Caroline Helwick

May 23, 2010

May 23, 2010 (New Orleans, Louisiana) — A study of 2 standard therapies for posttraumatic stress disorder (PTSD) shows that outcomes are better when patient preference is taken into account when prescribing treatment.

In other words, "one size does not fit all," said principal investigator Nora C. Feeny, PhD, from Case Western Reserve University, Cleveland, Ohio.

Although treatment with a selective serotonin reuptake inhibitor (SSRI) and treatment via prolonged exposure (PE) were both effective in the study, the best outcomes were observed in patients who got the treatment they preferred, Dr. Feeny reported at the American Psychiatric Association's 2010 Annual Meeting.

The study is apparently the first direct comparison of an SSRI and exposure-based treatment, and it highlights the need to consider the patient's preference when prescribing treatment, she said.

Randomized controlled trials have shown that both sertraline (SER) and PE are effective treatments for PTSD. However, they represent very different options. With PE, clients are encouraged to directly approach the trauma memory and trauma-related fears. With SER, this level of engagement with trauma-related stimuli is not necessary.

Dr. Feeny and colleagues sought to understand how the 2 approaches compare in efficacy for chronic PTSD and how patient preference for one over the other may influence the treatment effect.

To do so they employed a novel design called a doubly randomized preference trial. All subjects viewed a video in which the 2 treatments were described in a strictly nonbiased manner (ie, matched wording).

Patients were then randomly assigned to be further randomized to one treatment or the other or to be allowed to choose their treatment. Patients entering the randomization group were assigned to receive PE or SER without consideration of preference (although some patients inadvertently received the treatment of choice). Patients assigned to the choice group were allowed to select their treatment. This permitted the investigators to determine how patients' preferences affect treatment effect.

Extensive Trauma Exposure

Of 426 patients with a Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, primary diagnosis of chronic PTSD, 200 were ultimately allocated to either the randomization group or the choice group. In the randomization group, 55 participants received PE and 48 received SER. In the choice group, 61 patients elected to receive PE and 36 chose SER.

"Their preferences were consistent with previous literature," Dr. Feeny noted. Baseline characteristics were similar among patients who preferred PE vs SER, except that the SER group had more self-reported depression and less college education.

The study population was primarily women with a history of adult sexual assault (31%), adult nonsexual assault (22.5%), or childhood assault (24%). Only 2.5% of participants had PTSD as a result of combat.

The median time since the trauma exposure was 12 years, and the average participant reported 9 additional trauma types aside from the target trauma. Nearly all had received prior psychiatric treatment.

"This was a large, diverse, and clinically complex sample characterized by longstanding PTSD, extensive trauma exposure, high levels of previous treatment-seeking, and substantial comorbidity," Dr. Feeny noted.

Psychopathology and function scales included the Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, the PTSD Symptom Scale (PSS-I), the Hamilton Ratings Scale for Depression (HRSD-24), the PTSD Symptom Scale—Self-Report (PSS-SR), the Beck Depression Inventory (BDI), the State-Trait Anxiety Inventory (STAI), and the Sheehan Disability Scale (SDS).

SER was administered in a flexible dosing schedule (50 - 200 mg) under a standardized titration algorithm; patients receiving SER were not given exposure or antiexposure instructions. They did receive psychoeducation information. Patients were treated in 10 weekly 30-minute sessions by board-certified psychiatrists.

The PE group received education about common reactions to trauma, breathing retraining, prolonged repeated exposure to the trauma memory, and repeated in vivo exposure to situations they were avoiding. Treatment occurred in 10 weekly sessions of 90 to 120 minutes by Masters' and doctoral-level clinicians.

PE Demonstrates Slight Edge, but Preference Is Key

"Overall, both PE and SER showed good efficacy," Dr. Feeny reported. "Globally, almost all the treatment effects for both arms were large. PE may have had a slight advantage in terms of magnitude of change and loss of the PTSD diagnosis at 10 weeks, which is generally consistent with the broader literature."

The changes from baseline in key measures of PTSD are shown here.

Pre- to Posttreatment Effect Sizes (Intent to Treat)

  Choice No Choice
PSS-1 (severity) 1.42 1.95 .95 1.35
PDS (severity) 1.68 1.74 .99 1.41
HRSD (depression) .85 .99 .59 .71
BDI (depression) 1.27 1.31 .71 1.22
STAI-S (anxiety) 1.28 1.20 .52 1.02
SDS (functioning) .83 .81 .66 1.04

Treatment effects of 0.8 and higher indicate a "large impact of treatment," Dr. Feeny pointed out. Because this was an intention-to-treat analysis, which included patients who did not complete treatment, she believes these are underestimates of the actual treatment effect.

"Importantly, there were clear effects on PTSD outcome of being randomized to the choice arm," she said. "Patients who had no choice in their treatment had more diminished effects."

Response rates were higher among patients having a choice, as were rates of losing the PTSD diagnosis, which were approximately 80% and 70%, respectively, compared with approximately 55% for each category when patients were not allowed to choose their treatment.

Only 4 patients needed to be treated to obtain a response among patients who were given a choice.

The most prominent effects were seen when there was a discrepancy between the treatment assignment and the patient's preference. Among patients who did not receive their preferred treatment, 59% retained their PTSD diagnosis compared with 29% in nondiscrepant cases, and they tended to have more severe PTSD, as well as depression and anxiety, she reported.

"This may be asserting its effect through a lack of treatment adherence," said Dr. Feeny.

The study found evidence of lower SER dosage at the end of the study and less adherence to PE homework among the discrepant population. Mean SER dose was 144 mg/day for patients who chose SER compared with 62 mg/day for those who preferred PE but got SER (P < .001).

In addition, the number of sessions completed was 7.69 for patients lacking discrepancy and 5.14 for those with discrepancy (P < .001), and treatment completion rates were 73% and 49%, respectively (P = .002).

One Size May Not Fit All

"This is consistent with preference studies in depression. Pharmacotherapy compliance and psychotherapy outcomes have been shown to be impacted by treatment preferences," said Dr. Feeny.

"Our results highlight the need to rethink a 'one size fits all' approach to the treatment of PTSD," she added.

Murray B. Stein, MD, MPH, from the University of California–San Diego School of Medicine, moderated the session and commented for Medscape Psychiatry that it is not clear how often a patient with PTSD receives the treatment he or she may prefer, but "it does seem to make a difference in outcome."

"It appears that when patients do not get what they want, they are not likely to be very adherent to the treatment that did not make sense in the first place," he suggested.

"In a lot of settings, practitioners will make available what they know how to do, and for most, including psychiatrists, it's medication management for anxiety disorders, and this is also what they get paid for," said Dr. Stein.

"Many insurers will not pay for psychotherapy but will pay for medication, though for anxious patients, the preference is clearly for a psychosocial treatment," he added.

The study was funded by a supplemental grant from Pfizer, Inc. Dr. Feeny has reported no other conflicts of interest. Dr. Stein has reported receiving research funding from Hoffman-Laroche and consulting fees from AstraZeneca and Bristol Myers Squibb.

American Psychiatric Association's 2010 Annual Meeting: Symposium 12. Presented May 22, 2010.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.