Use of a two-phase process to identify possible cases of mental ill health in the UK military

Two-phase mental health screening methods, in which an abridged mental health measure is used to establish who should receive a more comprehensive assessment, may be more efficient and acceptable to respondents than a stand-alone complete questionnaire. Such two-phase methods are in use in US armed forces post-deployment mental health screening. This study assesses the sensitivity and specificity of abridged instruments (used in the first phase) compared to the full instruments (the second phase), and whether false negative cases resulting from the use of abridged tests were detected by another test, among a UK military screening sample. Data from a group of UK Armed Forces personnel (n=1464) who had completed full questionnaires assessing symptoms of post-traumatic stress disorder (PTSD) (PTSD Checklist - Civilian Version, PCL-C) and alcohol misuse (Alcohol Use Disorder Identification Test, AUDIT) were used. An abridged version of the PCL-C performed well in discriminating potential PTSD cases (as measured by the full instrument); AUDIT showed less discriminatory power, particularly due to poor specificity. Many cases missed by one abridged test would have been detected by an alternative test. Thus two-phase screening designs reduce the resource burden of a project without substantial loss of sensitivity for PTSD, but are less effective in discriminating potential cases of alcohol misuse.