Validation of a brief PTSD screener for underserved patients in federally qualified health centers

Gen Hosp Psychiatry. 2016 Jan-Feb:38:84-8. doi: 10.1016/j.genhosppsych.2015.07.009. Epub 2015 Jul 30.

Abstract

Objective: The objective was to validate the reliability and efficiency of alternative cutoff values on the abbreviated six-item Posttraumatic Stress Disorder (PTSD) Checklist (PCL-6) [1] for underserved, largely minority patients in primary care settings of Federally Qualified Health Centers (FQHCs).

Method: Using a sample of 760 patients recruited from six FQHCs in the New York City and New Jersey metropolitan area from June 2010 to April 2013, we compared the PCL-6 with the Clinician Administered PTSD Scale (CAPS) for the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. We used reliability statistics for single cutoff values on PCL-6 scores. We examined the relationship between probabilities of meeting CAPS diagnostic criteria and PCL-6 scores by nonparametric regression.

Results: PCL-6 scores range between 6 and 30. Reliability and efficiency statistics for cutoff between 12 and 26 were reported. There is a strong monotonic relationship between PCL-6 scores and the probability of meeting CAPS diagnostic criteria.

Conclusion: No single cutoff on PCL-6 scores has acceptable reliability on both false positive and false negative simultaneously. An ordinal decision rule (low risk: 12 or less, medium risk: 13 to 16, high risk: 17 to 25 and very high risk: 26 and above) can differentiate the risk of PTSD. A single cutoff (17 or higher as positive) may be suitable for identifying those with the greatest need for care given limited mental health capacity in FQHC settings.

Keywords: PCL-6; PTSD; Screening; Validation.

Publication types

  • Research Support, N.I.H., Extramural
  • Validation Study

MeSH terms

  • Adult
  • Case-Control Studies
  • Checklist*
  • Community Health Centers*
  • Female
  • Humans
  • Male
  • Mass Screening
  • Middle Aged
  • New Jersey
  • New York City
  • Psychometrics
  • Regression Analysis
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Stress Disorders, Post-Traumatic / diagnosis*
  • Vulnerable Populations*