The present paper provides a brief overview of methodological issues involved in the process of assessment related to the classification and prediction of relapse. These include conceptual and operational definitions of relapse, retrospective versus prospective assessment, attributional biases in recalling relapse events, single versus multiple determinants of relapse, static versus dynamic assessment models, and the necessary level of specificity involved in the assessment of relapse categories. Additionally, general domains representing distal personal characteristics, intermediate background variables and factors proximal in time to relapse situations are reviewed. Potential variables appropriate for assessment within each of these domains are described. It is concluded that relapse is best understood as a complex process having multiple and interactive determinants that vary in their temporal proximity from and their relative influence on relapse. An adequate assessment model must be sufficiently comprehensive to include theoretically relevant variables from each of the multiple domains and different levels of potential predictors.