Clinical interventions for patients with alcohol disorders

1
Clinical interventions for patients with alcohol
disorders

• David Kavanagh
• School of Medicine
• University of Queensland
• July, 2005

2
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

3
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 elementsmay include
• Establishing roles/contexts that reward alcohol
  control
• Community reinforcementothers reward change new
  social activities, roles
• Cognitive therapyexamines evidence for
  problematic thoughts
• Unrealistic, alcohol expectancies low
  self-efficacy, low expectancy of success
• Social Skills Trainingdiscuss, 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

4
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 gt 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
• AAalso social support social rewards for
  efforts non-medical

5
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 attendanceaverage 2/3 sessions
• High follow-up gt 90 of living participants
• Significant, well sustained treatment effects
• Little difference between treatments

6
(No Transcript)
7
Are all treatments equally effective?

• Not all are equally supported by evidence

8
Metanalysis alcohol treatmentprocedures
evaluated in gt2 studies

• Rank Order (Quality x Outcome) studies
  N studies
• Brief Intervention 68 31
• Motivation enhancement 71 17
• (acamprosate) 100 5
• (naltrexone/nalmafene) 83 6
• Social Skills Training 68 25
• Community reinforcement 100 4
• Behavior contracting 80 5
• Behavioral Marital 62 8
• Case management 67 6
• Self-monitoring 50 6
• Cognitive therapy 40 10
• 12.5 Client-centered therapy 57 7
• 12.5 (disulfiram) 50 24
• 14.5 Aversion therapy, apneic 67 3
• 14.5 Covert sensitization 38 8
• Miller Wilbourne (2002) Addiction 97, 265-277

9
A similar story in clinical populations

• Rank Order (Quality x Outcome) studies
• Brief Intervention 73
• Social Skills Training 63
• (GABA agonist- acamprosate) 100
• (Opiate antagonistnaltrexone, nalmafene) 83
• Community reinforcement 100
• Behavior contracting 80
• 7.5 Behavioral marital therapy 63
• 7.5 Case management 67
• Cognitive therapy 40
• (disulfiram) 50
• 11. Motivation enhancement 56
• 12. Self-help 67
• 13. Client-centred 67
• 14. Aversion therapy, nausea 40
• 15.5 Aversion therapy, apneic 67
• 15.5 Covert sensitization 38

10
What is not supported?

• Confrontational approaches 0
• Hypnosis 0 , relaxation training 17
• Education 27 (general), 0 (clinical)

11
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

12
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

13
Sowhat do we take from this?

• 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

14
Review of brief interventions

• Moyer, Finney, Swearingen Vergun (2002)
  Addiction, 97, 279-292.
• Comparison with control.

15
(No Transcript)
16
Review of brief interventions

• Moyer, Finney, Swearingen Vergun (2002)
  Addiction, 97, 279-292.
• 2. Comparison with more extended treatment

17
(No Transcript)
18
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
  minheterogeneous
• -65gm/week for multi-sessionheterogeneous
• Cuijers et al. (2004)mortality rates
• Relative risk .47 (2.5-.89) for studies with
  verified rates

19
Butwhat is brief intervention

• Substantial variability in what is included in
  different studies
• There is substantial assessment, often extensive
  other treatment in many

20
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?

21
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

22
Metanalysis alcohol treatmentprocedures
evaluated in gt2 studies

• Rank Order (Quality x Outcome) studies
  N studies
• Brief Intervention 68 31
• Motivation enhancement 71 17
• (acamprosate) 100 5
• (naltrexone/nalmafene) 83 6
• Social Skills Training 68 25
• Community reinforcement 100 4
• Behavior contracting 80 5
• Behavioral Marital 62 8
• Case management 67 6
• Self-monitoring 50 6
• Cognitive therapy 40 10
• Client-centered therapy 57 7
• (disulfiram) 50 24
• Miller Wilbourne (2002) Addiction 97, 265-277

23
Specific treatmentsBerglund 2005

• Effect size
• Cognitive-behavioural 0.73
• Community reinforcement 0.59
• Naltrexone CBT 0.28 (gt supportive)
• Acamprosate 0.26
• (1 study CBT ? supportive)
• Disulfiram 0
• Disulfiram supervision/reinf. 0.53
• (But again, echoing Project MATCH)
• Comparisons of active treatments typically n.s.d
• (raises issuehow maximise nonspecific
  factorsexpectancy/hope, alliance)

24
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

25
Issues Relapse

• 50 or more often return to problem use
• Retrospective analysesmost 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

26
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

27
Project Match

• Very few significant results variable over time
  by chance?e.g.
• Outpatients
• If lt psychiatric severity, 12 step gt abstinent
  days than CBT to 12 months
• If high in anger, MET gt outcomes than CBT at Post
• If heavy drinking friends, better in 12-Step at 3
  years
• Aftercare
• If high alcohol dependence, 12 step gt 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 gt in MET than CBT at
  12 months, not later

28
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
• Issuegives some people failures before get
  most powerful treatment

29
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

30
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?

31
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

32
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

33
Issuesaccess

• 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

34
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 gt flexibility, immediacy, attraction,
  reminders
• But still some limits on access for some people
• Tendency to dip into itneed strategies to
  maximise appropriate use

35
Randomised Controlled Trials of Correspondence
InterventionSitharthan , 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

36
Sample

• 166 eligible
• At 4 months 70 men, 51 women
• Age M 46
• 73 with partners
• Education M 12.9 yrs
• 53 full time employment 27 part-time
• SADQ-C M 8.5
• Self-identified problem Median 6 yrs
• At 12 months 90

37
Alcohol over 12 Months (Standard Alcohol Units
per Week)
?
Minimal received Full
38
Drinking Days per week
39
Alcohol Dependence (SADQ-C)
40
never ? 10 drinks
41
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)

42
Alcohol Consumption over 12 Months (Standard
Alcohol Units per Week)Uses participants at 2
months, average substitution for missing data.
?
?
Brief S-M, Wait received Full
Extended S-M received Full
43
Alcohol ConsumptionParticipants Completing to
12 Months
BSM, Wait receive CBT
ESM receives CBT
44
Demonstrates impact of strategies in addition to
information, monitoring
45
Challenges Iatragenic effects

• Moos (2005)7-15 get worse
• Younger, deviant peer modelling, unmarried,
  residentially unstable
• Comorbidityother 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?

46
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

47
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

48
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 gt alcohol outcomes, anxiety
• Overtaxing patient recall/implentation
• Is additional aspect needed?
• Maybe more complex conditions need simpler
  treatments?

49
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

50
Participants

• Referred by GP or media respondent
• Men drinking gt 4 alcohol units/day
• Women drinking gt 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.

51
Cognitive-Behavioural Treatment

• Information about alcohol
• Self-assessment of costs/benefits of drinking
• Self-monitoring of alcohol intake
• Identification avoidance of high-risk
  situations
• Development of drink refusal skills
• Challenges to maladaptive thoughts
• alcohol expectancies
• own abilities

52
CBT Cue Exposure

• 20 min CBT plus written notes
• Participants
• drink 2 x 10mg doses of alcohol
• hold 1 more drink -look, sniff
• 5 x 5-min trials each session
• Alcohol remains until end of session
• homework exposure (2 x 40-min/week) with audiotape

53
CBT Emotional Cue Exposure

• Identical to CBT CE except
• Cue exposure while recall unpleasant experience
• mood re-induced each 10 min
• Positive mood induction at end of session

54
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

55
Results Retention
p lt .05
56
Results Drinks / week
57
Percent abstinent days
58
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

59
Origins of Opportunistic Interventions

• e.g. Kristenson et al. 83 Malmo study
• Included repeated physical screens
• 4 yrs 80 drop in sick days
• 5 yrs 60 drop in hospital days
• 6 yrs 50 drop in mortality

60
Opportunistic interventions in primary care

• 5-15 mins of assessment and advice ? ? 20-25
  fall in alcohol intake
• If subtract control, about 21 for men, 8
  women
• Screening brief advice cost A19-21.5
• Costs per additional life saved lt 1873
• Wutke, Shiell, Gomel, Conigrave (2001)

61
Opportunistic interventionsmeta-analysis in
primary care

• 5-20 min interventions
• Est 70 gm/wk, but
• substantial heterogeneity
• NS for women (in 2 studies testing them)
• Multi-session interventions
• Est 65 gm/wk
• Heterogeneity seen for men, and overall data set.
• Based on studies to 1997
• Poikolainen (1999)

62
Brief treatment

• ? effect than no treatment
• Brief ? extended for low dependence
• (Several later studies by Miller confirmed this)
• ? impact if ongoing feedback of
• behaviour or
• health outcomes
• Ineffective long treatment?
• Seems rather, people have skills already