Abstract
BACKGROUND: There are currently no validated instruments to assess the burden of combat-related Posttraumatic Stress Disorder (PTSD) in Sinhalese-the main spoken language in Sri Lanka. OBJECTIVE: The purpose of this research was to establish the cross-cultural and structural validity of the PTSD Check List-Military Version (PCL-M) translated into Sinhalese. METHODS: Expert committee consensus generation as well as translation-back translation approaches were used to establish the semantic, conceptual, and content equivalence of the Sinhalese and English versions of the PCL-M. Four translations of each item were made. In the absence of any "gold standard" psychometric instrument in Sinhalese to establish the criterion validity for the PCL-M (SIN), the study utilized more informal checks for assessment of validity and Sri Lankan cutoffs for caseness for PTSD to establish the psychometric strength of the translated instrument along with standard reliability analysis. Confirmatory factor analysis was performed on PCL-M scoring of a random sample of 1,586 soldiers to examine construct validity. RESULTS: Thirteen of the 17 items were selected by popular vote, and the remaining 4 through discussion and consensus. Reliability measured by Cronbach's-α was 0.944 for the total scale and 0.812, 0.869, and 0.895 for the three DSM-IV sub-scales (re-experiencing, avoidance/numbing, and hyperarousal), respectively. The desired cutoff point for the translated instrument was determined to be 44. The five-factor model by Elhai et al. and the four-factor model by King et al. fitted best, demonstrating good fit to all three fit indices, while the four-factor model and the DSM-IV three-factor model by Simms et al. only had acceptable levels of fit for root mean squared error of approximation. χ(2) difference test comparing the two better-fitting models suggests that the five-factor model by Elhai et al. has the better fit. CONCLUSION: The PCL-M (SIN) version is suitable for use in the study of PTSD in the Sri Lankan military forces, as judged by cross-cultural and construct validity as well as reliability.