-M/CBT
Extended S-M
Immediate CBT
Wait/CBT
10
5
0
0
2
4
BSM, Wait
receive CBT
6
12
ESM
receives CBT
Demonstrates impact of strategies
in addition to information,
monitoring
Challenges: Iatragenic effects
 Moos (2005)—7-15% get worse
 Younger, deviant peer modelling, unmarried, residentially unstable
 Comorbidity—other drugs, mental health problem
 Conflict, social isolation, sexual abuse
 (& evocation of negative mood in treatment)
 Poor therapeutic alliance
 Attend fewer sessions; treatment goals not desired; low expectancy
of treatment
 Lack of alcohol monitoring
 Stigma of diagnosis, treatment?
Challenges: Takeup, fidelity,
effectiveness
Some demonstrations in brief intervention
literature
Key aspects appear to be
Appropriate knowledge, skills, self-efficacy
Low opportunity cost, high incentives
Cues to use
Dealing with comorbidity
No well established treatment as yet
Integrated treatment probably better that
parallel, sequential in psychosis
Our work suggests that a 3-hour motivational
treatment may help
But may not be better than rapport + assessment
Less clear in anxiety, depression
We are currently evaluating this in depression
Caution:
More is note always better
Randall et al. (2001)-anxiety comorbidity
Alcohol alone vs Alcohol + Anxiety
Parallel treatments, not tailored for comorbidity
Alcohol alone: > alcohol outcomes, = anxiety
Overtaxing patient recall/implentation
Is additional aspect needed?
Maybe more complex conditions need
simpler treatments?
Kavanagh, Young, T. Sitharthan,
G. Sitharthan, Saunders
Standard CBT for alcohol abuse
CBT + Cue Exposure
for moderation drinking (CE) or
CBT + CE within a negative mood
All had 8 x 70 min sessions over 10 weeks
Participants
Referred by GP or media respondent
Men drinking > 4 alcohol units/day
Women drinking > 2 alcohol units/day
Problems controlling drinking when dysphoric, &
 1 such occasion over 2 wks at Baseline
Excluded
psychosis; PTSD; current major depression;
insufficient English;
current abuse on other substances;
medical condition prohibiting drinking;
Lived 100km away.
Results
444 responded to publicity or referred
373 met criteria for inclusion
184 (49% of those eligible) consented to participate
After 21 pilots, 163 allocated
71 male, 92 female
55% married/de facto
Average 13.4 yrs education
11% unemployed
Drinking an average 36.6 alcohol units (366gm)/week
Results: Retention
80%
70%
60%
50%
CBT
CE
EmCE
40%
30%
20%
10%
0%
Completed 75% sessions
p < .05
Results: Drinks / week
50
50
Men
Women
CBT
CE
Number of drinks per week
EMCE
40
40
30
30
20
20
10
10
0
0
Baseline
Recommended Max
Post
3mths
6mths
9mths
Assessment Occasion
12mths
Baseline
Post
3mths
6mths
9mths
Assessment Occasion
12mths
Percent abstinent days
Percent Abstinent Days
50
50
Men
40
40
30
30
20
20
10
10
CBT
CE
EMCE
Recommended Min
0
0
Baseline
Women
Post
3mths
6mths
9mths
Assessment Occasion
12mths
Baseline
Post
3mths
6mths
9mths
Assessment Occasion
12mths
So
CBT had better retention than the cue exposure
CBT did a little better than cue exposure, but this
was due to worse initial drinking
 Was cue exposure insufficiently naturalistic?
 Joins other studies showing little effect from cue exposure
Good maintenance in all conditions to 12 months
Men and women reduced to similar levels,
but men closer to target