Unguided e-therapy for adults
with depressive symptoms:
IPT vs. CBT
Tara Donkera, Kathleen M. Griffithsb, Helen
Christensenb, Kylie Bennettb , Anthony
Bennettb, Annemieke van Stratena , Pim
Cuijpersa
a VU
University, Amsterdam
b Australian National University, Canberra
Introduction
Life time prevalence: 19% (Bijl et al., 1998)
Introduction
Top four leading causes of burden of disease worldwide (Lopez & Murray, 1998)
Introduction
Costs per capita for mood disorders:
€ 5,009 (Smit et al., 2006)
Low-intensity treatment for depression and anxiety
Marijke
»36 years
» Single
» Two children
»Town
»Part-time working
»Mild depression
Obstacles
Long waitlists
Obstacles
2 hrs of travelling to the clinic
Obstacles
Take leave from work
Obstacles
Arrange a babysit
Obstacles
Costs of babysit
Obstacles
+.. Costs of travelling,
babysit..
Low-intensity treatment for depression and
anxiety
Psychological self-help:
•|Standardized
•Independent
•Homework/tasks
•Via book, audio, internet..
•Guidance by email, chat, telephone
Advantages
No wait lists
Advantages
Working from home
Advantages
24/7
Advantages
Cost-effective
Advantages
Anonymous
Low-intensity treatment for depression and anxiety
Intensive face-to face
treatment
Low-intensity treatment
Obstacles
Advantages
•
•
•
•
•
•
•
•
Wait lists
Costs of therapy etc.
Time of travelling
Leave/babysit
Directly accessable
Cost-effective
Working from home
24/7
Introduction
Introduction
Introduction
Introduction
Introduction
E-CBT = effective (e.g. Carlbring et al., 2009; Christensen et
al., 2004;Kaldo et al., 2008; Ruwaard et al., 2009; Spek et al.,
2008; Warmerdam et al., 2008)
Introduction
Introduction
Introduction
Depression: IPT = CBT (Cuijpers et al., 2011)
Effect size: 0.04 (95% CI 0.14-0.21)
Introduction
=
?
Design
Internet-assisted
Cognitive Behavior Therapy (CBT)
&
Interpersonal Psychotherapy (IPT):
Design
Internet-assisted
Cognitive Behavior Therapy (CBT)
&
Interpersonal Psychotherapy (IPT):
Compared to
Internet-assisted
CBT MoodGYM (control group)
Design
• Non-inferiority trial
• H0: (IPT ≠ CBT) ≠ MoodGYM
• H1: (IPT= CBT) = MoodGYM
Design
MoodGYM
• Within effect size: 0.56 (95% CI: 0.33-0.79)
(MacKinnon et al., 2008)
• Determined sample size: N=450 on post-test
Design
MoodGYM
• Between effect size: 0.33 (95% CI: 0.11-0.55)
• Within effect size: 0.56 (95% CI: 0.33-0.79)
(MacKinnon et al., 2008)
• Determined sample size: N=450 on post-test
Design
• Non-inferiority trial
• 3 e-conditions (IPT, New CBT, CBT
MoodGYM)
• Fully automated trial, 4 weeks
• Online recruitment and screening
Design
• Inclusion: 18 years or older
• Exclusion: currently under treatment by
mental health specialist
• 3 measurements (baseline, after 4 weeks
and after 6-months)
• Primary outcome: depressive symptoms
Interpersonal Psychotherapy
• Developed by Klerman and Weissman
(1984)
• Relationship between depression and
experiences
• IPT aims at improvement of interpersonal
functioning and (inherent) depression
Internet-assisted IPT - overview
Internet-assisted IPT -week 1
Role disputes
• Identification of the dispute (with whom?)
• Modification of communication patterns
Internet-assisted IPT -week 2
Problems making relationships
• Realistic evaluation of can do`s/ can`t do`s
• Increase social contacts
Internet-assisted IPT -week 2
Problems making relationships
Internet-assisted IPT – week 2
Internet-assisted IPT - week 3
Role Changes
• Investigate which old role is given up
• Validate the loss
• Support letting go of old role
Internet-assisted IPT - week 3
Role Changes
• Develop new skills for the new role
• Develop new relationships and social
support
Internet-assisted IPT – week 3
Internet-assisted IPT - week 4
Grief
• Activate grief proces
– Psychoeducation
– Learn to express emotions
• Finding new meaningful activities and
social contacts replacing those which were
lost
Internet-assisted IPT – week 4
Results: Participant flow
N=1993
met inclusion criteria
n=69
did not fill in baseline Q
N=1924
included
IPT
n=641
MoodGYM
n=642
CBT
n=641
Results: demographics
•
•
•
•
•
•
•
•
Gender: 71% female
Age category: 25-29 years (15.1%)
Country: 1: Australia (38%)
2: United Kingdom (25%)
3: United States (19%)
Education: higher degree (26%)
Previous treatment by professional: 80%
Been depressed before: 90%
Results: depressive symptoms
Mean CES-D at baseline (n.s.):
• Moodgym: 35 (sd: 11.7) n=622
• IPT: 36 (sd: 11.9) n=623
• CBT: 36 (sd: 11.) n=615
Results: depressive symptoms
completers only
CES-D depression score
40
C BT
IPT
MG
35
30
25
20
15
0
4
weeks since randomization
Results: depressive symptoms
completers only
Pre-post test:
• No significant differences across the
condition (F=2.018; P=.13)
• Effect sizes:
MoodGYM: 0.81
CBT: 0.87
IPT: 0.77
Results: drop out
•
•
•
•
•
•
CES-D: p=.02* (higher CES-D)
Sexe: p =.<0.1* (more men)
Age: p =.<0.1* (25-29 years)
Education: p =.04* (lower education)
Condition: p = .02* (MoodGYM)
Been depressed: p = .80
Conclusions
Pre-post test for completers:
• H1: (CBT = IPT) = MoodGYM
• Effect sizes between 0.77-0.81
• E-IPT: more treatment choice
• 70% “drop out”
• Significant differences in drop out for
demographics, treatment allocation,
baseline CES-D score
Discussion
• ‘Gold standard’ MoodGYM:
Equivalence margin: 0.33
• Follow up data not yet available
• Drop out rate: 70% →
»
»
»
»
Missing data approaches
high symptoms: seek your GP
Unguided self-help
Characteristic of population
Disadvantages
Anonymous
Disadvantages
Guided self-help
Disadvantages
High drop out rate
Discussion
Unguided self-help….
• Ethical..?
• Under which conditions?
Questions?
Thank you for your attention
www.isrii.regroup.com
Email: T.Donker@psy.vu.nl