Interpersonal Psychotherapy for Posttraumatic Stress Disorder

Study
Design
IPT
Adaptation
Outcome
Remarks and  
Effect Sizes
Bleiberg & 
Markowitz,  
2005
Open trial
n =
 14
14 weekly sessions
Exposure to trauma 
reminders  
prohibited
Pre/ post  CAPS   
67 → 25
Attrition: 7%
Large effect sizes:
CAPS d = 1.8
Robertson  
et al., 2007
Open trial
n =
 13
8 weekly group IPT 
sessions
“Specially prepared”
treatment manual(?); 
standard group  
IPT?
“Modest” IES 
improvement
Attrition: 0%
Results stable on  
3- month f/ u;
ES: IES subscales 
r
 = 0.63– 0.67
Ray & Webster,  
2010
Open trial
n =
 9
8 weekly 2- hour  
group IPT  
sessions
Based on group IPT 
manual (Wilfley  
et al., 2000)
IES significantly 
improved  
(
p <
 .05)
Attrition: 0%
Some symptomatic 
slippage on  
2- month f/ u;
(ES: not calculable)
Krupnick  
et al., 2008
RCT: IPT vs. WL
n =
 48
16 weekly 2- hour  
group IPT  
sessions
Adapted for  
low- income,   
highly traumatized 
minority women
IPT > WL (CAPS,  
p <
 .001)
Attrition: 29% IPT
Gains persisted at  
4- month f/ u
ES: CAPS d = 1.31
Campanini  
et al., 2010
Open augmentation  
of med trial
n =
 40
16 weekly 2- hour  
group IPT  
sessions
Similar to Krupnick 
et al.; IPT did not 
focus on trauma 
exposure
CAPS 72 → 37,  
with large effect 
size (1.2)
Attrition: 17%
Medication 
non- responders;
ES: CAPS d = 1.17
CAPS = Clinician- Administered PTSD Scale; ES = effect size (Cohen’s d); F/ u = follow- up; IES = Impact of Events Scale; IPT = Interpersonal 
Psychotherapy; RCT = randomized controlled trial; WL = waiting list
Source: Markowitz et al., 2014.

---

6 
I P T   F O R   P T S D
researchers compared group IPT to a waiting list (admittedly not the strongest 
comparison condition) in which patients were repeatedly evaluated but did not 
receive psychotherapy. Medication was held stable for those who were taking 
psychotropic medication.
Paired woman therapists conducted 16 two- hour IPT sessions for groups 
of three to five women. The women in treatment had been multiply trauma-
tized, with an average of 6.8 (standard deviation [SD]  =  4.2) interpersonal 
traumas (!). The authors considered the dropout rate low in the context of 
this high- risk, non– treatment- seeking population: 71% of patients attended at 
least half the sessions. On the other hand, only 63% of IPT and 44% of waiting 
list subjects completed symptom ratings at treatment termination, and only 
81% of IPT patients and 63% of waiting list subjects were rated at four- month 
follow- up.
Scores on the Clinician- Administered PTSD Scale (CAPS; Blake et  al., 
1995; Weathers et  al., 2001)  fell from 65.2 (SD  =  20.9) in IPT to 40.6 post 
treatment and 38.5 (SD = 24.4) at four month follow- up: that is, from severe 
symptoms to mild to moderate symptoms. The waiting list subject scores 
were 62.6 (SD = 16.6), 56.6 (25.1), and 41.6 (SD = 26.7) at the three time points  
(
p <
 .001). IPT also produced significantly greater reductions on the Hamilton 
Depression Rating Scale (Hamilton, 1960) than did waiting list status. IPT pa-
tients improved on four of five subscales (all but Aggression) of the Inventory 
of Interpersonal Problems (IIP; Horowitz et  al., 1988), whereas waiting list 
group IIP scores worsened over time.
Summary:
 The Krupnick study is important as the first randomized con-
trolled trial of IPT for PTSD, and it generated interesting results. A  score 
greater than 60 on the CAPS, where the patients started before treatment, 
indicates severe PTSD; a score of 40 falls right on the diagnostic threshold 
(Weathers et al., 2001). Thus group IPT tended not to bring about remission, 
but did produce clinically meaningful improvement in a repeatedly battered 
population. A weakness of the study is the waiting list control condition: a 
comparison is only as strong as its comparator, and showing that IPT was 
better than no treatment is not definitively impressive. Nonetheless, it’s a start, 
a step into new comparative treatment territory. Although the five therapists 
in this study had some clinical experience, ranging from five to twenty years of 
work in group and psychodynamic psychotherapy, only one of them had had 
prior experience with IPT. If the therapists were not polished in delivering IPT, 
the results might then underestimate the potency of the treatment.