Prospective Analysis of Linguistic Analysis as a Method for Assessing Trauma Symptoms after Hurricane Harvey among Houstonian Adults



Journal Title

Journal ISSN

Volume Title



Hurricane Harvey was one of the most destructive hurricanes in United States’ history and negatively impacted a majority of Houstonians. Posttraumatic stress disorder (PTSD) symptoms are a common consequence for individuals who experience this form of trauma. Additionally, a dose-response effect has been documented in trauma symptoms following natural disaster, with more severe trauma related to increased symptomology. Given the severity of Hurricane Harvey, Houstonians constitute a high-risk population for experiencing heightened trauma symptoms. Limitations of current methods (e.g. self-report, clinical interview) for assessing trauma symptoms are particularly salient after a large-scale natural disaster, when the availability of mental health resources may be especially limited. The aim of the current study was to use the computer program Linguistic Inquiry and Word Count (LIWC) to analyze Houstonian adults’ narratives about Harvey (collected online in response to a brief prompt shortly after the event) to determine if specific linguistic markers are associated with an individual’s PTSD symptomology concurrently (one to two months post-disaster) and prospectively (six months post-disaster). Results indicate that greater use of biological process words (e.g., blood, pain), its subcategory body words (e.g., hands, spit), and fewer cause words (e.g., because, effect) were related to increased trauma symptoms at baseline. Additionally, use of fewer cognitive process words and greater use of bio words at baseline predicted greater symptom change at follow up, extending previous research findings. Findings suggest that linguistic analysis may be an important component of assessment and treatment monitoring of trauma symptoms after a hurricane.



Trauma symptoms, Linguistic analysis, Assessment, Natural disaster