Dadswell, Kara (2018) Predicting risk of repeat firelighting in young people: the development and evaluation of the Behaviour Risk Tool. PhD thesis, Victoria University.

Abstract

Firelighting is not an uncommon behaviour for young people. A review of the relevant literature shows that, fortunately, most young people who light fires, do so because they are curious, and fire safety education intervention is enough to prevent any future firelighting activity. However, a small number have significant psychosocial issues that may contribute to continued firelighting behaviour. The Juvenile Fire Awareness and Intervention Program (JFAIP) is a fire safety education program established in Victoria, Australia to target active young firelighters. While some young firelighters referred to the program have mental health problems, the JFAIP is not designed for such cases. Thus a need has been identified for a screening questionnaire to identify young people with psychosocial disturbance predictive of ongoing firelighting behaviour, in order to recommend supplementary mental health intervention. The overarching aim of this project was to develop such a screening tool. The first aim of Study 1 was to evaluate the validity of two internationally established firelighter screening tools, the FEMA Child Risk Survey (CRS) and Family Risk Survey (FRS). The second aim was to explore the fire-specific, family and psychosocial profile of young repeat firelighters and develop a new firelighter screening tool to be subsequently evaluated and compared to the CRS and FRS. Utilising a sample of 61 JFAIP families, data about fire-specific, family and psychosocial characteristics of the young firelighters (5-17 years of age) at the time of their initial pre-intervention interview was collected via the CRS, FRS, Fire Risk Interview and Child Behaviour Checklist. A 12-month follow-up obtained information about continued firelighting activity. Study 1 results indicated that the CRS was unable to distinguish between the groups (repeat and non-repeat), but the FRS demonstrated some utility, accurately detecting 86% of repeat firelighters. However it was overly inclusive, erroneously predicting that half of the non-repeat firelighters were also at risk. The results also identified significant differences on a number of fire-specific and psychosocial variables between repeat and non-repeat firelighters. The 25 specific items or themes across the four instruments used to collect data demonstrating the greatest differences between the repeat and non-repeat firelighters were combined into a new screening tool, termed the Behaviour Risk Tool (BRT). Study 2 was designed to measure the validity and reliability of the BRT using new samples. In Part A the BRT was completed by the parent/guardian of 63 young people in the JFAIP (5-17 years of age) at their initial pre-intervention interview. A minimum six month follow up contact determined the presence or absence of repeat firelighting. Results demonstrated the BRT had a sensitivity of 0.8 and specificity of 0.7 at the cut-off score of 57.5. Thus the BRT detected 80% of the repeat firelighters and 70% of the nonrepeat firelighters correctly. In Part B, the test-retest reliability and internal consistency of the BRT was assessed. Parents/guardians of 76 children (5-17 years) in the general population completed the questionnaire twice, two weeks apart. The results indicated high repeatability across time (r=.93) and high internal consistency (.88-.93) for the BRT. Hence, Study 2 found that the BRT had better sensitivity and specificity than the CRS and FRS and was a reliable questionnaire. The expectation is that the BRT will be used as a preliminary screening measure in the JFAIP, to identify cases where additional mental health support may be necessary. This is the first such tool to be developed in an Australia context. Furthermore, the JFAIP is the first young firelighter program in Australia to adopt a screening tool of this kind. Ideally the BRT will continue to be used over many years with many families, and continue to be evaluated for effectiveness and further improvements.

Search Google Scholar

Repository staff login