Early life trauma is highly prevalent among marginalised populations. For example, the prevalence of childhood sexual abuse among opioid-dependent individuals has been reported as 36% and 72% for males and females, respectively (e.g. Conroy et al 2009) while the rate of trauma exposure prior to age 16 was found to be 71% among a homeless sample (e.g. Taylor & Sharpe, 2008). Early life trauma can impart a long-standing vulnerability for mental disorder and is associated with greater severity of disorder and poor treatment outcome (e.g. Nanni et al 2011). It is important to quantify the trauma histories of marginalised populations to inform the development of appropriate interventions and service responses. Service providers, however, are not always comfortable with research that asks highly sensitive questions such as those measuring trauma exposure or trauma-related symptomatology. This paper comments on the experiences of the authors in measuring child maltreatment and other early life trauma among drug treatment and homeless populations. Experience is drawn from several studies including a large case-control study of trauma and mental disorder among an opioid-dependent sample, two longitudinal surveys of chronically homeless adults accessing integrated homelessness and psychosocial support services, and a longitudinal survey of homeless young people with complex needs. Specifically, the paper discusses: a) whether we should be asking ‘vulnerable’ individuals about past trauma and the consequences of doing so; b) the concerns raised by services and other stakeholders and how these can be addressed; c) the extent to which marginalised populations are ‘vulnerable’ and hence need to be ‘protected’ from particular research inquiries.