Portuguese version of the PTSD checklist-military version (PCL-M) – II: Diagnostic utility

Teresa Carvalho, José Pinto-Gouveia, Marina Cunha, Joana Duarte

Research output: Contribution to journalArticlepeer-review


Objective: War veterans are at high risk of developing posttraumatic stress disorder (PTSD), and the development of brief self-report instruments that enable screening for PTSD in this population is crucial. The PTSD Checklist-Military Version (PCL-M) is widely used for this purpose. This study sought to explore the diagnostic utility of the Portuguese version of the PCL-M. Methods: The participants were 86 Portuguese Colonial War veterans (42 with a PTSD diagnosis and 44 without PTSD). Participants completed a self-report instrument designed to collect sociodemographic data, the PCL-M, and the Clinician-Administered PTSD Scale (CAPS). Results: The area under the receiver operator characteristic (ROC) curve showed excellent discriminant ability between subjects with and without PTSD (AUC = 0.94). To achieve a positive PTSD diagnosis, an optimal cutoff point of 49 for the PCL-M total score and cutoff points for each of its 17 items are recommended. Conclusions: This work is a relevant contribution for research and clinical practice in the vast population of Portuguese Colonial War veterans. Use of the PCL-M as a screening tool for PTSD symptoms will allow easier, resource-aware targeting of subjects with a potential PTSD diagnosis, adding to the improvement of public health in Portugal.

Original languageEnglish
Pages (from-to)55-62
Number of pages8
JournalRevista Brasileira de Psiquiatria
Issue number1
Publication statusPublished - 2015
Externally publishedYes

Subject classification (UKÄ)

  • Psychology

Free keywords

  • Diagnosis and classification
  • Military psychiatry
  • Posttraumatic stress disorder
  • Psychometric
  • Statistics
  • Tests/interviews


Dive into the research topics of 'Portuguese version of the PTSD checklist-military version (PCL-M) – II: Diagnostic utility'. Together they form a unique fingerprint.

Cite this