Date of Award

Spring 5-15-2015

Degree Type


Degree Name

Master of Public Health (MPH)


Public Health

First Advisor

Daniel J. Whitaker, PhD

Second Advisor

Kim Ramsey-White, PhD


Background: In 2013, there were over 3 million reported cases of child maltreatment to state agencies, and 693,484 unique children were victims of child maltreatment. To address this public health problem, there is a move towards implementation of evidenced-based practice in the child welfare system. However, evidence based programs that are disseminated to community settings often fail to deliver the same positive outcomes found in research studies. Concern about this knowledge- to- practice gap has led to a burgeoning research focusing on factors that predict successful implementation including high fidelity.

Purpose: This study aims to fill the gap in knowledge regarding the relationship between provider fidelity for SafeCare, an evidenced-based parenting program, and parent behavior change. By improving our knowledge about fidelity in implementation we can work to improve the translation from laboratory to “real life.”

Methods: The data was derived from a longitudinal study aimed at examining the implementation of the SafeCare model in Georgia. As part of this implementation the National SafeCare Research and Training Center (NSTRC) conducted trainings with 54 agencies across the state of Georgia. Trainings were delivered to individuals (“home visitors”) under NSTRC implementation model who then began delivering SafeCare in the child welfare population. Home visitor fidelity monitoring was conducted by SafeCare trainers via audiotaped sessions in the home.

Family outcome data were collected as part of home visitors assessments of parents done as part of the SafeCare model. Specifically, home visitors measured home safety, parent health knowledge, and parent-child interactions.

Results: Correlations between fidelity and parenting outcomes ranged from r = -0.347 to r = .330, but none were statistically significant. To examine whether fidelity predicts behavior when controlling for other factors that may be predictive of family behavior, regression models were conducted controlling for age, sex, race, and work experience. In all regression models, fidelity remained a non-significant predictor (p > .05).

Conclusions: There are many possible reasons for the failure to find a statistically significant relationship between fidelity and behavior change including restricted variability, sample size, measures—or the possibility that no true relationship exists. By understanding fidelity in this implementation we can begin to think about successful components of broader implementation of this behavioral parenting model.