Health Services Research & Development

Veterans Crisis Line Badge
Go to the ORD website
Go to the QUERI website

DHI 07-150 – HSR&D Study

New | Current | Completed | DRA | DRE | Portfolios/Projects | Centers | Career Development Projects

DHI 07-150
Soldier to Civilian: RCT of an Intervention to Promote Post-Deployment Reintegration
Nina A. Sayer PhD
Minneapolis VA Health Care System, Minneapolis, MN
Minneapolis, MN
Funding Period: September 2009 - August 2013

BACKGROUND/RATIONALE:
Veterans returning from war with difficulty reintegrating into civilian life experience barriers to seeking and obtaining care. There is a need for interventions to improve mental health and reintegration among new veterans that are accessible and non-stigmatizing. Prior research indicates that new veterans prefer to receive help for reintegration problems over the internet and that expressive writing methods can help people cope with past trauma and major life transitions.

OBJECTIVE(S):
To determine if a brief, highly scalable online intervention - writing expressively about transitioning to civilian life - improves symptoms and functioning in Iraq and Afghanistan war Veterans. We hypothesized that expressive writing would be more effective than factual writing and no writing in reducing symptoms of PTSD, depression, anxiety, and hostility and in improving physical symptoms, reintegration, social support, and life satisfaction. We further hypothesized that expressive writing effects would be maintained over time. In secondary analyses, we explored possible differences in effectiveness by gender, race, time since deployment, and VA user status.

METHODS:
Design: A 3-group, randomized controlled trial of 1,292 Iraq and Afghanistan war veterans with self-reported reintegration difficulty recruited from a national database. Study procedures were implemented over the internet from July, 2011 through April, 2013 and included follow-up assessments at 3- and 6-month post-baseline. We used gender as the blocking variable.
Intervention: Participants were randomized into groups that completed either four sessions of expressive writing (n = 508), four sessions of factual control writing (n = 507), or no writing (n = 277).
Randomization and blinding: After completing the consent form, participants were randomized to expressive writing, factual writing or no writing control conditions at a rate of 2:2:1. To generate the allocation sequence, we sequentially generated ordered restricted permutations of length 10, restricted to have 4 As, 4 Bs and 2Cs within each gender. As participants consented, the website randomized them into one of the study arms by assigning the first unused position in the randomization tables. Participants were blinded with respect to writing options; statisticians and investigators were blind to assignment.
Measures of Adherence: We assessed emotional state before and after each writing session. After each writing session, we asked participants to rate their writing experience. We used a computer program that analyzes text using word categories to examine word use in participants' essays. We expected those randomized to expressive writing to experience a larger increase in negative emotion, report that their essays were more personal and that they revealed more emotion, and to use more negative emotion words than those in factual writing.
Main Outcome Measures: Self-reported Posttraumatic Stress Disorder (PTSD) symptoms, hostility, distress, physical complaints, reintegration difficulty, satisfaction with life and social support assessed with standardized measures at baseline and 3- and 6-month follow-up.
Analysis: Power calculations based on a small effect indicated that 288 participants per arm was sufficient to achieve .96 power at = .05. However, we used 2:2:1 allocation, thereby doubling the sizes of the writing groups and increasing power to detect small effects even if responses to writing instructions were highly variable. Primary analyses were performed using intention to treat. We analyzed outcomes using generalized linear mixed models. We computed within and between group effect sizes as d, the standardized mean difference.

FINDINGS/RESULTS:
Participants and Adherence to Writing Instructions: Baseline characteristics were the same across conditions, with the exception that a smaller proportion of those randomized to factual writing had a service-connected mental health condition than those randomized to expressive and no writing conditions (P< .05). The average number of completed writing sessions was 2.54 in expressive writing and 2.84 in factual writing (P = .002). As we expected, expressive writing participants demonstrated greater increase in negative emotion after each writing session than those in factual writing (Ps < .0001). Those in expressive writing also reported after each writing session that their writing was more personal and that they revealed more emotions (Ps < .0001). Consistent with this, the first, second, and fourth writing session samples contained a higher proportion of negative emotion words for expressive writers than for factual control writers (P < .0001); in the third session, the proportion was similar across intervention and control (P > .69).

Intent to Treat Analysis: Veterans who wrote expressively experienced significantly greater reductions in PTSD symptoms, distress, hostility, physical complaints, and reintegration difficulty compared with veterans who did not write (Ps < .01). The standardized mean difference - Cohen's d effect size - between expressive and no writing conditions ranged from 0.22 to 0.33 at 3-month and 0.18 to 0.35 at 6-month post-baseline. Compared with veterans who wrote factually, veterans who wrote expressively experienced greater reductions in physical complaints at 3-month (effect size, 0.14, P = .03) and greater reductions in physical complaints (effect size, 0.16, P = .009), PTSD symptoms and distress (effect size, 0.15, P = .02) at 6-month follow-up. Expressive writing did not lead to improved social support or life satisfaction relative to the control conditions. There were no differences in the effectiveness of expressive writing by gender, race, time since deployment, or VA user status.
Adverse Events: There were no serious adverse events in any condition. However, as per study safety protocol, we had a clinical psychologist external to the study review writing samples submitted by expressive writing participants who expressed possible suicidal or homicidal thoughts and one expressive writing participant who describe concerns about violence in a family member. This external reviewer did not identify immediate risk in any of these cases.

IMPACT:
Expressive writing, a simple, resource-efficient intervention that can be widely implemented via the internet, is a promising strategy for improving health and functioning among veterans who experience postdeployment reintegration difficulty. More research on expressive writing in veterans is warranted. In addition to examining mechanisms, it would be important to determine whether we can magnify expressive writing effects by, for example, altering instructions or including feedback on the writing. In addition, there is a need for studies that identify individuals most likely to benefit from expressive writing.

PUBLICATIONS:

Journal Articles

  1. Orazem RJ, Frazier PA, Schnurr PP, Oleson HE, Carlson KF, Litz BT, Sayer NA. Identity adjustment among Afghanistan and Iraq war veterans with reintegration difficulty. Psychological trauma : theory, research, practice and policy. 2017 Aug 1; 9(Suppl 1):4-11.
  2. Sayer NA, Noorbaloochi S, Frazier PA, Pennebaker JW, Orazem RJ, Schnurr PP, Murdoch M, Carlson KF, Gravely A, Litz BT. Randomized Controlled Trial of Online Expressive Writing to Address Readjustment Difficulties Among U.S. Afghanistan and Iraq War Veterans. Journal of traumatic stress. 2015 Oct 1; 28(5):381-90.
  3. Sayer NA, Orazem RJ, Noorbaloochi S, Gravely A, Frazier P, Carlson KF, Schnurr PP, Oleson H. Iraq and Afghanistan War Veterans with Reintegration Problems: Differences by Veterans Affairs Healthcare User Status. Administration and policy in mental health. 2015 Jul 1; 42(4):493-503.
  4. Sayer NA, Carlson K, Frazier P. Reintegration Challenges in U.S. Service Members and Veterans Following Combat Deployment. Social issues and policy review. 2014 Jan 13; 8(1):33-73.
  5. Sayer NA, Frazier P, Orazem RJ, Murdoch M, Gravely A, Carlson KF, Hintz S, Noorbaloochi S. Military to civilian questionnaire: a measure of postdeployment community reintegration difficulty among veterans using Department of Veterans Affairs medical care. Journal of traumatic stress. 2011 Dec 7; 24(6):660-70.
Conference Presentations

  1. Sayer NA, Noorbaloochi S, Frazier PA, Pennebaker JW, Orazem RJ, Schnurr PP, Murdoch M, Carlson KF, Gravely AA, Litz B. RCT of an online expressive writing intervention for Veteran reintegration difficulty. Poster session presented at: American Psychological Association Annual Convention; 2015 Aug 6; Toronto, Canada.
  2. Sayer NA, Rosen CS, Nugent S, Kehle-Forbes SM, Chard K, Bernardy NC, Schnurr P, Orazem R, Mohr D. A Preliminary Look at Use of PE and CPT in VHA Specialty Outpatient PTSD Programs. Paper presented at: VA HSR&D / QUERI National Meeting; 2015 Jul 9; Philadelphia, PA.
  3. Sayer NA, Noorbaloochi S, Frazier P, Pennebaker JW, Orazem RJ, Schnurr PP, Murdoch M, Carlson KF, Gravely A, Litz B. Randomized Controlled Trial of Online Expressive Writing for Veteran Reintegration Difficulty. Paper presented at: International Center of Mental Health Policy and Economics Annual Conference; 2014 Sep 26; Washington, DC.


DRA: Military and Environmental Exposures, Mental, Cognitive and Behavioral Disorders, Acute and Combat-Related Injury
DRE: Technology Development and Assessment, Treatment - Observational, Treatment - Efficacy/Effectiveness Clinical Trial
Keywords: Deployment, Gender Differences, Operation Enduring Freedom, Outcomes - Patient, PTSD, Reintegration Post-Deployment, Social Support, Telemedicine/Telehealth
MeSH Terms: none