Behavior change across implementations of the SafeCare model in real world settings.

• Very large changes were observed in the behaviors targeted by SafeCare across implementations. • SafeCare can be used with a range of families and is likely to result in changes to behaviors targeted. • Effect sizes across targeted behaviors of SafeCare were comparable to prior research. Child maltreatment is a significant public health problem affecting more than 3.4 million children per year in the U.S. Child protective systems (CPS) are increasingly adopting evidenced-based practices (EBP), with many specifically adopting behaviorally-based parenting programs (BPTs). One challenge with scaling-up potentially promising interventions it to continue monitoring outcomes for evaluation purposes because program impacts tend to decline when disseminated. Little is known regarding the influence of client diversity on program impacts in the field. This paper examined immediate target skill behavior change among parents receiving the SafeCare model across six implementations. Program impacts were also examined according to family characteristics. Data for 493 families were collected as part of 64 SafeCare implementations between 2008 and 2014, and included family characteristics and behavior change indices. Behavior change indices for the three SafeCare modules (parenting-child interactions, home hazards, and child health care skills) were collected at pre- and post-training sessions by the SafeCare provider. Behavior change across the three modules were substantial. Parents healthcare skills specifically for their children increased by 29.4 percentage points (sd = 26.8, d = 1.74, p <.01). Home safety hazards decreased by 12.54 hazards (sd = 13.9; d = 3.00, p <.01), and parent-child interaction scores increased by 25.0 percentage points (sd = 23.9, d = 2.10, p <.01). There was no evidence that behavior change was moderated by family characteristics including parent age, sex, child protective service status, number of children, or income. [ABSTRACT FROM AUTHOR]