Skip to main content
Effective Health Care Program

Pharmacologic and Nonpharmacologic Treatments for Posttraumatic Stress Disorder: Groundwork for a Publicly Available Repository of Randomized Controlled Trials

Technical Brief Draft
Download PDF527.5 KB

This item is available in PDF and XLSX only (Full Report [PDF, 527.5 KB]; Appendixes A to D [PDF, 865.8 KB]; Appendix E [XLSX, 252.6 KB]; Appendix F [XLSX, 471.6 KB]). People using assistive technology may not be able to fully access information in these files. For additional assistance, please contact us.

Purpose

The purpose of this project is to identify and abstract data from randomized controlled trials (RCTs) of posttraumatic stress disorder (PTSD) interventions to support the development of a publicly accessible data repository by the Veterans Affairs' National Center for Posttraumatic Stress Disorder.

Key Messages

  • Three hundred eighteen RCTs of PTSD interventions published between 1988 and 2018 were included—55% evaluated psychotherapeutic interventions, 30% studied pharmacologic interventions and 15% looked at complementary and integrative or nonpharmacologic biological treatments.
  • Sixty-one percent of RCTs were conducted in the U.S., 78% enrolled < 100 participants, 57% enrolled participants from community populations (versus military/veteran), and 51% enrolled participants with mixed types of traumas rather than a specific trauma type.
  • Less than half of the studies reported on the loss of PTSD diagnosis or clinically meaningful response/remission of symptoms. Reporting was incomplete for many data elements, e.g., history and number of traumatic brain injuries and the number of trauma types.
  • This work will be useful in developing a publicly accessible repository of studies of PTSD treatments as well as providing information on gaps in the evidence to inform future research.

Structured Abstract

Background. Posttraumatic stress disorder (PTSD) reduces quality of life and functioning. People with PTSD have symptoms such as intrusive thoughts, nightmares, flashbacks, avoidance of trauma-related stimuli, negative beliefs about oneself and/or others, and hypervigilance. The symptoms may be due to direct or indirect exposure to trauma, such as witnessing actual or threatened death, injury, or violence including sexual violence, and threats of harm. While recent clinical practice guidelines and reviews exist, providing a single, updatable source of PTSD treatment trials would be useful for clinicians, researchers, and policymakers.

Purpose. To provide detailed information on PTSD treatment research, we systematically abstracted data from randomized controlled trials (RCTs) of PTSD interventions. The Veterans Affairs' National Center for Posttraumatic Stress Disorder (NCPTSD) intends to use the data to develop a publicly available data repository.

Data Sources. We searched the Published International Literature on Traumatic Stress (PILOTS), Ovid® MEDLINE®, Cochrane CENTRAL, PsycINFO®, Embase®, CINAHL®, and Scopus® for eligible RCTs and reviewed reference lists of selected systematic reviews and clinical practice guidelines.

Methods. In consultation with the NCPTSD, we established inclusion criteria for RCTs and specific data elements to be abstracted. We dually reviewed citations from the literature search, and then the full-text of potentially includable articles for eligibility, resolving any disagreements using consensus. One team member abstracted data from included RCTs into evidence tables, and a second reviewer checked abstracted data for accuracy and completeness. The primary publication for each RCT was abstracted; data and citations from any secondary publications appear in the same record.

Findings. We identified 318 RCTs of PTSD interventions for abstraction (106 pharmacologic studies and 212 nonpharmacologic studies) published from 1988 to 2018, with a peak in 2014. Psychotherapeutic interventions were the most commonly studied (55%), while 30% evaluated pharmacologic interventions. Most studies were conducted in the U.S. (61%), and enrolled fewer than 100 participants (78%). More participants were enrolled from a community population (57%) than from a military, veteran or other population, and the majority of studies were conducted in the outpatient setting (67%). Studies most often enrolled participants with a mix of trauma types (51%), followed by studies of participants with combat-related trauma (20%).

Although there was wide variation, the most commonly used PTSD diagnostic instruments were the Clinician-Administered PTSD Scale (CAPS) and the Structured Clinical Interview for DSM (SCID). Less than half of the studies reported loss of PTSD diagnosis and clinically meaningful response/remission of symptoms. Several other data elements were infrequently reported, including the number of participants with a history of traumatic brain injuries and the number of trauma types.

Conclusions. The data abstracted for 318 RCTs of treatments for PTSD can be used to create a publicly accessible data repository. By identifying important gaps in the research, such a data repository can inform future study design and conduct.

Executive Summary

What is this project and why are we doing it? The purpose of this project is to identify and abstract data from randomized controlled trials of treatments for posttraumatic stress disorder (PTSD). The Veterans Affairs' National Center for Posttraumatic Stress Disorder (NCPTSD) intends to use the data to develop a publicly available data repository of randomized controlled trials of PTSD treatments.

People who have experienced direct or indirect exposure to trauma, such as witnessing actual or threatened death, injury, or violence, including sexual violence and threats of harm, may develop PTSD. People with PTSD can present with a diverse and uneven collection of symptoms such as intrusive thoughts, nightmares, flashbacks, avoidance of trauma-related stimuli, negative beliefs about oneself and/or others, and hypervigilance. Untreated, these symptoms can last for years and reduce quality of life and functioning. While there are treatments that have been found to improve symptoms, there is not one single treatment known to be most effective, and selecting a treatment for a given patient can be fraught with uncertainty.

The NCPTSD data repository will serve multiple stakeholders and purposes—

  • Support clinicians, policymakers, researchers, and patients and their families in selecting treatments, and understanding the benefits and harms of different treatments
  • Offer a place to search for evidence on specific interventions, including the participant characteristics and settings in which they have been studied (or identify which treatments have not been studied)
  • Identify evidence gaps to determine research priorities, and serve as a data source for researchers to understand these gaps more fully as they design new research

How did we look for studies and what did we do with studies we found? We consulted with the Agency for Healthcare Research and Quality (AHRQ), sponsoring partner at the Veterans Affairs' NCPTSD, and members of a multidisciplinary Technical Expert Panel to guide our work on this project. We searched multiple databases for eligible studies: Published International Literature on Traumatic Stress (PILOTS), Ovid® MEDLINE®, Cochrane CENTRAL, PsycINFO®, Embase®, CINAHL®, and Scopus®. In addition, we reviewed reference lists of selected systematic reviews and clinical practice guidelines. With input from the NCPTSD and from AHRQ, the EPC team established criteria to determine eligibility for inclusion and exclusion of studies. We abstracted data on population, study characteristics, and outcomes reported in the included studies into evidence tables. The evidence tables for this report are more detailed than the typical systematic review evidence tables in order to achieve the end goal of displaying information on PTSD treatments in a searchable and interactive repository that will be formatted for public availability. We devoted considerable time and attention to developing and documenting standard conventions for recording study data. This documentation supports consistent and comprehensive reporting of study data in the current project and potential future projects.

What did we find? After reviewing 7,843 article abstracts and 1,101 full-text publications, we identified 318 randomized controlled trials that met inclusion criteria. Studies were published from 1988 to 2018, with increased volume in the past 10 years, particularly for nonpharmacologic intervention studies. Psychotherapeutic interventions were the most commonly studied (55% of studies), while 30% evaluated pharmacologic interventions. The vast majority of studies were conducted in the U.S. (61%), and enrolled fewer than 100 participants (78%). More participants were enrolled from a community population (57%) than from a military, veteran or other population for both pharmacologic and nonpharmacologic intervention studies. Similarly, the majority of studies were conducted in the outpatient setting (66%). These studies most often enrolled participants with a mix of trauma types (51%), followed by participants with combat-related trauma (19%).

Although there was wide variation, the most commonly used PTSD diagnostic instruments used were the Clinician-Administered PTSD Scale (CAPS) and the Structured Clinical Interview for DSM (SCID). Less than half of the studies reported key outcomes of loss of PTSD diagnosis and clinically meaningful response/remission of symptoms. Only 13% of pharmacologic studies reported diagnosis change. In addition, there are several other data elements that are infrequently reported in publications of both types of studies. For instance, history and mean number of trauma types is reported in only a small number of nonpharmacologic (9%) and pharmacologic (11%) studies. In addition, very few of the pharmacologic studies reported number of trauma types.

What comes next? Future work to help support and expand the eventual NCPTSD repository may include adding either studies or outcomes and analyses that were not eligible to be included here. Conversion of the abstracted data into an interactive and searchable repository will be completed by the sponsoring partners at the NCPTSD during a later phase of this project.