Factor Structure and Measurement Invariance of the Revised Brief Addiction Monitor

There are currently 2 versions of the Brief Addiction Monitor (BAM) being widely used within Department of Veterans Affairs (VA) medical centers and other treatment settings: the BAM, which entails use of discrete response options for all items, and the revised version, the BAM-R, which consists of the same items but uses continuous response options for several of the items. There is also conflicting evidence about the factor structure of the original BAM, with a 4-factor structure proposed by 1 study that refutes a 3-factor structure proposed from the original study of the measure. The BAM-R is widely administered in substance use treatment settings across the country and is overtaking the discrete BAM as the preferred instrument, although little research has examined the factor structure or longitudinal performance of this version of the measure. The purpose of this study is to examine the factor structure and temporal stability of the BAM-R among a large national sample of veterans across multiple treatment settings (i.e., all VA veterans with at least 2 complete BAM-R administrations reflected in the medical record; N = 22,453). Findings suggest that the 4-factor structure is superior to the commonly used 3-factor structure for both model fit and stability over two occasions of measurement and should be the factor structure used for clinical and research purposes pending further measure revisions. Public Significance Statement A widely used assessment in substance use disorder treatment settings, the Brief Addiction Monitor (BAM), has 2 versions: the BAM and the BAM-R. In this study, we demonstrate that the original 3-factor structure of the BAM does not fit the data from the BAM-R as well as does an alternative 4-factor structure, and we discuss implications for future research and clinical use of the BAM-R.