Posttraumatic stress disorder (PTSD) is a commonly occurring and seriously impairing disorder that occurs after exposure to traumatic events (TEs). Symptoms typically begin shortly after TE exposure and evolve with time to either chronicity or recovery. PTSD is one of the most preventable mental disorders, as many people exposed to TEs come to clinical attention in first response settings. Controlled clinical trials show that PTSD risk can be significantly reduced by early preventive interventions. However, these interventions have nontrivial costs, making it infeasible to offer them to all persons exposed to TEs given that only a small minority goes on to develop PTSD. They are also unnecessary for many survivors who recovery spontaneously. To be cost-effective, risk prediction rules are needed to identify which exposed persons are at high risk of PTSD taking into consideration that predictors may vary between samples, within samples (e.g., between male and female survivors) and at different time lags from the TE. A number of research studies have collected longitudinal data addressing this issue by assessing potential predictors of PTSD among TE victims starting in first response healthcare settings, following participants over time, and using baseline data to predict subsequent PTSD. However, these studies'results have often been presented as changes in groups'average likelihood and were not synthesized in a way that would be practical, useful and predictive of individual risk. Therefore, we created a consortium of the principal investigators of the most important such studies to combine their individual- and item-level data towards carrying out a pooled secondary analysis to synthesize information about the predictors of PTSD.
Our Specific Aims are: (1): To construct a consolidated dataset of individual-level data from 16 of the most important longitudinal studies of predictors of PTSD among TE victims starting in first response healthcare settings. These studies assessed a total of 6,390 respondents, 14% of whom have developed acute PTSD;(2): To estimate a latent growth mixture model (LGMM) of PTSD symptom trajectories in the roughly 92% of the consolidated sample (n = 5,917) assessed between one and three times after baseline with the CAPS and then to evaluate the sensitivity of model results to between-sample differences in trajectories and PTSD symptom measures;(3): To estimate the magnitude and cross-study consistency of associations between baseline predictors and PTSD outcomes (acute PTSD in the total sample;PTSD persistence among acute cases;LGMM PTSD class membership and symptom trajectories);(4): To use the results in Aim 3 to develop recommendations for the PTSD risk factors to be assessed in the future in first response settings along with software to facilitate systematic data collection and inform clinical decision making. We seek support to construct this consolidated dataset, to carry out and report the results of analyses, and to develop a risk prediction tool that can be used in first response settings.

Public Health Relevance

The aims of this project are to combine data from prospective studies of PTSD in order to identify the most robust combination of predictors of onset and course of the disorder. The results of this study will inform research, policy and practice by identifying malleable and robust risk factors for the different phases of PTSD: disease onset, and course.

National Institute of Health (NIH)
Research Project (R01)
Project #
Application #
Study Section
Behavioral Genetics and Epidemiology Study Section (BGES)
Program Officer
Tuma, Farris K
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Columbia University (N.Y.)
Public Health & Prev Medicine
Schools of Public Health
New York
United States
Zip Code