Abstract
BACKGROUND: Children's active travel to school is associated with physical activity and thus health. Safe Routes to School (SR2S) programs identify 'safe routes' to promote children's active travel to school. No field audit tool exists specifically to assess the microscale built environment of these 'safe routes' within Australia. This study describes the reliability assessment of the Field Audit for Children's Active Transport to School (FACTS) tool. METHODS: The FACTS tool was developed using a multi-step process, including a literature search, expert opinion, and pilot testing. For the reliability assessment, two trained auditors collected data at eight schools over three weeks in April 2021. For inter-rater reliability, auditors conducted audits on the 'safe routes' for the same six schools. For intra-rater reliability, auditors conducted repeat measures on the 'safe routes' for four schools each (eight schools total), including three different schools each from the six used for inter-rater reliability and one additional school. Item-by-item reliability was assessed using Cohen's Kappa, Cohen's Weighted Kappa, and percentage agreement. The reliability of calculated domain scores was assessed using intraclass correlation coefficients. RESULTS: For inter-rater reliability, 31 of the 45 (68.9%) items had moderate to almost perfect agreement, seven items (15.6%) had below moderate agreement, and a Kappa statistic could not be calculated for seven items (15.6%) due to constant values. For intra-rater reliability, 37 of the 45 (82.2%) items had moderate to almost perfect agreement, two items (4.4%) had below moderate agreement, and a Kappa statistic could not be calculated for six items (13.3%) due to constant values. For inter- and intra-rater reliability of the segment domain scores, three of the four domains had substantial to almost perfect agreement. For inter- and intra-rater reliability of the crossing domain scores, all four domains had moderate to almost perfect agreement. For inter- and intra-rater reliability of the segment, crossing, route, and school scores, all had substantial to perfect agreement. CONCLUSIONS: The FACTS tool can reliably characterise the microscale built environment of promoted 'safe routes' for their use within SR2S programs, and should be considered for use in future SR2S programs within the suburban Australian context.