Thousands of injuries, hundreds of fatalities and millions of dollars in property loss are among the annual costs of youth firesetting in North America and the UK (Hall, 2005; Arson Prevention Bureau, 2006). Given the scope of the negative outcomes associated with juvenile firesetting, it is surprising that there is so little empirical study of this behaviour. One central reason for the lack of research may be that fire involvement by children and adolescents is relatively common. Almost half of all boys will engage in some unsanctioned fire-starting (Kafry, 1978; Grolnick et al, 1990), which may have led to the perception that juvenile fire involvement is a normal behaviour of little clinical relevance (Pollack-Nelson et al, 2006). Indeed, the average caregiver does not report their child's fire involvement (Kafry, 1978) or view it as requiring expert consultation (Webb et al, 1990). As only about 5% of 300 randomly polled mental health workers received any training about juvenile firesetting (Sparber, 2005), many mental health professionals also may believe that juvenile firesetting is benign.

Juvenile firesetting is a heterogeneous behaviour that occurs during all developmental periods (i.e., pre-school, childhood, adolescence) and many children and adolescents stop their fire involvement without intervention. However, there is a small subgroup whose firesetting will persist, to the peril of both themselves and their communities. Fire involvement or even just fire interest during childhood is an important predictor of both adolescent arson (Hanson et al, 1994) and adult arson (Rice & Harris, 1991). One central task for researchers and clinicians working with youth firesetters has been to differentiate juveniles at low risk for further fire involvement from those at high risk, in order to develop and implement programmes that fit the needs of the individual. Much of the scant empirical literature on juvenile firesetting has examined the characteristics of juvenile firesetters, how they differ from non-firesetters and more recently the differences within the firesetting group. The empirical advances have been slow and the knowledge base remains small. Although specific risk factors have been identified, a comprehensive multifactorial biopsychosocial profile has yet to emerge.