As evidence-based programs are implemented in real world settings, there is a strong need to effectively and efficiently monitor fidelity, or adherence to a program, in order to maintain the expected effects demonstrated in research settings. The purpose of this study was to compare two methods of assessing fidelity to an evidence-based, parent-training model (SafeCare®) as implemented by community service providers. Specifically, analyses compared fidelity assessed via video versus audio recordings. SafeCare modules often require mobility and high interaction, thus, video recordings may provide a more accurate view of home visitor and family interactions for scoring fidelity. However, videos are more expensive and cumbersome in comparison to audio recordings. Trained coders were randomly assigned to score a video or audio recording of the same session for 25 SafeCare sessions and the codes were compared for agreement. Two types of SafeCare sessions were assessed: assessment and training. Average agreement was somewhat higher for assessment sessions than for training sessions. Average agreement, across all sessions, was higher among items pertaining to SafeCare content than items pertaining to the therapeutic process. Several specific items were identified that are difficult to code via audio recordings. However, more research is needed to determine agreement levels across all SafeCare modules and session types in order to provide insight on the implications for SafeCare's future use of audio and video methods of measuring fidelity.